Evaluation of Impact of Security Measures on Container-based and VM-based Application Performance in Cloud

Evaluation of Impact of Security Measures on Container-based and VM-based Application Performance in Cloud | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Evaluation of Impact of Security Measures on Container-based and VM-based Application Performance in Cloud Avleen Kaur, Zoheir Ezziane This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7368391/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract This paper explores the trade-off between security and performance of web applications hosted on Virtual Machine based environments and container-based environments. This paper also investigates the comparison between container and Virtual Machine performance on several aspects. With the increasing adoption of Virtual Machines and container technologies for web applications, understanding the significance of security measures becomes crucial. A series of tests were conducted on Virtual Machine based and container-based web applications, implementing encryption, and network isolation. This paper measures the key performance indicators, such as response time, CPU utilization, Memory utilization, and network utilization. The results indicate a 40% increase in the response time of a container than a Virtual Machine in the case where security measures have been applied to the environment. In the analysis of secured and unsecured VMs, all the metrics show a difference of less than 10%. However, in the case of containers, there is a significant difference in the performance ranging from 28–85% across the board. These findings highlight the need for optimized security strategies that mitigate performance degradation. Future works will focus on exploring more advanced cloud setups and additional security measures. This paper provides foundational analysis to balance security and performance in the design and deployment of modern web applications. Cloud computing Virtual Machine Containers Web Applications Performance Security Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Figure 14 Figure 15 Figure 16 Figure 17 Figure 18 Figure 19 Figure 20 Figure 21 Figure 22 1. Introduction In today’s rapidly evolving digital landscape, web services have become the basis for businesses and individuals aiming to deliver several online services. They facilitate a wide range of daily interactions, from checking the weather, to more complex tasks. These services are increasingly being deployed to cloud platforms. This helps provide scalability, flexibility, and cost efficiency, allowing organizations to dynamically adjust resources in response to fluctuating demands. Cloud service providers, such as, Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform (GCP) are leveraged by the organizations to deploy web applications. New considerations for performance and security are introduced while deploying web services to the cloud. There is a necessity for a balance between scalability and the management of overheads introduced by virtualized technologies. In the last few years, there has been a significant shift towards web services, driven by the need for higher scalability, accessibility, and flexibility in web solutions. The adoption for cloud-based architectures, such as microservices has increased which ensures the seamless integration and maintenance of applications across various platforms. This has led to a significant increase in the use of APIs, serverless computing, and containerisation technologies. The quality of a web service has been defined as a non-functional characteristic [1]. Meaning, factors like total execution time taken by a service, cost of invoking the service, availability, and security features need to be analysed. If we explore the different types of web services, it becomes clear that the selection of the type of web services influences the architectural design and functionality of applications. It introduces the topic of performance of web services and how it becomes crucial to its adoption. Performance enhances user satisfaction by providing faster response times and reliable interactions. It also plays an important role in scaling operations seamlessly and maintaining cost-effectiveness in the long run. Performance issues may lead to loss of sales, especially in e-commerce and online services. Efficient web services reduce the load on servers and infrastructure, leading to cost savings in terms of resource utilization and maintenance. With the optimized performance, developers are free to experiment with new technologies and architectures without being constrained by performance bottlenecks. The performance of web services in this paper will be tested and evaluated based on the below three performance statistics: CPU Usage Memory Usage Network Usage The central issue which needs to be addressed during the communication of web services is security [1,2]. There are a lot of threats which may affect the web service security including, message alteration, loss of confidentiality, denial of services, and man-in-the-middle to name a few. In the above image, the leading workload for containers are databases and web servers. Both services require high security measures to prevent any data breaches and malicious activities. Hence showcasing the importance of security in a containerised environment. The primary aim of this paper is to investigate how various security measures affect the performance of web services hosted on Virtual Machines as well as containers. The investigation will compare the performance implications between these two hosting environments. The below two questions will be answered in this work. Q1. Do security measures impact the performance of VM-based and Container-based applications? Q2. Is there a significant difference in performance of secured VMs and Containers? To answer the above questions, the below objectives have been established: Objective 1 : Examining the impact of Security Measures on a VM-based web service. Objective 2: Examining the Impact of Security Measures on Container based Web Services Objective 3: Comparative Analysis of VM based and Container based Web Services By setting these detailed objectives, the project aims to provide a comprehensive analysis of how security measures can impact the performance of web services in different environments. The findings are expected to guide developers, IT security professionals, and system architects in making informed decisions regarding the deployment and management of secure and efficient web services. 2. Background and related works The existing literature is explored on the themes of virtualization, containerization, and the interplay between security and performance. According to VMWare [ 3 ], a Virtual Machine is a compute resource which, instead of using a physical computer, utilises software to run programs and deploy applications and web services. Instead of relying on physical hardware, VMs operate within a software layer called the hypervisor. This hypervisor sits between the hardware and the Virtual Machines, allowing the VM to run on a single physical server, each with its own operating system and applications. Containers are considered as executive lightweight, limited, executable units of software that package up the code and all its dependencies, so the application runs quickly and reliably from one environment to another [ 4 , 5 ]. Docker is described as a platform designed for developing, deploying, and running applications in containers [ 6 , 7 ]. These applications are isolated in their own containers. Virtualised applications, which are lightweight and portable can be run in these containers across different environments without the need of an entire Virtual Machine. Yadav [ 8 ] defines the below comparison points between Virtual Machines and containers. It can be noted that security becomes an overhead in containers as the OS is shared between the containers. Hence, the containers could be classified as being less secure than Virtual Machines. After a brief definition of VMs, containers, and the basic differences between them, lets discuss the common security strategies which are implemented in both. Tsifountidis [ 9 ] defines three common security strategies which are generally implemented in VMs and container-based environments [ 10 ]. Alyas [ 11 ], emphasises the importance of network segmentation in preventing unauthorised access and limiting the blast radius in case of a network breach. Network segmentation in containerisation aims to isolate the container network interface from one another, reducing the risk of attacks and network threats. Network isolation strategies for VMs and containerised environments are summarised in Table 1 . Table 1 Comparison on Network Isolation Virtual Machines Containers Typically achieved using virtual private networks and virtual LANs. These tools create segmented networks that restrict the VMs ability to interact with each other except through defined pathways. This protects sensitive data and systems from unauthorized access in case one VM is compromised. Involves configuring network namespaces that provide each container with its own isolated network stack. This means that containers do not share IP addresses, port ranges, or routing tables unless explicitly configured. Network policies can also be enforced at the container orchestration level. To control the flow of traffic between services. --------- Table 1 ----------- The impact of encryption in the web and security layers have already been explored [ 12 , 13 ]. It is also argued that the impact and overhead of security measures is not adequately quantified and factored into the designing process of web services. Table 2 provides the encryption strategies for VMs and containerised environments. Table 2 Comparison on Encryption Strategies Virtual Machines Containers Encryption in VMs can be applied at both the storage and the network levels. For storage, disk encryption like LUKS or BitLocker ensure that data remains secure at rest. For network traffic, TLS/SSL encryption is commonly used for data in transit, especially when data moves between VMs or from VMs to external systems. Secret Management tools like HashiCorp Vault or Kubernetes secrets are used to manage and store sensitive data securely. These systems encrypt the data at rest and provide secure mechanisms for access. Network encryption is also enforced using TLS to secure API endpoints and service-to-services communication. ------------ Table 2 ------------ Several key areas that help gauge the efficiency, scalability, and operational effectiveness of VMs and containers are reporter [ 14 , 15 ]. Few of the commonly evaluated performance metrics are: a. Startup time b. Resources Utilization c. Throughput and Latency The comparison of performance between Virtual Machines and Containers is a focal area in modern computing environments. Virtual Machines provide complete operating system isolation which impacts resource allocation and performance, whereas containers, by sharing their host OS, can start quicker and use fewer resources. Sultan [ 16 ] describe several security challenges for containers, such as, protection of a container from internal applications, inter-container protection, protection of the host from the containers. They also discusse both software solutions and hardware-based solutions to address these challenges. Various studies surveyed vulnerabilities in Docker images, libcontainer, Docker daemon, and orchestration tools, highlighting the security challenges in multitenant infrastructures [ 17 , 18 ]. Efforts to integrate user namespaces into Docker and Kubernetes aimed to mitigate privilege escalation by allowing them to run as non-root users. However, using namespaces also had vulnerabilities due to unexpected exposure of kernel interfaces to unprivileged users. Implementing robust security measures can impact the performance of web services. Data encryption involves computational overhead that can increase the response time and reduce the throughput. Runtime monitoring and security tools which actively scan and analyse the transactions can slow down the service response times. Network segmentation can introduce additional latency as data moves through segmented paths. The aspects which impact performance of a web application are described below [ 19 , 20 ]. a. Transport layer encryption adds additional computational overhead, increasing the response time and reducing throughput. b. Authentication and Authorization mechanisms can introduce latency, as each request may require validation against a database or an external service. c. Intrusion Detection systems and application firewalls inspect incoming traffic and outgoing traffic, introducing processing delays. d. Ensuring data integrity through hashing and checksums can add to the processing load. e. Security patches and update installation can cause temporary downtimes and require system resources, affecting performance. Containers are often thought to offer superior performance compared to Virtual Machines. However, from the analysis done by Randal [ 21 ] for different research papers, docker containers and QEMU + KVM Virtual Machines have comparable CPU and memory overhead usages, with KVM (Kernel Virtual Machines) showing a 40% higher overhead for I/O operations. This overhead was mainly due to the extra cycles per I/O operations, which could be mitigated by batching small I/O operations into larger ones. In another research paper referred by Yadav [ 8 ], no significant performance differences were found in CPU execution time and network throughput Docker containers and various types of Virtual Machines. Manco [ 22 ] demonstrate that with minor modifications to Xen’s control stack, Virtual Machine instantiation could be increased by 50 to 200 times compared to docker container instantiation. Their paper compares Virtual Machines and containers based on their security and performance. It is mentioned that due to the weaker isolation of containers as compared to Virtual Machines, it is often the case that containers are run inside Virtual Machines for better security. Containers suffer from significant security issues due to the extensive syscall API they rely on. From the above studies, it can be concluded that if a Virtual Machine can be made lighter and considering they do offer more security in terms of isolation, a Virtual Machines’ performance may become comparable to that of a container. The trade-offs were reported between different virtualization technologies in terms of performance, security, and resource management [ 23 ]. Containers offer better performance but weaker isolation, while traditional Virtual Machines provide strong security at the cost of performance. Lightweight hypervisors attempt to balance these trade-offs by combining the strengths of both containers and Virtual Machines. With the increase in research for lighter Virtual Machines it is important to note the various security aspects and how they affect each of the two helping organizations make informed decisions. According to Li [ 23 ], while containers generally offer lower performance overhead due to their lightweight nature, there are some cases where VMs performed better when handling data. This study demonstrated there are no consistent performance advantages across all metrics for either type of architecture, suggesting that the choice between containers and VMs should be influenced by specific use-case requirements rather than a one-size-fits-all approach. An in-depth examination was provided how Container based virtualization through docker compares to traditional VM based virtualization in terms of performance [ 24 ]. The Input Output performance testing has been performed by utilizing files greater than the available memory to prevent caching effects. Load Testing, Memory Utilization Testing has been performed to study the various aspects of performance of VMs and Containers. The key finding for both docker and VMs show competitive CPU handling capabilities. File I/O performance was observed to be better in VMs potentially due to the optimization of file system access in the VMs. Abuabdo [ 25 ] highlighted the necessity of considering both security and performance when choosing between virtualisation and containerisation for deploying web services or applications. 3. Proposed scheme The proposed scheme discusses the tasks that were designed to create the application and testing it based on various parameters mentioned below: 3.1. Tasks a. Application Creation and Testing b. Creation of a Performance Testing Jenkins pipeline and Monitoring tool integration. c. Running the testing pipeline to record the performance metrics. d. Applying the security measures on both the environments and running the testing pipeline e. Compare the Performance Results This approach will help in understanding the impact of security on the two different scenarios. The application will test the login/registration and the security of the users’ passwords. The encryption of the files will help understand the impact of data encryption at rest. The encryption method will determine the impact on the CPU and memory performance. Network performance will be judged in accordance to the movement of large files from the EC2 machine where the application will be hosted to the s3 bucket. To make the application more secure, network segmentation will be tested to check the performance degradation, if any. 4. Implementation The implementation of the proposed solution starts by describing the survey, the application development, workflow, monitoring, and testing. 4.1. Survey The survey “Understanding the usage of Containers and Virtual Machines” has been sent to working professionals across the industries and geographical regions. The survey uses Likert scale to ask questions maintaining consistency across the questions. These questions are used to gauge the responders’ opinions and can be useful for understanding specific feedback. The first two questions have been asked to get to know the responders and what industry they work in: Q1 Please select the industry you work in ---- Figure 1 ------------ The result shows that the responders seem to be almost equally divided across the board. This helps in getting a result from a broader area of industries. Q2 Please select the designation that best suits your role ----------------- Figure 2 ----------------- The above questions shows that while 34% of the responders are developers, the rest of the domains are fairly divided showing a diverse set of experiences in different areas. The next few questions have been asked in regards the usage of containers and VMs in their experience/day-to-day tasks Q3 Would you prefer using Virtual Machines or containers for business-critical applications? --------------- Figure 3 --------------- The above result shows that about 57% of the responders prefer using Virtual Machines for their business-critical applications. Q4 Do Virtual Machines provide better security protocols than containers? ---------------- Figure 4 ---------------- The results of this question show that there is almost an equal divide between all the options. This shows that containers and VMs provide approximately the same output to security measures. It can be said that VMs provide slightly better security protocols than the containers taking into consideration the ratio of answers for “Often” +” Always” to “Rarely” +” Never.” Q5 Do containers provide better performance than Virtual Machines? ------------- Figure 5 ------------- In this question, we can see there is a clear lean towards “Always” defining that containers do provide better performance than VMs. Q6 Do you think security measures add to the processing time of an application hosted on Virtual Machines irrespective of cloud deployment or on-premise deployment? ---------------------- Figure 6 ------------------ This question highlights that there is a major lean towards “Often.” From this the conclusion can be gathered that there is in fact an impact of security measures on performance of VMs. Since there is an impact there is a need to research more on the performance and security trade-offs, their impact, and how to mitigate any large trade-offs. Q7 Do you think security measures add to the processing time of an application hosted on Containers irrespective of cloud deployment or on-premise deployment? --------------- Figure 7 --------------- This question also helps solidify the fact that there is an impact of security on performance on containers and the need for further investigation on the topic. The below questions are asked to determine how security and performance impact the business requirements of the companies and enterprises. Q8 Does system capacity planning impact customer satisfaction? --------------- Figure 8 --------------- The above question shows how the capacity planning of a system is impacting the customer satisfaction with about 66% of the responders agreeing to it being an import part of the architecture planning phase. Q9 Should system capacity planning account for the performance impact of security measures in system design? ------------------- Figure 9 ------------------- The above question shows how the impact of security may lower the performance of an application and how the system capacity planning should take into consideration the impact of security and how much extra resource may be needed to cope with it. Q10 Is it necessary to consider security measures when determining sizes of system resources? ------------- Figure 10 ------------- Q11 In case of a failure of a business-critical application, which one of the following will be considered a priority? ------------ Figure 11 ------------ This question determines how having the application run a bit slower is acceptable in the industry, but the security of the application should be intact. If this is the case, then the trade-off between performance and security becomes a major concern considering the business may lose customers in case the application is not up to suitable standards for the clients while taking care that the application security to remain intact. 4.2. Deployment of the Application Figures 12 and 13 illustrate how the deployment of the application has been divided into two parts: a. Deployment of application without security measures on VMs and Containers (Fig. 12) b. Deployment of application with security measures on VMs and Containers (Fig. 13) ------------ Figure 12 ------------------ ----------------- Figure 13 ------------------- The basic workflow and the configuration of the application is as below: 4.2.1 The frontend The front end of the application has been created using HTML, CSS, JavaScript and has been deployed on both EC2 machines as well as docker containers. The web server has been deployed using expressjs and nodejs. 4.2.2 The Storage The S3 bucket used is a general-purpose bucket deployed on AWS with bucket versioning disabled. Each user will have a folder created with their email IDs which will contain all the files uploaded by that user. This will help keep the files separate from other users and maintain security. 4.2.3 The backend The backend of the application has been deployed through Python Flask and Gunicorn for the server deployment. The APIs responsible for uploading the files to S3 bucket, and all the interactions with the DB are created in python. PostgreSQL has been chosen as the database of choice due to its high compliance with SQL standards. The Network 4.3. Application Workflow The application workflow is depicted in Fig. 14. ------------------- Figure 14 ------------------ 4.4. Monitoring and Testing The testing of the deployment will be conducted through a Jenkins pipeline with the help of the load testing tool - Locust. The Jenkins instance which will host the testing has also been deployed on an EC2 instance in the same Subnet as the application servers. The server monitoring has been done via Prometheus and Grafana. 5. Evaluation The evaluations of the deployment include the differences between the VM-based and container-based deployments. 5.1. Scenario 1 – Without the Security Measures In this scenario, with a total of 50 users, a file of size 2.4MB has been used which will be uploaded to the application by the python API. The testing has been conducted for a time range of 150 seconds. The files will not be encrypted and will be directly pushed into the s3 bucket under the respective username folder. The PostgreSQL database is maintained in the same subnet group as the web application. 5.1.1 Virtual Machine based Application The application running on the Virtual Machine has been tested with the scenario 1 and is illustrated in Fig. 15. ----------------- Figure 15 -------------------- From the above locust charts, it can be observed that the response times for the application peaked at about 9000-9500ms, as illustrated in Fig. 16. ----------------- Figure 16 ----------------- From the above Grafana dashboard, it can be observed that the CPU usage peaked at about 40%, and the memory utilization was about 75% for the “cache + buffer” memory and about 50% for the total RAM used, with the average RAM being used at about 66% which would roughly be about 660MiB. The system was under a load percentage of 34%. 5.1.2 Container based Application The application running on the docker container has been tested as per scenario 1. ------------------- Figure 17 ---------------------- From the above locust charts shown in Fig. 17, it can be observed that the response times for the application peaked at about 6000ms. ----------------- Figure 18 -------------------- From the Grafana dashboard depicted in Fig. 18, it can be observed that the CPU utilization of the backend container, handling all the API requests, was about 6%. The memory utilization for the container peaked at 94MiB. The network usage for the containers peaked at 6Mb/s during the test runtime. 5.2. Scenario 2 – With the Security Measures In this scenario, similar to the above, a total of 50 users, and a file of size 2.4MB has been used which will be uploaded to the application by the python API. The testing has been conducted for a time range of 150 seconds. 5.2.1 Virtual Machine based Application The application running on the Virtual Machine has been tested with scenario 2. --------------------- Figure 19 ------------------------ From locust charts shown in Fig. 19, it can be observed that the response times for the application peaked at about 10000ms. -------------------- Figure 20 ------------------ From the Grafana dashboard depicted in Fig. 20, it can be observed that the CPU usage again peaked at about 40%, and the memory utilization was about 90% for the “cache + buffer” memory and about 50% for the total RAM used, with the average RAM being used at about 66% which would roughly be about 660MiB. The system was under a load percentage of 25%. 5.2.2 Container based Application ----------------- Figure 21 -------------------- From the locust charts shown in Fig. 21, it can be observed that the response times for the application peaked at about 15000ms. ------------------ Figure 22 -------------------- As per the above Grafana dashboard depicted in Fig. 22, it can be observed that the CPU utilization of the backend container, handling all the API requests, was about 10%. The memory utilization for the container peaked at 128MiB. The network usage for the containers peaked at 8Mb/s during the test runtime. 6. Analysis This analysis section will dive into the performance comparison of Virtual Machine based and container-based web application on cloud. The section will also cover how the security measures added to both Virtual Machine based and container-based web application impact the performance of both deployment scenarios[ 5 ]. The analysis will focus on three separate scenarios CPU usage, Memory usage, and network usage. The insights gained here will help determine the impact of security on performance as well as how the Virtual Machine based, and container-based application perform in comparison to one another. The EC2 machines used to conduct the tests are equipped with AWS t2.micro instance type, with an OS Image of Amazon Linux, storage of 8GB EBS connected to the EC2 machine and 1vCPU. The results have been co-related to the two research questions of this paper and the results of the testing have been provided below for each of them. Q1. Do security measures impact the performance of VM-based and Container-based applications? To answer this question, the findings for the CPU, Memory, and Network usage will be compared for both scenarios where security measures applied and not applied as illustrated in Table 3 . Table 3 Comparison without security measures Virtual Machine Containers Response Time 9000-9500ms 6000ms CPU Usage 40% 6% Memory Usage 660MiB 94MiB ------------- Table 3 ----------------- From Table 3 , considering the scenario where security measures where not applied, the Response time for containers is less than that of the Virtual Machine. The CPU usage has a significant difference with that of the Virtual Machine being 40% and that of the container being only 6%. The memory usage also shows the containers being able to handle the application processing without taking on much load. Because the Virtual Machines CPU and Memory are responsible for the OS and any other background processes running, Table 4 shows a higher percentage of resource utilization on the Virtual Machine side was expected. Table 4 Comparison with Security Measures Virtual Machine Containers Response Time 10000ms 15000ms CPU Usage 40% 10% Memory Usage 723MiB 128MiB ----------------- Table 4 ---------------- As per Table 4 , it can be seen that the container still performs better in the case of CPU and Memory usage. However, the response time of the containers is more than that of the Virtual Machine. The difference in response time in percentage can be calculated by using the formula $$\:Percentage\:Difference=\:\left(\frac{\left(x-y\right)}{\frac{x+y}{2}}\right)\times\:100\%$$ Using the above formula the percentage difference of the response time comes out to be 40%. To put into perspective, there is a 40% increase in the response time of a container as compared to that of the Virtual Machine. Considering both the above scenarios, it can be concluded that the container out-performs the Virtual Machine in almost all the aspects. The response time in the case of applied security measures being the only case where the Virtual Machine performed better than the container. Q2. Is there a significant difference in performance of secured VMs and Containers? To understand the impact of security measures on both the deployment types, the comparison for unsecured VM to secured VM, and unsecured container to secured container must be evaluated. -------------- Table 5 Table 5 Comparison between Virtual Machines Un-secured VM Secured VM Response Time 9000-9500ms 10000ms CPU Usage 40% 40% Memory Usage 66% 72.30% ------------------ Table 5 shows the difference in the response time, CPU usage, and memory usage of the two different VMs. The response time and the memory usage are higher in the secured VM. To calculate the percentage difference, the same formula can be used again. For the response time, the average value for the unsecured VM can be taken as 9250ms. Calculating the percentage difference between this and 10000ms will be roughly around 7.79%. For the Memory usage, the percentage difference comes out to be approximately 9.11%. The Memory usage being the same, no percentage difference has been calculated, and the percentage usage of 40% will be considered in the conclusion. Although there is a bit of a difference in the two scenarios, with the secured VM utilizing more resources, the percentage difference shows values less than 10%. This shows that in the case of Virtual Machines, there is indeed an impact of security measures on the performance of a web application hosted on a Virtual Machines. But the difference is not too significant. For the next scenario, performance comparison between secured containers and non-secured containers are shown in Table 6 . Table 6 Comparison between Containers Un-secured Container Secured Container Response Time 6000ms 15000ms CPU Usage 6% 10% Memory Usage 94MiB 128MiB Network Usage 6Mb/s 8Mb/s ---------------------- Table 6 ------------------------ Calculating the percentage difference between the values shown in Table 6 using the same formula. Hene the obtained results are as follows: • Response Time : approximately 85% • CPU Usage : approximately 50% • Memory Usage : approximately 30.63% • Network Usage : approximately 28.57% Those values show a significant increase in the resource usage of the container when the security measures are applied. This result shows there is a major impact of security measures on performance of web applications hosted on containers. Overall, the analysis can be concluded by stating the following: 1. There is a better performance outcome in terms of containerised application as compared to Virtual Machines if seen as a whole. The Virtual Machine performs better only in the case of response time for the application APIs. 2. There is a difference in the performance of the secured and unsecured Virtual Machine but the difference is somewhat manageable. 3. There is a significant performance and resource usage in terms of containers, with the container with security measures applied performing worse than the one where no security measures are applied. 7. Conclusions, limitations and future work In this work, the concept of cloud, virtualization, and containerization were discussed. The different deployment scenarios, literature review and the deployment process helped dive into the architectural and infrastructural setups of today’s web services in the cloud. This paper also discussed the significance of performance in today’s world and the impact of security on performance of web applications. A survey was conducted which targeted the working professionals in the IT industry. Their responses though being mixed do show that there is a known trade-off between security and performance in web applications. Through the implementation and testing, it can be observed that there is an impact of security on performance in both the cases of virtualization as well as containerization. Hence, proving the trade-off between the two to be true. The analysis done in this paper shows how much of a difference is between the two deployment scenarios. This project has been conducted on a small scale with a limited application with only a few APIs. The infrastructure chosen is limited in terms of EC2 machines though deployed on cloud. The project solely focuses on Python APIs running on Flask and Gunicorn with basic HTML, CSS, and JavaScript running on the frontend. To further enhance this project and dive deeper into the security and performance trade-off, this test can be conducted on a larger scale with a number of EC2 machines and containers load balanced to provide a more real-world scenario. The complexity of the application can be increased to further add pressure on the machines’ resources forcing it to auto scale and understand the costing of this trade-off as well. Declarations Acknowledgements (Competing Interests): No funds, grants, or other support was received. The authors have no financial or proprietary interests in any material discussed in this article. Statement on clinical trial : Not Applicable. Consent to Publish: We give our full consent for Springer Nature to publish the manuscript in Discover Computing, should it be accepted after peer review. Ethics and Consent to Participate declarations : not applicable. Ethics statement : Not Applicable. Data availability: Data available upon request. Author contribution statements A.K. conceived the presented idea and developed the theory and performed the computations. Z.E. verified the design and analytical methods as well as supervised the findings of this work. reviewed the entire work and selected the sections that needed to be focused on for publication purposes. Both authors discussed the results and contributed to the final manuscript. References Tanwar J, Sharma SK, Mittal M. Generation of Quality Web Services: A survey 10th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO). 2022 https://doi.org/10.1109/icrito56286.2022.9964560. Accessed 27 Apr 2025 Shukla A, Katt B, Yamin MM. A quantitative framework for security assurance evaluation and selection of cloud services: a case study. Int. J. Inf. Secur . 2023; 22: 1621–1650. https://doi.org/10.1007/s10207-023-00709-8 Accessed 19 May 2025 VMWare. What is a Virtual Machine? Available at https://www.vmware.com/topics/glossary/content/virtual-machine.html/ Accessed 26 May 2024 Stergiopoulos G, Dedousis P, Gritzalis D. Automatic analysis of attack graphs for risk mitigation and prioritization on large-scale and complex networks in Industry 4.0. Int. J. Inf. Secur. 2022; 21:37–59. https://doi.org/10.1007/s10207-020-00533-4 Jaiswal K. Dockerize A Spring Boot Application - Sixt Research & Development India - Medium. 2020 Available at https://medium.com/the-sixt-india-blog/dockerise-a-spring-boot-application-bdfce1d6eb15 Accessed 27 May 2025 Antunes P, Guimarães N. Guiding the implementation of data privacy with microservices. Int. J. Inf. Secur . 2024;23:3591–3608. https://doi.org/10.1007/s10207-024-00907-y Kropp A, Torre R. Docker: containerize your application. In: 2020 Elsevier eBooks; 2020. p. 231–244 https://doi.org/10.1016/b978-0-12-820488-7.00026-8 Yadav AK, Garg ML, Ritika. Docker Containers Versus Virtual Machine-Based Virtualization. Advances in Intelligent Systems and Computing, pp.141–150 (2018) http://dx.doi.org/10.1007/978-981-13-1501-5_12. Accessed 16 May 2025 Docker containers versus virtual machine-based virtualization. In: Abraham A, Dutta P, Mandal J, Bhattacharya A, Dutta S, editors. Emerging technologies in data mining and information security. Advances in intelligent systems and computing. Singapore: Springer; 2019. 814:5-12. Tsifountidis F. Virtualization Security: Virtual Machine Monitoring and Introspection. [dissertation]: University of London; 2011. Available at https://cdn.ttgtmedia.com/rms/pdf/RHUL_Tsifountidis_Final.pdf Accessed 15 Feb 2025 Patil R, Dudeja H, Modi C. Designing in-VM-assisted lightweight agent-based malware detection framework for securing virtual machines in cloud computing. Int. J. Inf. Secur. 2020;19:147–162. Alyas T, Ali S, Khan HU, Samad A, Alissa K, Saleem MA. Container performance and vulnerability management for container security using Docker engine. Shafiq M, editor. Security and communication networks; 2022. p.1–11. Available at http://dx.doi.org/10.1155/2022/6819002 Accessed 25 Feb 2025 Babatunde DJ. Evaluating the Impact of Security Measures on Performance of secure Web Applications hosted on Virtualized Platforms. Available at https://repository.uel.ac.uk/download/860cd9c0802ea8ba5d31f5df9 1e3bdbfb919eee114bac2e69ee2a03f9cb7612f/20753811/John%2520Babatunde.pdf 2015. Accessed 15 April 2025 Walid R, Joshi KP, Choi SG. Comparison of attribute-based encryption schemes in securing healthcare systems. Sci Rep. 2024;14:7147. Available at https://doi.org/10.1038/s41598-024-57692-w Accessed 15 March 2025 Fang Z, Ma H, Chen G, Hartmann S. A group genetic algorithm for energy-efficient resource allocation in container-based clouds with heterogeneous physical machines. In: Liu T, Webb G, Yue L, Wang D., editors. Advances in Artificial Intelligence. AI 2023. Lecture Notes in Computer Science. Singapore: Springer; 2023. p.14472. https://doi.org/10.1007/978-981-99-8391-9_36 Accessed 15 Feb 2025 Felter W, Ferreira A, Rajamony R, Rubio J. An updated performance comparison of virtual machines and Linux containers. IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS); 2015. p.171-172. https://ieeexplore.ieee.org/document/7095802 Accessed 30 April 2025 Sultan S, Ahmad I, Dimitriou T. Container security: Issues, challenges, and the road ahead. IEEE Access. 2019;7:52976–52996 https://doi.org/10.1109/ACCESS.2019.2911732 Malhotra R, Bansal A, Kessentini M. Deployment and performance monitoring of docker based federated learning framework for software defect prediction. Cluster Comput. 2024;27:6039–6057. https://doi.org/10.1007/s10586-024-04266-0 Martin A, Raponi S, Combe T, Di Pietro R. Docker ecosystem – Vulnerability Analysis. Computer Communications. 2018;122:30–43. Su JM. WebHOLE: Developing a web-based hands-on learning environment to assist beginners in learning web application security. Educ Inf Technol. 2024; 29:6579–6610. Iwendi DC, Kryvinska DN, Awan I-U, editors. Recent Advancements in Cloud Security using Performance Technologies and Techniques; 2022. SpringerOpen; 2022.Available at https://www.springeropen.com/collections/raicsuptat Accessed 20 May 2025 Randal A. The ideal versus the real. ACM Computing Surveys. 2020;53(1):1–31 https://doi.org/10.1145/3365199. Manco F et al. My VM is kighter (and safer) than your container. SOSP 2017. Proceedings of the 26th ACM Symposium on Operating Systems Principles; 2017 October 28–31; Shanghai, China. ACM Press; 2017. p.218–233. Li Z, Kihl M, Lu Q, Andersson JA. Performance overhead comparison between hypervisor and container based virtualization. Proceedings of the IEEE 31st International Conference on Advanced Information Networking and Applications; 2017 March 27-29; Taipei, Taiwan, IEEE; 2017. p. 955-962. http://dx.doi.org/10.1109/aina.2017.79. Kaur P, Josan JK, Neeru N. Performance analysis of Docker containerization and virtualization. In: Bindhu V, Tavares JMRS, Du KL, editors. Proceedings of Third International Conference on Communication, Computing and Electronics Systems; 2022. Lecture Notes in Electrical Engineering; Singapore: Springer; 2022. 844; p.863–877 http://dx.doi.org/10.1007/978-981-16-8862-1_56. Abuabdo A, Al-Sharif ZA. Virtualization vs. Containerization: Towards a multithreaded performance evaluation approach. AICCSA 2019. Proceedings of the IEEE/ACS 16th International Conference on Computer Systems and Applications. 2019 Nov. 3-7; Abu Dhabi, United Arab Emirates, p. 1-6. http://dx.doi.org/10.1109/aiccsa47632.2019.9035233 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7368391","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":515869980,"identity":"863bb8fa-fd35-46a3-abeb-9a6f493d8329","order_by":0,"name":"Avleen Kaur","email":"","orcid":"","institution":"Nottingham Trent University","correspondingAuthor":false,"prefix":"","firstName":"Avleen","middleName":"","lastName":"Kaur","suffix":""},{"id":515869981,"identity":"d043da44-a36e-4221-be11-650fc31debf1","order_by":1,"name":"Zoheir Ezziane","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABBElEQVRIiWNgGAWjYBAC+wYeBoYEMJOxAUjYwGUMcGlhRNOSxsBDlBYkcJiwFmb23oMPHu5hkJPvP9z4uaDifJ69RALjhx8Mh41xaWHjOZdskPCMwZix4WCz9Iwzt4t5JBKYJXsYDpvh0sIjkWMmkXCAIbGZsbFBmrftdmKPRAKDNNCFNri0SEjkmP8AaqlvY2Zs/s377xxIC/NvfFoMgLYwALUk8LAxtknzNhwAaWED2YLTYQY8Z4yBDpMwnMHD2GbNcyw5sefMwzbLHoN0nN63b+8x/PjjgI28fP/xx7d5auwS29uTD9/4UWFt2IBLD9RLyBxQnOKMlVEwCkbBKBgFxAAAnC9OhoDQjsMAAAAASUVORK5CYII=","orcid":"","institution":"Nottingham Trent University","correspondingAuthor":true,"prefix":"","firstName":"Zoheir","middleName":"","lastName":"Ezziane","suffix":""}],"badges":[],"createdAt":"2025-08-13 23:08:19","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7368391/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7368391/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":91932668,"identity":"60fe00a2-4322-4c9b-a7b6-08c2b47088c8","added_by":"auto","created_at":"2025-09-23 02:35:41","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":76416,"visible":true,"origin":"","legend":"","description":"","filename":"Draftv2withoutfigsandtables.docx","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/024425518f05f83cafddb520.docx"},{"id":91938363,"identity":"76669669-4a12-4734-ab2d-71b07b44503c","added_by":"auto","created_at":"2025-09-23 03:07:45","extension":"tif","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":253632,"visible":true,"origin":"","legend":"","description":"","filename":"Fig1.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/cbdac349c0c4739133f7961b.tif"},{"id":91932669,"identity":"823d139c-bc28-4b8d-9d2d-078041d8a655","added_by":"auto","created_at":"2025-09-23 02:35:41","extension":"tif","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":200630,"visible":true,"origin":"","legend":"","description":"","filename":"Fig10.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/ff54a518589815dadc9ad5e3.tif"},{"id":91938384,"identity":"0090bd5f-bf96-487b-a5a0-d46e96c685bc","added_by":"auto","created_at":"2025-09-23 03:08:12","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":14573,"visible":true,"origin":"","legend":"","description":"","filename":"Tab1.docx","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/5447e5e4966bc45dc45cdfdf.docx"},{"id":91932683,"identity":"cb1ff688-afbe-432b-8f25-00166e2e301e","added_by":"auto","created_at":"2025-09-23 02:35:42","extension":"tif","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":228966,"visible":true,"origin":"","legend":"","description":"","filename":"Fig11.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/1a76aee859f088195d201c80.tif"},{"id":91932677,"identity":"2677f76e-1e36-4eee-9876-d60f31dc86d7","added_by":"auto","created_at":"2025-09-23 02:35:41","extension":"docx","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":14740,"visible":true,"origin":"","legend":"","description":"","filename":"Tab2.docx","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/035a0f60150d753f45f1c470.docx"},{"id":91935269,"identity":"f0f07860-87b8-439e-a167-43d9d4afe728","added_by":"auto","created_at":"2025-09-23 02:43:42","extension":"tif","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":587248,"visible":true,"origin":"","legend":"","description":"","filename":"Fig12.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/7910f868d5142d735f871409.tif"},{"id":91937568,"identity":"7c172e7c-3923-4a9b-ab75-704f404dd7e1","added_by":"auto","created_at":"2025-09-23 02:59:41","extension":"docx","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":14623,"visible":true,"origin":"","legend":"","description":"","filename":"Tab3.docx","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/0ca1af2bc98f43c562900048.docx"},{"id":91936689,"identity":"88619b49-260a-4100-b337-4f36289f0401","added_by":"auto","created_at":"2025-09-23 02:51:42","extension":"tif","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":606648,"visible":true,"origin":"","legend":"","description":"","filename":"Fig13.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/e0564ee48daadcdfdc325fe4.tif"},{"id":91935266,"identity":"068ac79c-d8e4-45a4-93bc-6cb51734526a","added_by":"auto","created_at":"2025-09-23 02:43:41","extension":"docx","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":14622,"visible":true,"origin":"","legend":"","description":"","filename":"Tab4.docx","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/f0fda1d74a9ad8e6d4adcc2b.docx"},{"id":91932704,"identity":"068d83ad-6b32-49f1-9d42-976f351cab46","added_by":"auto","created_at":"2025-09-23 02:35:42","extension":"tif","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":231872,"visible":true,"origin":"","legend":"","description":"","filename":"Fig14.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/fb166e934590dad807c4feea.tif"},{"id":91932687,"identity":"1357165f-928d-4fbe-9bf7-1214ea77a6bf","added_by":"auto","created_at":"2025-09-23 02:35:42","extension":"docx","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":14593,"visible":true,"origin":"","legend":"","description":"","filename":"Tab5.docx","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/b0f87ab0bcf66797f1cbc1db.docx"},{"id":91932684,"identity":"32a90235-a3f6-41b8-8158-53275dfac7da","added_by":"auto","created_at":"2025-09-23 02:35:42","extension":"tif","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":186350,"visible":true,"origin":"","legend":"","description":"","filename":"Fig15.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/fb3ae7863acb89b81c15ac41.tif"},{"id":91932707,"identity":"0bb0595b-b657-4f00-b060-2428e1c00654","added_by":"auto","created_at":"2025-09-23 02:35:42","extension":"docx","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":14664,"visible":true,"origin":"","legend":"","description":"","filename":"Tab6.docx","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/4f0fc810c332a3221e183164.docx"},{"id":91935277,"identity":"e31b7ce0-32b1-41cb-a299-04072ee6dd0d","added_by":"auto","created_at":"2025-09-23 02:43:42","extension":"tif","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":360074,"visible":true,"origin":"","legend":"","description":"","filename":"Fig16.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/ecd6be09ba98b7c46fabc564.tif"},{"id":91935283,"identity":"6404312c-286c-473e-a2cf-b66b580de94d","added_by":"auto","created_at":"2025-09-23 02:43:42","extension":"tif","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":182318,"visible":true,"origin":"","legend":"","description":"","filename":"Fig17.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/ca5bec73204899c4de041702.tif"},{"id":91935278,"identity":"8f5dbf8a-e0be-43ea-81c8-371a880ae552","added_by":"auto","created_at":"2025-09-23 02:43:42","extension":"tif","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":277588,"visible":true,"origin":"","legend":"","description":"","filename":"Fig18.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/fd898a47c85519c60b87c5f0.tif"},{"id":91932697,"identity":"70b81259-8d3c-4541-8c05-570b26ab3ec0","added_by":"auto","created_at":"2025-09-23 02:35:42","extension":"tif","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":157702,"visible":true,"origin":"","legend":"","description":"","filename":"Fig19.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/427e4f2bfdff861720542df8.tif"},{"id":91932695,"identity":"476eef8c-1049-4756-ad38-76b9f55c6908","added_by":"auto","created_at":"2025-09-23 02:35:42","extension":"tif","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":233918,"visible":true,"origin":"","legend":"","description":"","filename":"Fig2.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/80e197ae3f85753150382e00.tif"},{"id":91935295,"identity":"34f093d4-cbe4-40da-aa5c-ac23721dc54e","added_by":"auto","created_at":"2025-09-23 02:43:43","extension":"tif","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":326892,"visible":true,"origin":"","legend":"","description":"","filename":"Fig20.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/d09205e066e6856a85712b4f.tif"},{"id":91932712,"identity":"23d4e167-27da-421b-ba81-307e7dd0165d","added_by":"auto","created_at":"2025-09-23 02:35:42","extension":"tif","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":154520,"visible":true,"origin":"","legend":"","description":"","filename":"Fig21.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/d5e0b72ddfd1317971da9e17.tif"},{"id":91932706,"identity":"5ac7ffcb-0416-4ef7-9c64-c59ac7f795c8","added_by":"auto","created_at":"2025-09-23 02:35:42","extension":"tif","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":312650,"visible":true,"origin":"","legend":"","description":"","filename":"Fig22.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/851a711ff74334d9e05c91a0.tif"},{"id":91937578,"identity":"46a5f457-152a-48c1-a8fc-2099dd493987","added_by":"auto","created_at":"2025-09-23 02:59:44","extension":"tif","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":174838,"visible":true,"origin":"","legend":"","description":"","filename":"Fig3.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/d0113c4f94bc8fd15d204e41.tif"},{"id":91936693,"identity":"cadb9819-0689-4237-92fa-7ff57b10581e","added_by":"auto","created_at":"2025-09-23 02:51:42","extension":"tif","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":187062,"visible":true,"origin":"","legend":"","description":"","filename":"Fig4.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/5b30bb819a35629ebaf905bc.tif"},{"id":91935351,"identity":"6bb2eecd-6d33-4acb-ae62-719406eab8b8","added_by":"auto","created_at":"2025-09-23 02:43:44","extension":"tif","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":174368,"visible":true,"origin":"","legend":"","description":"","filename":"Fig5.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/a30d866ed537bef3d3a8095f.tif"},{"id":91936694,"identity":"89eaa816-ad33-4b34-85ff-b69236a7ce6b","added_by":"auto","created_at":"2025-09-23 02:51:42","extension":"tif","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":263264,"visible":true,"origin":"","legend":"","description":"","filename":"Fig6.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/f0ac68017df07474a0a2f773.tif"},{"id":91935286,"identity":"42bcd37e-2c14-4974-a4fb-bfe8b97d4431","added_by":"auto","created_at":"2025-09-23 02:43:42","extension":"tif","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":273034,"visible":true,"origin":"","legend":"","description":"","filename":"Fig7.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/d25b90e649f64a2feece1371.tif"},{"id":91932713,"identity":"0a2411a2-b09a-43bd-8ef5-08ecb731736c","added_by":"auto","created_at":"2025-09-23 02:35:42","extension":"tif","order_by":27,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":170434,"visible":true,"origin":"","legend":"","description":"","filename":"Fig8.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/e465ff28c7de479e9c1f1e31.tif"},{"id":91932730,"identity":"dbec6c4a-f771-43b1-9a02-1565313adebf","added_by":"auto","created_at":"2025-09-23 02:35:43","extension":"tif","order_by":28,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":212450,"visible":true,"origin":"","legend":"","description":"","filename":"Fig9.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/14504b7550e9276e2f285a3a.tif"},{"id":91936698,"identity":"18bc0dcc-9e46-4dd0-a9a3-a60c048a80ae","added_by":"auto","created_at":"2025-09-23 02:51:43","extension":"json","order_by":29,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4413,"visible":true,"origin":"","legend":"","description":"","filename":"7bbebc5d64a34d2dbc585a4168472a7f.json","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/e94c6288c01fe1066607ac86.json"},{"id":91936695,"identity":"bacc8942-47d0-4738-8aea-07b5397112b1","added_by":"auto","created_at":"2025-09-23 02:51:42","extension":"xml","order_by":30,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":101333,"visible":true,"origin":"","legend":"","description":"","filename":"7bbebc5d64a34d2dbc585a4168472a7f1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/0cc5d40a4c3425ff5e63cb3b.xml"},{"id":91932719,"identity":"63bb013d-00f3-4b8f-b256-cc50e5df440e","added_by":"auto","created_at":"2025-09-23 02:35:43","extension":"tif","order_by":31,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":253632,"visible":true,"origin":"","legend":"","description":"","filename":"Fig1.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/abb7bb12c409ce5f800cf632.tif"},{"id":91935291,"identity":"087f00e8-147b-407c-abee-91ff23779508","added_by":"auto","created_at":"2025-09-23 02:43:43","extension":"tif","order_by":32,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":200630,"visible":true,"origin":"","legend":"","description":"","filename":"Fig10.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/8cd03274621e78a2b8a99a4c.tif"},{"id":91935285,"identity":"c9b60705-68a6-47ed-a8a3-f8354a65dae2","added_by":"auto","created_at":"2025-09-23 02:43:42","extension":"tif","order_by":33,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":228966,"visible":true,"origin":"","legend":"","description":"","filename":"Fig11.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/c8acd8538b494502b249ab7a.tif"},{"id":91935344,"identity":"806936e4-32f7-4ce6-b52e-bb56fd8621cd","added_by":"auto","created_at":"2025-09-23 02:43:44","extension":"tif","order_by":34,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":587248,"visible":true,"origin":"","legend":"","description":"","filename":"Fig12.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/9fda51a50c0c78afcd56f552.tif"},{"id":91932722,"identity":"9931ed02-f8ae-45a2-b0d3-f81d96ac5aa5","added_by":"auto","created_at":"2025-09-23 02:35:43","extension":"tif","order_by":35,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":606648,"visible":true,"origin":"","legend":"","description":"","filename":"Fig13.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/caa22c3e5eed827a4f3c0c25.tif"},{"id":91932721,"identity":"9bac4054-3876-49c8-80f1-b4eebd039ebc","added_by":"auto","created_at":"2025-09-23 02:35:43","extension":"tif","order_by":36,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":231872,"visible":true,"origin":"","legend":"","description":"","filename":"Fig14.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/4211d7bc17ba45bdd7d0446b.tif"},{"id":91935314,"identity":"cfbb3b91-ab0b-4844-80c2-fe6c303bcfc6","added_by":"auto","created_at":"2025-09-23 02:43:43","extension":"tif","order_by":37,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":186350,"visible":true,"origin":"","legend":"","description":"","filename":"Fig15.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/73a271c05f3dfc1a95044be8.tif"},{"id":91935307,"identity":"c0b614c6-bd49-4e37-b910-c30239cdf7bd","added_by":"auto","created_at":"2025-09-23 02:43:43","extension":"tif","order_by":38,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":360074,"visible":true,"origin":"","legend":"","description":"","filename":"Fig16.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/b237b811e5e7ff368925b819.tif"},{"id":91936701,"identity":"b6c3d23a-9821-4d73-916f-c24600a5b8cc","added_by":"auto","created_at":"2025-09-23 02:51:43","extension":"tif","order_by":39,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":182318,"visible":true,"origin":"","legend":"","description":"","filename":"Fig17.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/43881f7d400c379e209061a3.tif"},{"id":91936715,"identity":"6568924b-6c35-4b84-9f80-a68b14ee60c2","added_by":"auto","created_at":"2025-09-23 02:51:44","extension":"tif","order_by":40,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":277588,"visible":true,"origin":"","legend":"","description":"","filename":"Fig18.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/299425b5f010eb349d40bc17.tif"},{"id":91935301,"identity":"c080ad6a-e79b-4f4f-8630-346a4be1f03a","added_by":"auto","created_at":"2025-09-23 02:43:43","extension":"tif","order_by":41,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":157702,"visible":true,"origin":"","legend":"","description":"","filename":"Fig19.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/32b9726bac8dbd8423b32798.tif"},{"id":91932717,"identity":"5b27953e-f821-4a29-a6de-b29ba736b5fe","added_by":"auto","created_at":"2025-09-23 02:35:42","extension":"tif","order_by":42,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":233918,"visible":true,"origin":"","legend":"","description":"","filename":"Fig2.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/c1f5ddd0544071210266ba8b.tif"},{"id":91932724,"identity":"71f67070-6ad7-4067-bc6b-69e02d1b59fc","added_by":"auto","created_at":"2025-09-23 02:35:43","extension":"tif","order_by":43,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":326892,"visible":true,"origin":"","legend":"","description":"","filename":"Fig20.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/9938d624570a837f4e848ae8.tif"},{"id":91935293,"identity":"64b5394a-cd0d-4fde-8e43-3412fa236ab3","added_by":"auto","created_at":"2025-09-23 02:43:43","extension":"tif","order_by":44,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":154520,"visible":true,"origin":"","legend":"","description":"","filename":"Fig21.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/63b4dced1e9498c1d0fe0699.tif"},{"id":91935292,"identity":"496aede0-9d4b-495d-9aae-15801c97b102","added_by":"auto","created_at":"2025-09-23 02:43:43","extension":"tif","order_by":45,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":312650,"visible":true,"origin":"","legend":"","description":"","filename":"Fig22.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/dd0b198a27f6e483543b21b0.tif"},{"id":91936714,"identity":"2ab6d41a-ccf5-4e5c-98c4-0930576dabe6","added_by":"auto","created_at":"2025-09-23 02:51:44","extension":"tif","order_by":46,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":174838,"visible":true,"origin":"","legend":"","description":"","filename":"Fig3.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/def20488fc17b0c5f5dd8554.tif"},{"id":91936700,"identity":"3cc63046-8262-47da-bfb3-af85c10733b0","added_by":"auto","created_at":"2025-09-23 02:51:43","extension":"tif","order_by":47,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":187062,"visible":true,"origin":"","legend":"","description":"","filename":"Fig4.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/a3b4430e5b79341c726ee01e.tif"},{"id":91932733,"identity":"51b65329-f18e-4bab-b670-6054de366d41","added_by":"auto","created_at":"2025-09-23 02:35:43","extension":"tif","order_by":48,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":174368,"visible":true,"origin":"","legend":"","description":"","filename":"Fig5.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/3a2c213ebde503b194181d9c.tif"},{"id":91936717,"identity":"f53e18d5-888c-4e80-b427-3a0f718777ad","added_by":"auto","created_at":"2025-09-23 02:51:44","extension":"tif","order_by":49,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":263264,"visible":true,"origin":"","legend":"","description":"","filename":"Fig6.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/d9e12b98f05e406a7c0f4098.tif"},{"id":91937573,"identity":"063aa6a3-07e4-4297-aa2a-042dc46cfff1","added_by":"auto","created_at":"2025-09-23 02:59:43","extension":"tif","order_by":50,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":273034,"visible":true,"origin":"","legend":"","description":"","filename":"Fig7.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/ac1ea0fc8f20b0d75db907cb.tif"},{"id":91936696,"identity":"43ec134b-9f54-43eb-aed6-36daac48ca94","added_by":"auto","created_at":"2025-09-23 02:51:42","extension":"tif","order_by":51,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":170434,"visible":true,"origin":"","legend":"","description":"","filename":"Fig8.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/0502dbaf12d6bea35315e57f.tif"},{"id":91936708,"identity":"e5e1f5ac-36d2-445c-a174-40af52e31868","added_by":"auto","created_at":"2025-09-23 02:51:43","extension":"tif","order_by":52,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":212450,"visible":true,"origin":"","legend":"","description":"","filename":"Fig9.tif","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/71ff62095e8df4f5e27ce05f.tif"},{"id":91932736,"identity":"01d370ab-d21c-4b84-9999-9e8f31865c31","added_by":"auto","created_at":"2025-09-23 02:35:43","extension":"png","order_by":53,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":39096,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig1.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/4a09b780ebc44d5c88d996e3.png"},{"id":91935318,"identity":"7d323bd3-3c7c-4ff9-ac54-5c7a30795f5b","added_by":"auto","created_at":"2025-09-23 02:43:43","extension":"png","order_by":54,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":40165,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig10.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/a097efdf260c2d89c440fddb.png"},{"id":91935288,"identity":"8b20b845-0370-47b8-8641-137d6cb18144","added_by":"auto","created_at":"2025-09-23 02:43:42","extension":"png","order_by":55,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":43655,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig11.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/83e1265b06aeaa26a54f69f6.png"},{"id":91937572,"identity":"1ba1b330-2ac0-4c26-b4c7-f5edad5715b7","added_by":"auto","created_at":"2025-09-23 02:59:42","extension":"png","order_by":56,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":32611,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig12.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/9ffb4d18bc0da40544c192aa.png"},{"id":91935294,"identity":"b453849f-228e-44ad-9b0f-d4806ed650a3","added_by":"auto","created_at":"2025-09-23 02:43:43","extension":"png","order_by":57,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":33881,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig13.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/f4415d3e8dcd1dc09ba8be28.png"},{"id":91932766,"identity":"7d70be64-1aa1-4688-93fe-cccb9e372ab6","added_by":"auto","created_at":"2025-09-23 02:35:44","extension":"png","order_by":58,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":35598,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig14.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/94c22ba009e10ed9906b8efd.png"},{"id":91935303,"identity":"1eb1257e-f91e-4eed-b820-1f7a990426d9","added_by":"auto","created_at":"2025-09-23 02:43:43","extension":"png","order_by":59,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":33401,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig15.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/48c0a2eff56d137df4b2ddd4.png"},{"id":91932746,"identity":"4a7d368d-f546-41db-af33-0ef05b0f8e51","added_by":"auto","created_at":"2025-09-23 02:35:43","extension":"png","order_by":60,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":50537,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig16.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/936444ec091cdbef392dd9e9.png"},{"id":91932751,"identity":"ad4fd367-b0bc-452d-bc13-db742d8a00ee","added_by":"auto","created_at":"2025-09-23 02:35:44","extension":"png","order_by":61,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":33032,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig17.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/fc455de87733ad3f5669ec62.png"},{"id":91937577,"identity":"9761985e-f895-4dd6-b1cd-0fc87e03ee74","added_by":"auto","created_at":"2025-09-23 02:59:44","extension":"png","order_by":62,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":39112,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig18.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/a279242f784157c5ab9d57ff.png"},{"id":91932757,"identity":"92a3bab8-977c-4ae6-8eda-f2c5326a191b","added_by":"auto","created_at":"2025-09-23 02:35:44","extension":"png","order_by":63,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":28928,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig19.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/32b1d4158831b2caefe6a69f.png"},{"id":91932756,"identity":"6ecfc07f-a322-46ce-b6b6-76e6b3545bd1","added_by":"auto","created_at":"2025-09-23 02:35:44","extension":"png","order_by":64,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":35322,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig2.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/d11f12c080cd8e1c39c6efa2.png"},{"id":91932747,"identity":"ade896da-e569-428e-82a3-0cb9b0d2c72c","added_by":"auto","created_at":"2025-09-23 02:35:43","extension":"png","order_by":65,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":48922,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig20.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/7b7cda0cfde87dc3b821ea43.png"},{"id":91932753,"identity":"59658d1e-9cf0-44d6-9ae0-176c6fc144d7","added_by":"auto","created_at":"2025-09-23 02:35:44","extension":"png","order_by":66,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":28984,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig21.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/95ea6fdc28f5e0e5ee47d57f.png"},{"id":91932763,"identity":"8eab8915-1e5c-4f53-9182-1a6acbd94bd3","added_by":"auto","created_at":"2025-09-23 02:35:44","extension":"png","order_by":67,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":45209,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig22.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/8b965578970d41978bf02fa6.png"},{"id":91932758,"identity":"e317e5cd-a507-4bef-b169-f72802ca4208","added_by":"auto","created_at":"2025-09-23 02:35:44","extension":"png","order_by":68,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":33757,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig3.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/250bd7a34268c9a97761097d.png"},{"id":91936712,"identity":"b79a4131-7fb4-469f-b7af-f957e52e0a67","added_by":"auto","created_at":"2025-09-23 02:51:44","extension":"png","order_by":69,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":39010,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig4.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/993237313a31fea7b7a49cb0.png"},{"id":91935337,"identity":"b6675a0a-4cd5-4aea-94a8-44a5976105b3","added_by":"auto","created_at":"2025-09-23 02:43:44","extension":"png","order_by":70,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":35751,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig5.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/814d3bb9ee6a35e92b78ddf5.png"},{"id":91935353,"identity":"0eeb0781-2793-4f37-97c0-9da208263a4d","added_by":"auto","created_at":"2025-09-23 02:43:44","extension":"png","order_by":71,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":50824,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig6.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/c5246fde1f839a6f1bb3e31b.png"},{"id":91935305,"identity":"8847287b-bbe6-41ad-837b-a8c35c53f458","added_by":"auto","created_at":"2025-09-23 02:43:43","extension":"png","order_by":72,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":53036,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig7.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/0d0e6a0a28f2c1e5c49c8e53.png"},{"id":91932743,"identity":"fe0a0ddb-b8d2-417b-8724-e9a6cdb51e0c","added_by":"auto","created_at":"2025-09-23 02:35:43","extension":"png","order_by":73,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":35428,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig8.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/bca713aff046deddf0618353.png"},{"id":91932748,"identity":"b26a512c-f066-4bea-897b-682fdf6d191a","added_by":"auto","created_at":"2025-09-23 02:35:44","extension":"png","order_by":74,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":42484,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig9.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/79cc54297ba26bdfd4506266.png"},{"id":91935348,"identity":"5fba3504-cfa2-4ddf-88c4-101101898863","added_by":"auto","created_at":"2025-09-23 02:43:44","extension":"xml","order_by":75,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":96855,"visible":true,"origin":"","legend":"","description":"","filename":"7bbebc5d64a34d2dbc585a4168472a7f1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/fe513d0f217342ddb68918af.xml"},{"id":91932728,"identity":"8f221662-7a48-4546-ae4d-ef5b078c211f","added_by":"auto","created_at":"2025-09-23 02:35:43","extension":"html","order_by":76,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":114777,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/6901d4e15546d614d7ea3671.html"},{"id":91935272,"identity":"23466099-e7bd-4f49-bb93-67ed35e41c3d","added_by":"auto","created_at":"2025-09-23 02:43:42","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":357398,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"Fig1.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/3dcedad6b763d9390671f72a.png"},{"id":91932674,"identity":"17c4c775-ee5a-4552-8e3d-3f88aefb032f","added_by":"auto","created_at":"2025-09-23 02:35:41","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":320918,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"Fig2.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/5e6bc3daff81014d6fcfb082.png"},{"id":91932671,"identity":"ce6a7198-ff5c-4387-89c1-fa477369dde5","added_by":"auto","created_at":"2025-09-23 02:35:41","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":233137,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"Fig3.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/0032847ff14bda1011b22953.png"},{"id":91936684,"identity":"9fa54f63-229a-470b-9f93-7da435fd5250","added_by":"auto","created_at":"2025-09-23 02:51:41","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":259120,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"Fig4.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/e51a2dde0024f13f490f4c37.png"},{"id":91935262,"identity":"e10cbe98-6011-44c8-bc3f-f42587d4b943","added_by":"auto","created_at":"2025-09-23 02:43:41","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":238453,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"Fig5.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/845d3c99afa149e6ce7e73f5.png"},{"id":91936692,"identity":"4e5248e2-fa18-42a4-a35f-23a19165e5f1","added_by":"auto","created_at":"2025-09-23 02:51:42","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":344004,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"Fig6.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/ed7f43901e1bffa2c509df73.png"},{"id":91932672,"identity":"6347e2dd-cc6e-46ba-a864-fa59ac0e9fe4","added_by":"auto","created_at":"2025-09-23 02:35:41","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":361265,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"Fig7.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/aafc264d2dde2a95f083b956.png"},{"id":91938362,"identity":"164c7d83-0a6d-49e5-b06a-a1cba7fd9c3c","added_by":"auto","created_at":"2025-09-23 03:07:45","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":235883,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"Fig8.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/75b9d1620eae10a06297ec36.png"},{"id":91932681,"identity":"188a25c7-b7c7-405b-a05e-f923008e6d59","added_by":"auto","created_at":"2025-09-23 02:35:41","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":290819,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"Fig9.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/c8daa6e753f96c2e3f3b8220.png"},{"id":91935270,"identity":"d4a7c86c-8300-4e7d-acbc-ac279acd2000","added_by":"auto","created_at":"2025-09-23 02:43:42","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":274663,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"Fig10.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/733f1814b794d15f110f70f5.png"},{"id":91935274,"identity":"9e4de5dd-a238-4ee0-9d6b-835310da7fd9","added_by":"auto","created_at":"2025-09-23 02:43:42","extension":"png","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":301147,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"Fig11.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/d8fbb60dd7a916add50b89ae.png"},{"id":91932691,"identity":"89946ddd-80fd-4f49-978f-96f1846754f5","added_by":"auto","created_at":"2025-09-23 02:35:42","extension":"png","order_by":12,"title":"Figure 12","display":"","copyAsset":false,"role":"figure","size":551888,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"Fig12.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/8b296a87dae09c5ed5adf40a.png"},{"id":91935279,"identity":"52aaf792-a1a4-42f3-8f5d-ca6f07ab518d","added_by":"auto","created_at":"2025-09-23 02:43:42","extension":"png","order_by":13,"title":"Figure 13","display":"","copyAsset":false,"role":"figure","size":571321,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"Fig13.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/5f7252f4eefef81eb57ce890.png"},{"id":91932703,"identity":"9fa44d68-8c54-4f9f-999d-dd1c882908f8","added_by":"auto","created_at":"2025-09-23 02:35:42","extension":"png","order_by":14,"title":"Figure 14","display":"","copyAsset":false,"role":"figure","size":316158,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"Fig14.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/ad385f73619002ce1ee87ef0.png"},{"id":91935273,"identity":"84de63ce-587c-47d7-b90e-f6895bb6b85e","added_by":"auto","created_at":"2025-09-23 02:43:42","extension":"png","order_by":15,"title":"Figure 15","display":"","copyAsset":false,"role":"figure","size":257347,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"Fig15.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/c74a52c6147de97d0fbcb356.png"},{"id":91936690,"identity":"817d9e5f-29ce-4729-a570-d50e99be9779","added_by":"auto","created_at":"2025-09-23 02:51:42","extension":"png","order_by":16,"title":"Figure 16","display":"","copyAsset":false,"role":"figure","size":478718,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"Fig16.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/6769f3b38a55b6d075593243.png"},{"id":91932700,"identity":"12fb67bb-e9f8-4af8-9c31-12d834e77627","added_by":"auto","created_at":"2025-09-23 02:35:42","extension":"png","order_by":17,"title":"Figure 17","display":"","copyAsset":false,"role":"figure","size":252654,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"Fig17.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/051ccd5bbd153d2d9d20bd9e.png"},{"id":91932688,"identity":"41b2397e-0044-4f3b-96e9-3adcb69e7c0d","added_by":"auto","created_at":"2025-09-23 02:35:42","extension":"png","order_by":18,"title":"Figure 18","display":"","copyAsset":false,"role":"figure","size":394496,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"Fig18.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/e368ee6c17d2c3e0091bd8da.png"},{"id":91935275,"identity":"c54e65da-a6fe-4fff-8543-70fe84e50509","added_by":"auto","created_at":"2025-09-23 02:43:42","extension":"png","order_by":19,"title":"Figure 19","display":"","copyAsset":false,"role":"figure","size":228705,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"Fig19.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/b870ced9c862ea1d3c1aabdb.png"},{"id":91932739,"identity":"2c167882-ed72-4b8e-a454-d0f78676ac42","added_by":"auto","created_at":"2025-09-23 02:35:43","extension":"png","order_by":20,"title":"Figure 20","display":"","copyAsset":false,"role":"figure","size":460586,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"Fig20.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/462848c14ca6680612c0d70a.png"},{"id":91935340,"identity":"323a3d48-97f8-49e2-963d-ad5b0b7700f5","added_by":"auto","created_at":"2025-09-23 02:43:44","extension":"png","order_by":21,"title":"Figure 21","display":"","copyAsset":false,"role":"figure","size":223248,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"Fig21.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/5f6a2efba27cc0cf93d342e9.png"},{"id":91935281,"identity":"4ac87c96-42b4-485d-9899-8372331ce11f","added_by":"auto","created_at":"2025-09-23 02:43:42","extension":"png","order_by":22,"title":"Figure 22","display":"","copyAsset":false,"role":"figure","size":459827,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"Fig22.png","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/7b5ae95f0f6c39781976d271.png"},{"id":96045018,"identity":"356a5855-b736-41ee-9be3-5562782f0efe","added_by":"auto","created_at":"2025-11-17 05:24:02","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":8044243,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7368391/v1/6eacb935-ab9b-4ece-8329-ec5b9995ef9e.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"\u003cp\u003eEvaluation of Impact of Security Measures on Container-based and VM-based Application Performance in Cloud\u003c/p\u003e","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eIn today\u0026rsquo;s rapidly evolving digital landscape, web services have become the basis for businesses and individuals aiming to deliver several online services. They facilitate a wide range of daily interactions, from checking the weather, to more complex tasks. These services are increasingly being deployed to cloud platforms. This helps provide scalability, flexibility, and cost efficiency, allowing organizations to dynamically adjust resources in response to fluctuating demands.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eCloud service providers, such as, Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform (GCP) are leveraged by the organizations to deploy web applications.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eNew considerations for performance and security are introduced while deploying web services to the cloud. There is a necessity for a balance between scalability and the management of overheads introduced by virtualized technologies.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn the last few years, there has been a significant shift towards web services, driven by the need for higher scalability, accessibility, and flexibility in web solutions. The adoption for cloud-based architectures, such as microservices has increased which ensures the seamless integration and maintenance of applications across various platforms. This has led to a significant increase in the use of APIs, serverless computing, and containerisation technologies.\u003c/p\u003e\n\u003cp\u003eThe quality of a web service has been defined as a non-functional characteristic [1]. Meaning, factors like total execution time taken by a service, cost of invoking the service, availability, and security features need to be analysed.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIf we explore the different types of web services, it becomes clear that the selection of the type of web services influences the architectural design and functionality of applications. It introduces the topic of performance of web services and how it becomes crucial to its adoption. Performance enhances user satisfaction by providing faster response times and reliable interactions. It also plays an important role in scaling operations seamlessly and maintaining cost-effectiveness in the long run. Performance issues may lead to loss of sales, especially in e-commerce and online services. Efficient web services reduce the load on servers and infrastructure, leading to cost savings in terms of resource utilization and maintenance. With the optimized performance, developers are free to experiment with new technologies and architectures without being constrained by performance bottlenecks.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe performance of web services in this paper will be tested and evaluated based on the below three performance statistics:\u003c/p\u003e\n\u003col\u003e\n \u003cli\u003eCPU Usage\u003c/li\u003e\n \u003cli\u003eMemory Usage\u003c/li\u003e\n \u003cli\u003eNetwork Usage\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eThe central issue which needs to be addressed during the communication of web services is security [1,2]. There are a lot of threats which may affect the web service security including, message alteration, loss of confidentiality, denial of services, and man-in-the-middle to name a few.\u003c/p\u003e\n\u003cp\u003eIn the above image, the leading workload for containers are databases and web servers. Both services require high security measures to prevent any data breaches and malicious activities. Hence showcasing the importance of security in a containerised environment. \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe primary aim of this paper is to investigate how various security measures affect the performance of web services hosted on Virtual Machines as well as containers. The investigation will compare the performance implications between these two hosting environments. The below two questions will be answered in this work.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eQ1.\u003c/strong\u003e Do security measures impact the performance of VM-based and Container-based applications?\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eQ2.\u0026nbsp;\u003c/strong\u003eIs there a significant difference in performance of secured VMs and Containers?\u003c/p\u003e\n\u003cp\u003eTo answer the above questions, the below objectives have been established:\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eObjective 1\u003c/strong\u003e: Examining the impact of Security Measures on a VM-based web service.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eObjective 2:\u0026nbsp;\u003c/strong\u003eExamining the Impact of Security Measures on Container based Web Services\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eObjective 3:\u0026nbsp;\u003c/strong\u003eComparative Analysis of VM based and Container based Web Services\u003c/p\u003e\n\u003cp\u003eBy setting these detailed objectives, the project aims to provide a comprehensive analysis of how security measures can impact the performance of web services in different environments. The findings are expected to guide developers, IT security professionals, and system architects in making informed decisions regarding the deployment and management of secure and efficient web services.\u003c/p\u003e"},{"header":"2. Background and related works","content":"\u003cp\u003eThe existing literature is explored on the themes of virtualization, containerization, and the interplay between security and performance.\u003c/p\u003e\u003cp\u003eAccording to VMWare [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e], a Virtual Machine is a compute resource which, instead of using a physical computer, utilises software to run programs and deploy applications and web services. Instead of relying on physical hardware, VMs operate within a software layer called the hypervisor. This hypervisor sits between the hardware and the Virtual Machines, allowing the VM to run on a single physical server, each with its own operating system and applications.\u003c/p\u003e\u003cp\u003eContainers are considered as executive lightweight, limited, executable units of software that package up the code and all its dependencies, so the application runs quickly and reliably from one environment to another [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eDocker is described as a platform designed for developing, deploying, and running applications in containers [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. These applications are isolated in their own containers. Virtualised applications, which are lightweight and portable can be run in these containers across different environments without the need of an entire Virtual Machine.\u003c/p\u003e\u003cp\u003eYadav [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e] defines the below comparison points between Virtual Machines and containers. It can be noted that security becomes an overhead in containers as the OS is shared between the containers. Hence, the containers could be classified as being less secure than Virtual Machines.\u003c/p\u003e\u003cp\u003eAfter a brief definition of VMs, containers, and the basic differences between them, lets discuss the common security strategies which are implemented in both. Tsifountidis [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e] defines three common security strategies which are generally implemented in VMs and container-based environments [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eAlyas [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e], emphasises the importance of network segmentation in preventing unauthorised access and limiting the blast radius in case of a network breach. Network segmentation in containerisation aims to isolate the container network interface from one another, reducing the risk of attacks and network threats.\u003c/p\u003e\u003cp\u003eNetwork isolation strategies for VMs and containerised environments are summarised in Table \u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eComparison on Network Isolation\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"2\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVirtual Machines\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eContainers\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTypically achieved using virtual private networks and virtual LANs. These tools create segmented networks that restrict the VMs ability to interact with each other except through defined pathways. This protects sensitive data and systems from unauthorized access in case one VM is compromised.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eInvolves configuring network namespaces that provide each container with its own isolated network stack. This means that containers do not share IP addresses, port ranges, or routing tables unless explicitly configured. Network policies can also be enforced at the container orchestration level. To control the flow of traffic between services.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e---------\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e\u003c/p\u003e\u003cp\u003e-----------\u003c/p\u003e\u003cp\u003eThe impact of encryption in the web and security layers have already been explored [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. It is also argued that the impact and overhead of security measures is not adequately quantified and factored into the designing process of web services.\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e provides the encryption strategies for VMs and containerised environments.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eComparison on Encryption Strategies\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"2\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVirtual Machines\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eContainers\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEncryption in VMs can be applied at both the storage and the network levels.\u003c/p\u003e\u003cp\u003eFor storage, disk encryption like LUKS or BitLocker ensure that data remains secure at rest. For network traffic, TLS/SSL encryption is commonly used for data in transit, especially when data moves between VMs or from VMs to external systems.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSecret Management tools like HashiCorp Vault or Kubernetes secrets are used to manage and store sensitive data securely. These systems encrypt the data at rest and provide secure mechanisms for access. Network encryption is also enforced using TLS to secure API endpoints and service-to-services communication.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e------------\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e\u003c/p\u003e\u003cp\u003e------------\u003c/p\u003e\u003cp\u003eSeveral key areas that help gauge the efficiency, scalability, and operational effectiveness of VMs and containers are reporter [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Few of the commonly evaluated performance metrics are:\u003c/p\u003e\u003cp\u003ea. Startup time\u003c/p\u003e\u003cp\u003eb. Resources Utilization\u003c/p\u003e\u003cp\u003ec. Throughput and Latency\u003c/p\u003e\u003cp\u003eThe comparison of performance between Virtual Machines and Containers is a focal area in modern computing environments. Virtual Machines provide complete operating system isolation which impacts resource allocation and performance, whereas containers, by sharing their host OS, can start quicker and use fewer resources.\u003c/p\u003e\u003cp\u003eSultan [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e] describe several security challenges for containers, such as, protection of a container from internal applications, inter-container protection, protection of the host from the containers. They also discusse both software solutions and hardware-based solutions to address these challenges.\u003c/p\u003e\u003cp\u003eVarious studies surveyed vulnerabilities in Docker images, libcontainer, Docker daemon, and orchestration tools, highlighting the security challenges in multitenant infrastructures [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Efforts to integrate user namespaces into Docker and Kubernetes aimed to mitigate privilege escalation by allowing them to run as non-root users. However, using namespaces also had vulnerabilities due to unexpected exposure of kernel interfaces to unprivileged users.\u003c/p\u003e\u003cp\u003eImplementing robust security measures can impact the performance of web services. Data encryption involves computational overhead that can increase the response time and reduce the throughput. Runtime monitoring and security tools which actively scan and analyse the transactions can slow down the service response times. Network segmentation can introduce additional latency as data moves through segmented paths.\u003c/p\u003e\u003cp\u003eThe aspects which impact performance of a web application are described below [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e].\u003c/p\u003e\u003cp\u003ea. Transport layer encryption adds additional computational overhead, increasing the response time and reducing throughput.\u003c/p\u003e\u003cp\u003eb. Authentication and Authorization mechanisms can introduce latency, as each request may require validation against a database or an external service.\u003c/p\u003e\u003cp\u003ec. Intrusion Detection systems and application firewalls inspect incoming traffic and outgoing traffic, introducing processing delays.\u003c/p\u003e\u003cp\u003ed. Ensuring data integrity through hashing and checksums can add to the processing load.\u003c/p\u003e\u003cp\u003ee. Security patches and update installation can cause temporary downtimes and require system resources, affecting performance.\u003c/p\u003e\u003cp\u003eContainers are often thought to offer superior performance compared to Virtual Machines. However, from the analysis done by Randal [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e] for different research papers, docker containers and QEMU\u0026thinsp;+\u0026thinsp;KVM Virtual Machines have comparable CPU and memory overhead usages, with KVM (Kernel Virtual Machines) showing a 40% higher overhead for I/O operations. This overhead was mainly due to the extra cycles per I/O operations, which could be mitigated by batching small I/O operations into larger ones. In another research paper referred by Yadav [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e], no significant performance differences were found in CPU execution time and network throughput Docker containers and various types of Virtual Machines.\u003c/p\u003e\u003cp\u003eManco [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e] demonstrate that with minor modifications to Xen\u0026rsquo;s control stack, Virtual Machine instantiation could be increased by 50 to 200 times compared to docker container instantiation. Their paper compares Virtual Machines and containers based on their security and performance. It is mentioned that due to the weaker isolation of containers as compared to Virtual Machines, it is often the case that containers are run inside Virtual Machines for better security. Containers suffer from significant security issues due to the extensive syscall API they rely on.\u003c/p\u003e\u003cp\u003eFrom the above studies, it can be concluded that if a Virtual Machine can be made lighter and considering they do offer more security in terms of isolation, a Virtual Machines\u0026rsquo; performance may become comparable to that of a container.\u003c/p\u003e\u003cp\u003eThe trade-offs were reported between different virtualization technologies in terms of performance, security, and resource management [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. Containers offer better performance but weaker isolation, while traditional Virtual Machines provide strong security at the cost of performance. Lightweight hypervisors attempt to balance these trade-offs by combining the strengths of both containers and Virtual Machines.\u003c/p\u003e\u003cp\u003eWith the increase in research for lighter Virtual Machines it is important to note the various security aspects and how they affect each of the two helping organizations make informed decisions.\u003c/p\u003e\u003cp\u003eAccording to Li [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e], while containers generally offer lower performance overhead due to their lightweight nature, there are some cases where VMs performed better when handling data. This study demonstrated there are no consistent performance advantages across all metrics for either type of architecture, suggesting that the choice between containers and VMs should be influenced by specific use-case requirements rather than a one-size-fits-all approach.\u003c/p\u003e\u003cp\u003eAn in-depth examination was provided how Container based virtualization through docker compares to traditional VM based virtualization in terms of performance [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. The Input Output performance testing has been performed by utilizing files greater than the available memory to prevent caching effects. Load Testing, Memory Utilization Testing has been performed to study the various aspects of performance of VMs and Containers. The key finding for both docker and VMs show competitive CPU handling capabilities. File I/O performance was observed to be better in VMs potentially due to the optimization of file system access in the VMs.\u003c/p\u003e\u003cp\u003eAbuabdo [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e] highlighted the necessity of considering both security and performance when choosing between virtualisation and containerisation for deploying web services or applications.\u003c/p\u003e"},{"header":"3. Proposed scheme","content":"\u003cp\u003eThe proposed scheme discusses the tasks that were designed to create the application and testing it based on various parameters mentioned below:\u003c/p\u003e\u003cp\u003e3.1. Tasks\u003c/p\u003e\u003cp\u003ea. Application Creation and Testing\u003c/p\u003e\u003cp\u003eb. Creation of a Performance Testing Jenkins pipeline and Monitoring tool integration.\u003c/p\u003e\u003cp\u003ec. Running the testing pipeline to record the performance metrics.\u003c/p\u003e\u003cp\u003ed. Applying the security measures on both the environments and running the testing pipeline\u003c/p\u003e\u003cp\u003ee. Compare the Performance Results\u003c/p\u003e\u003cp\u003eThis approach will help in understanding the impact of security on the two different scenarios. The application will test the login/registration and the security of the users\u0026rsquo; passwords. The encryption of the files will help understand the impact of data encryption at rest. The encryption method will determine the impact on the CPU and memory performance. Network performance will be judged in accordance to the movement of large files from the EC2 machine where the application will be hosted to the s3 bucket. To make the application more secure, network segmentation will be tested to check the performance degradation, if any.\u003c/p\u003e"},{"header":"4. Implementation","content":"\u003cp\u003eThe implementation of the proposed solution starts by describing the survey, the application development, workflow, monitoring, and testing.\u003c/p\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e4.1. Survey\u003c/h2\u003e\u003cp\u003eThe survey \u0026ldquo;Understanding the usage of Containers and Virtual Machines\u0026rdquo; has been sent to working professionals across the industries and geographical regions.\u003c/p\u003e\u003cp\u003eThe survey uses Likert scale to ask questions maintaining consistency across the questions. These questions are used to gauge the responders\u0026rsquo; opinions and can be useful for understanding specific feedback.\u003c/p\u003e\u003cp\u003eThe first two questions have been asked to get to know the responders and what industry they work in:\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eQ1\u003c/strong\u003e\u003cp\u003ePlease select the industry you work in\u003c/p\u003e\u003c/p\u003e\u003cp\u003e----\u003c/p\u003e\u003cp\u003eFigure 1\u003c/p\u003e\u003cp\u003e------------\u003c/p\u003e\u003cp\u003eThe result shows that the responders seem to be almost equally divided across the board. This helps in getting a result from a broader area of industries.\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eQ2\u003c/strong\u003e\u003cp\u003ePlease select the designation that best suits your role\u003c/p\u003e\u003c/p\u003e\u003cp\u003e-----------------\u003c/p\u003e\u003cp\u003eFigure 2\u003c/p\u003e\u003cp\u003e-----------------\u003c/p\u003e\u003cp\u003eThe above questions shows that while 34% of the responders are developers, the rest of the domains are fairly divided showing a diverse set of experiences in different areas.\u003c/p\u003e\u003cp\u003eThe next few questions have been asked in regards the usage of containers and VMs in their experience/day-to-day tasks\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eQ3\u003c/strong\u003e\u003cp\u003eWould you prefer using Virtual Machines or containers for business-critical applications?\u003c/p\u003e\u003c/p\u003e\u003cp\u003e---------------\u003c/p\u003e\u003cp\u003eFigure 3\u003c/p\u003e\u003cp\u003e\u003cb\u003e---------------\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe above result shows that about 57% of the responders prefer using Virtual Machines for their business-critical applications.\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eQ4\u003c/strong\u003e\u003cp\u003eDo Virtual Machines provide better security protocols than containers?\u003c/p\u003e\u003c/p\u003e\u003cp\u003e----------------\u003c/p\u003e\u003cp\u003eFigure 4\u003c/p\u003e\u003cp\u003e\u003cb\u003e----------------\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe results of this question show that there is almost an equal divide between all the options. This shows that containers and VMs provide approximately the same output to security measures. It can be said that VMs provide slightly better security protocols than the containers taking into consideration the ratio of answers for \u0026ldquo;Often\u0026rdquo; +\u0026rdquo; Always\u0026rdquo; to \u0026ldquo;Rarely\u0026rdquo; +\u0026rdquo; Never.\u0026rdquo;\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eQ5\u003c/strong\u003e\u003cp\u003eDo containers provide better performance than Virtual Machines?\u003c/p\u003e\u003c/p\u003e\u003cp\u003e-------------\u003c/p\u003e\u003cp\u003eFigure 5\u003c/p\u003e\u003cp\u003e-------------\u003c/p\u003e\u003cp\u003eIn this question, we can see there is a clear lean towards \u0026ldquo;Always\u0026rdquo; defining that containers do provide better performance than VMs.\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eQ6\u003c/strong\u003e\u003cp\u003eDo you think security measures add to the processing time of an application hosted on Virtual Machines irrespective of cloud deployment or on-premise deployment?\u003c/p\u003e\u003c/p\u003e\u003cp\u003e----------------------\u003c/p\u003e\u003cp\u003eFigure 6\u003c/p\u003e\u003cp\u003e\u003cb\u003e------------------\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThis question highlights that there is a major lean towards \u0026ldquo;Often.\u0026rdquo; From this the conclusion can be gathered that there is in fact an impact of security measures on performance of VMs. Since there is an impact there is a need to research more on the performance and security trade-offs, their impact, and how to mitigate any large trade-offs.\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eQ7\u003c/strong\u003e\u003cp\u003eDo you think security measures add to the processing time of an application hosted on Containers irrespective of cloud deployment or on-premise deployment?\u003c/p\u003e\u003c/p\u003e\u003cp\u003e---------------\u003c/p\u003e\u003cp\u003eFigure 7\u003c/p\u003e\u003cp\u003e---------------\u003c/p\u003e\u003cp\u003eThis question also helps solidify the fact that there is an impact of security on performance on containers and the need for further investigation on the topic.\u003c/p\u003e\u003cp\u003eThe below questions are asked to determine how security and performance impact the business requirements of the companies and enterprises.\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eQ8\u003c/strong\u003e\u003cp\u003eDoes system capacity planning impact customer satisfaction?\u003c/p\u003e\u003c/p\u003e\u003cp\u003e---------------\u003c/p\u003e\u003cp\u003eFigure 8\u003c/p\u003e\u003cp\u003e---------------\u003c/p\u003e\u003cp\u003eThe above question shows how the capacity planning of a system is impacting the customer satisfaction with about 66% of the responders agreeing to it being an import part of the architecture planning phase.\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eQ9\u003c/strong\u003e\u003cp\u003eShould system capacity planning account for the performance impact of security measures in system design?\u003c/p\u003e\u003c/p\u003e\u003cp\u003e-------------------\u003c/p\u003e\u003cp\u003eFigure 9\u003c/p\u003e\u003cp\u003e-------------------\u003c/p\u003e\u003cp\u003eThe above question shows how the impact of security may lower the performance of an application and how the system capacity planning should take into consideration the impact of security and how much extra resource may be needed to cope with it.\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eQ10\u003c/strong\u003e\u003cp\u003eIs it necessary to consider security measures when determining sizes of system resources?\u003c/p\u003e\u003c/p\u003e\u003cp\u003e-------------\u003c/p\u003e\u003cp\u003eFigure 10\u003c/p\u003e\u003cp\u003e\u003cb\u003e-------------\u003c/b\u003e\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eQ11\u003c/strong\u003e\u003cp\u003eIn case of a failure of a business-critical application, which one of the following will be considered a priority?\u003c/p\u003e\u003c/p\u003e\u003cp\u003e------------\u003c/p\u003e\u003cp\u003eFigure 11\u003c/p\u003e\u003cp\u003e\u003cb\u003e------------\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThis question determines how having the application run a bit slower is acceptable in the industry, but the security of the application should be intact. If this is the case, then the trade-off between performance and security becomes a major concern considering the business may lose customers in case the application is not up to suitable standards for the clients while taking care that the application security to remain intact.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e4.2. Deployment of the Application\u003c/h2\u003e\u003cp\u003eFigures 12 and 13 illustrate how the deployment of the application has been divided into two parts:\u003c/p\u003e\u003cp\u003ea. Deployment of application without security measures on VMs and Containers (Fig.\u0026nbsp;12)\u003c/p\u003e\u003cp\u003eb. Deployment of application with security measures on VMs and Containers (Fig.\u0026nbsp;13)\u003c/p\u003e\u003cp\u003e------------\u003c/p\u003e\u003cp\u003eFigure 12\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003e------------------\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003e-----------------\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003eFigure 13\u003c/p\u003e\u003cp\u003e-------------------\u003c/p\u003e\u003cp\u003eThe basic workflow and the configuration of the application is as below:\u003c/p\u003e\u003cdiv id=\"Sec6\" class=\"Section3\"\u003e\u003ch2\u003e4.2.1 The frontend\u003c/h2\u003e\u003cp\u003eThe front end of the application has been created using HTML, CSS, JavaScript and has been deployed on both EC2 machines as well as docker containers. The web server has been deployed using expressjs and nodejs.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec7\" class=\"Section3\"\u003e\u003ch2\u003e4.2.2 The Storage\u003c/h2\u003e\u003cp\u003eThe S3 bucket used is a general-purpose bucket deployed on AWS with bucket versioning disabled. Each user will have a folder created with their email IDs which will contain all the files uploaded by that user. This will help keep the files separate from other users and maintain security.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec8\" class=\"Section3\"\u003e\u003ch2\u003e4.2.3 The backend\u003c/h2\u003e\u003cp\u003eThe backend of the application has been deployed through Python Flask and Gunicorn for the server deployment. The APIs responsible for uploading the files to S3 bucket, and all the interactions with the DB are created in python.\u003c/p\u003e\u003cp\u003ePostgreSQL has been chosen as the database of choice due to its high compliance with SQL standards. The Network\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003e4.3. Application Workflow\u003c/h2\u003e\u003cp\u003eThe application workflow is depicted in Fig.\u0026nbsp;14.\u003c/p\u003e\u003cp\u003e-------------------\u003c/p\u003e\u003cp\u003eFigure 14\u003c/p\u003e\u003cp\u003e------------------\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003e4.4. Monitoring and Testing\u003c/h2\u003e\u003cp\u003eThe testing of the deployment will be conducted through a Jenkins pipeline with the help of the load testing tool - Locust. The Jenkins instance which will host the testing has also been deployed on an EC2 instance in the same Subnet as the application servers. The server monitoring has been done via Prometheus and Grafana.\u003c/p\u003e\u003c/div\u003e"},{"header":"5. Evaluation","content":"\u003cp\u003eThe evaluations of the deployment include the differences between the VM-based and container-based deployments.\u003c/p\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003e5.1. Scenario 1 \u0026ndash; Without the Security Measures\u003c/h2\u003e\u003cp\u003eIn this scenario, with a total of 50 users, a file of size 2.4MB has been used which will be uploaded to the application by the python API. The testing has been conducted for a time range of 150 seconds.\u003c/p\u003e\u003cp\u003eThe files will not be encrypted and will be directly pushed into the s3 bucket under the respective username folder. The PostgreSQL database is maintained in the same subnet group as the web application.\u003c/p\u003e\u003cdiv id=\"Sec13\" class=\"Section3\"\u003e\u003ch2\u003e5.1.1 Virtual Machine based Application\u003c/h2\u003e\u003cp\u003eThe application running on the Virtual Machine has been tested with the scenario 1 and is illustrated in Fig.\u0026nbsp;15.\u003c/p\u003e\u003cp\u003e-----------------\u003c/p\u003e\u003cp\u003eFigure\u0026nbsp;15\u003c/p\u003e\u003cp\u003e--------------------\u003c/p\u003e\u003cp\u003eFrom the above locust charts, it can be observed that the response times for the application peaked at about 9000-9500ms, as illustrated in Fig.\u0026nbsp;16.\u003c/p\u003e\u003cp\u003e-----------------\u003c/p\u003e\u003cp\u003eFigure 16\u003c/p\u003e\u003cp\u003e-----------------\u003c/p\u003e\u003cp\u003eFrom the above Grafana dashboard, it can be observed that the CPU usage peaked at about 40%, and the memory utilization was about 75% for the \u0026ldquo;cache\u0026thinsp;+\u0026thinsp;buffer\u0026rdquo; memory and about 50% for the total RAM used, with the average RAM being used at about 66% which would roughly be about 660MiB. The system was under a load percentage of 34%.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section3\"\u003e\u003ch2\u003e5.1.2 Container based Application\u003c/h2\u003e\u003cp\u003eThe application running on the docker container has been tested as per scenario 1.\u003c/p\u003e\u003cp\u003e-------------------\u003c/p\u003e\u003cp\u003eFigure 17\u003c/p\u003e\u003cp\u003e----------------------\u003c/p\u003e\u003cp\u003eFrom the above locust charts shown in Fig.\u0026nbsp;17, it can be observed that the response times for the application peaked at about 6000ms.\u003c/p\u003e\u003cp\u003e-----------------\u003c/p\u003e\u003cp\u003eFigure 18\u003c/p\u003e\u003cp\u003e--------------------\u003c/p\u003e\u003cp\u003eFrom the Grafana dashboard depicted in Fig.\u0026nbsp;18, it can be observed that the CPU utilization of the backend container, handling all the API requests, was about 6%. The memory utilization for the container peaked at 94MiB. The network usage for the containers peaked at 6Mb/s during the test runtime.\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003e5.2. Scenario 2 \u0026ndash; With the Security Measures\u003c/h2\u003e\u003cp\u003eIn this scenario, similar to the above, a total of 50 users, and a file of size 2.4MB has been used which will be uploaded to the application by the python API. The testing has been conducted for a time range of 150 seconds.\u003c/p\u003e\u003cdiv id=\"Sec16\" class=\"Section3\"\u003e\u003ch2\u003e5.2.1 Virtual Machine based Application\u003c/h2\u003e\u003cp\u003eThe application running on the Virtual Machine has been tested with scenario 2.\u003c/p\u003e\u003cp\u003e---------------------\u003c/p\u003e\u003cp\u003eFigure 19\u003c/p\u003e\u003cp\u003e------------------------\u003c/p\u003e\u003cp\u003eFrom locust charts shown in Fig.\u0026nbsp;19, it can be observed that the response times for the application peaked at about 10000ms.\u003c/p\u003e\u003cp\u003e--------------------\u003c/p\u003e\u003cp\u003eFigure 20\u003c/p\u003e\u003cp\u003e------------------\u003c/p\u003e\u003cp\u003eFrom the Grafana dashboard depicted in Fig.\u0026nbsp;20, it can be observed that the CPU usage again peaked at about 40%, and the memory utilization was about 90% for the \u0026ldquo;cache\u0026thinsp;+\u0026thinsp;buffer\u0026rdquo; memory and about 50% for the total RAM used, with the average RAM being used at about 66% which would roughly be about 660MiB. The system was under a load percentage of 25%.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec17\" class=\"Section3\"\u003e\u003ch2\u003e5.2.2 Container based Application\u003c/h2\u003e\u003cp\u003e-----------------\u003c/p\u003e\u003cp\u003eFigure 21\u003c/p\u003e\u003cp\u003e--------------------\u003c/p\u003e\u003cp\u003eFrom the locust charts shown in Fig.\u0026nbsp;21, it can be observed that the response times for the application peaked at about 15000ms.\u003c/p\u003e\u003cp\u003e------------------\u003c/p\u003e\u003cp\u003eFigure 22\u003c/p\u003e\u003cp\u003e--------------------\u003c/p\u003e\u003cp\u003eAs per the above Grafana dashboard depicted in Fig.\u0026nbsp;22, it can be observed that the CPU utilization of the backend container, handling all the API requests, was about 10%. The memory utilization for the container peaked at 128MiB. The network usage for the containers peaked at 8Mb/s during the test runtime.\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e"},{"header":"6. Analysis","content":"\u003cp\u003eThis analysis section will dive into the performance comparison of Virtual Machine based and container-based web application on cloud. The section will also cover how the security measures added to both Virtual Machine based and container-based web application impact the performance of both deployment scenarios[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. The analysis will focus on three separate scenarios CPU usage, Memory usage, and network usage. The insights gained here will help determine the impact of security on performance as well as how the Virtual Machine based, and container-based application perform in comparison to one another.\u003c/p\u003e\u003cp\u003eThe EC2 machines used to conduct the tests are equipped with AWS t2.micro instance type, with an OS Image of Amazon Linux, storage of 8GB EBS connected to the EC2 machine and 1vCPU.\u003c/p\u003e\u003cp\u003eThe results have been co-related to the two research questions of this paper and the results of the testing have been provided below for each of them.\u003c/p\u003e\u003cp\u003e\u003cb\u003eQ1. Do security measures impact the performance of VM-based and Container-based applications?\u003c/b\u003e\u003c/p\u003e\u003cp\u003eTo answer this question, the findings for the CPU, Memory, and Network usage will be compared for both scenarios where security measures applied and not applied as illustrated in Table \u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eComparison without security measures\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"3\"\u003e\u003c/colgroup\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eVirtual Machine\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eContainers\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eResponse Time\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e9000-9500ms\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e6000ms\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCPU Usage\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e40%\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e6%\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMemory Usage\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e660MiB\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e94MiB\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/table\u003e\u003c/div\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e-------------\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e\u003c/p\u003e\u003cp\u003e-----------------\u003c/p\u003e\u003cp\u003eFrom Table \u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, considering the scenario where security measures where not applied, the Response time for containers is less than that of the Virtual Machine. The CPU usage has a significant difference with that of the Virtual Machine being 40% and that of the container being only 6%. The memory usage also shows the containers being able to handle the application processing without taking on much load.\u003c/p\u003e\u003cp\u003eBecause the Virtual Machines CPU and Memory are responsible for the OS and any other background processes running, Table \u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e shows a higher percentage of resource utilization on the Virtual Machine side was expected.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eComparison with Security Measures\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"3\"\u003e\u003c/colgroup\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eVirtual Machine\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eContainers\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eResponse Time\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e10000ms\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e15000ms\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCPU Usage\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e40%\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e10%\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMemory Usage\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e723MiB\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e128MiB\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/table\u003e\u003c/div\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e-----------------\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e\u003c/p\u003e\u003cp\u003e----------------\u003c/p\u003e\u003cp\u003eAs per Table \u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, it can be seen that the container still performs better in the case of CPU and Memory usage. However, the response time of the containers is more than that of the Virtual Machine. The difference in response time in percentage can be calculated by using the formula\u003c/p\u003e\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:Percentage\\:Difference=\\:\\left(\\frac{\\left(x-y\\right)}{\\frac{x+y}{2}}\\right)\\times\\:100\\%$$\u003c/div\u003e\u003c/div\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eUsing the above formula the percentage difference of the response time comes out to be 40%. To put into perspective, there is a 40% increase in the response time of a container as compared to that of the Virtual Machine.\u003c/p\u003e\u003cp\u003eConsidering both the above scenarios, it can be concluded that the container out-performs the Virtual Machine in almost all the aspects. The response time in the case of applied security measures being the only case where the Virtual Machine performed better than the container.\u003c/p\u003e\u003cp\u003e\u003cb\u003eQ2. Is there a significant difference in performance of secured VMs and Containers?\u003c/b\u003e\u003c/p\u003e\u003cp\u003eTo understand the impact of security measures on both the deployment types, the comparison for unsecured VM to secured VM, and unsecured container to secured container must be evaluated.\u003c/p\u003e\u003cp\u003e--------------\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eComparison between Virtual Machines\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"3\"\u003e\u003c/colgroup\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eUn-secured VM\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSecured VM\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eResponse Time\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e9000-9500ms\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e10000ms\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCPU Usage\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e40%\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e40%\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMemory Usage\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e66%\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e72.30%\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/table\u003e\u003c/div\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e------------------\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e shows the difference in the response time, CPU usage, and memory usage of the two different VMs. The response time and the memory usage are higher in the secured VM. To calculate the percentage difference, the same formula can be used again.\u003c/p\u003e\u003cp\u003eFor the response time, the average value for the unsecured VM can be taken as 9250ms. Calculating the percentage difference between this and 10000ms will be roughly around 7.79%.\u003c/p\u003e\u003cp\u003eFor the Memory usage, the percentage difference comes out to be approximately 9.11%.\u003c/p\u003e\u003cp\u003eThe Memory usage being the same, no percentage difference has been calculated, and the percentage usage of 40% will be considered in the conclusion.\u003c/p\u003e\u003cp\u003eAlthough there is a bit of a difference in the two scenarios, with the secured VM utilizing more resources, the percentage difference shows values less than 10%. This shows that in the case of Virtual Machines, there is indeed an impact of security measures on the performance of a web application hosted on a Virtual Machines. But the difference is not too significant.\u003c/p\u003e\u003cp\u003eFor the next scenario, performance comparison between secured containers and non-secured containers are shown in Table \u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eComparison between Containers\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"3\"\u003e\u003c/colgroup\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eUn-secured Container\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSecured Container\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eResponse Time\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e6000ms\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e15000ms\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCPU Usage\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e6%\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e10%\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMemory Usage\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e94MiB\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e128MiB\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNetwork Usage\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e6Mb/s\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e8Mb/s\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/table\u003e\u003c/div\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e----------------------\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e\u003c/p\u003e\u003cp\u003e------------------------\u003c/p\u003e\u003cp\u003eCalculating the percentage difference between the values shown in Table \u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e using the same formula. Hene the obtained results are as follows:\u003c/p\u003e\u003cp\u003e• \u003cb\u003eResponse Time\u003c/b\u003e: approximately 85%\u003c/p\u003e\u003cp\u003e• \u003cb\u003eCPU Usage\u003c/b\u003e: approximately 50%\u003c/p\u003e\u003cp\u003e• \u003cb\u003eMemory Usage\u003c/b\u003e: approximately 30.63%\u003c/p\u003e\u003cp\u003e• \u003cb\u003eNetwork Usage\u003c/b\u003e: approximately 28.57%\u003c/p\u003e\u003cp\u003eThose values show a significant increase in the resource usage of the container when the security measures are applied. This result shows there is a major impact of security measures on performance of web applications hosted on containers.\u003c/p\u003e\u003cp\u003eOverall, the analysis can be concluded by stating the following:\u003c/p\u003e\u003cp\u003e1. There is a better performance outcome in terms of containerised application as compared to Virtual Machines if seen as a whole. The Virtual Machine performs better only in the case of response time for the application APIs.\u003c/p\u003e\u003cp\u003e2. There is a difference in the performance of the secured and unsecured Virtual Machine but the difference is somewhat manageable.\u003c/p\u003e\u003cp\u003e3. There is a significant performance and resource usage in terms of containers, with the container with security measures applied performing worse than the one where no security measures are applied.\u003c/p\u003e"},{"header":"7. Conclusions, limitations and future work","content":"\u003cp\u003eIn this work, the concept of cloud, virtualization, and containerization were discussed. The different deployment scenarios, literature review and the deployment process helped dive into the architectural and infrastructural setups of today’s web services in the cloud. This paper also discussed the significance of performance in today’s world and the impact of security on performance of web applications. A survey was conducted which targeted the working professionals in the IT industry. Their responses though being mixed do show that there is a known trade-off between security and performance in web applications. Through the implementation and testing, it can be observed that there is an impact of security on performance in both the cases of virtualization as well as containerization. Hence, proving the trade-off between the two to be true. The analysis done in this paper shows how much of a difference is between the two deployment scenarios.\u003c/p\u003e\u003cp\u003eThis project has been conducted on a small scale with a limited application with only a few APIs. The infrastructure chosen is limited in terms of EC2 machines though deployed on cloud. The project solely focuses on Python APIs running on Flask and Gunicorn with basic HTML, CSS, and JavaScript running on the frontend. To further enhance this project and dive deeper into the security and performance trade-off, this test can be conducted on a larger scale with a number of EC2 machines and containers load balanced to provide a more real-world scenario. The complexity of the application can be increased to further add pressure on the machines’ resources forcing it to auto scale and understand the costing of this trade-off as well.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eAcknowledgements (Competing Interests):\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eNo funds, grants, or other support was received.\u003c/p\u003e\n\u003cp\u003eThe authors have no financial or proprietary interests in any material discussed in this article.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatement on clinical trial\u003c/strong\u003e: Not Applicable.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eConsent to Publish:\u0026nbsp;\u003c/strong\u003eWe give our full consent for Springer Nature to publish the manuscript in\u0026nbsp;Discover Computing, should it be accepted after peer review.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eEthics and Consent to Participate declarations\u003c/strong\u003e: not applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics statement\u003c/strong\u003e\u003cstrong\u003e:\u0026nbsp;\u003c/strong\u003eNot Applicable.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eData availability:\u0026nbsp;\u003c/strong\u003eData available upon request.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contribution statements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA.K. conceived the presented idea and developed the theory and performed the computations.\u003c/p\u003e\n\u003cp\u003eZ.E. verified the design and analytical methods as well as supervised the findings of this work. reviewed the entire work and selected the sections that needed to be focused on for publication purposes. Both authors discussed the results and contributed to the final manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eTanwar J, Sharma SK, Mittal M. Generation of Quality Web Services: A survey 10th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO). 2022 https://doi.org/10.1109/icrito56286.2022.9964560. Accessed 27 Apr 2025\u003c/li\u003e\n\u003cli\u003eShukla A, Katt B, Yamin MM. A quantitative framework for security assurance evaluation and selection of cloud services: a case study. Int. J. Inf. Secur\u003cem\u003e.\u003c/em\u003e 2023; 22: 1621\u0026ndash;1650. https://doi.org/10.1007/s10207-023-00709-8 Accessed 19 May 2025\u003c/li\u003e\n\u003cli\u003eVMWare. What is a Virtual Machine? Available at https://www.vmware.com/topics/glossary/content/virtual-machine.html/ Accessed 26 May 2024\u003c/li\u003e\n\u003cli\u003eStergiopoulos G, Dedousis P, Gritzalis D. Automatic analysis of attack graphs for risk mitigation and prioritization on large-scale and complex networks in Industry 4.0. \u003cem\u003eInt. J. Inf. Secur.\u003c/em\u003e 2022; 21:37\u0026ndash;59. https://doi.org/10.1007/s10207-020-00533-4\u003c/li\u003e\n\u003cli\u003eJaiswal K. Dockerize A Spring Boot Application - Sixt Research \u0026amp; Development India - Medium. 2020 Available at https://medium.com/the-sixt-india-blog/dockerise-a-spring-boot-application-bdfce1d6eb15 Accessed 27 May 2025\u003c/li\u003e\n\u003cli\u003eAntunes P, Guimar\u0026atilde;es N. Guiding the implementation of data privacy with microservices. Int. J. Inf. Secur\u003cem\u003e.\u003c/em\u003e 2024;23:3591\u0026ndash;3608. https://doi.org/10.1007/s10207-024-00907-y \u003c/li\u003e\n\u003cli\u003eKropp A, Torre R. Docker: containerize your application. In: 2020 Elsevier eBooks; 2020. p. 231\u0026ndash;244 https://doi.org/10.1016/b978-0-12-820488-7.00026-8\u003c/li\u003e\n\u003cli\u003eYadav AK, Garg ML, Ritika. Docker Containers Versus Virtual Machine-Based Virtualization. Advances in Intelligent Systems and Computing, pp.141\u0026ndash;150 (2018) http://dx.doi.org/10.1007/978-981-13-1501-5_12. Accessed 16 May 2025 Docker containers versus virtual machine-based virtualization. In: Abraham A, Dutta P, Mandal J, Bhattacharya A, Dutta S, editors. Emerging technologies in data mining and information security. Advances in intelligent systems and computing. Singapore: Springer; 2019. 814:5-12.\u003c/li\u003e\n\u003cli\u003eTsifountidis F. Virtualization Security: Virtual Machine Monitoring and Introspection. [dissertation]: University of London; 2011. Available at https://cdn.ttgtmedia.com/rms/pdf/RHUL_Tsifountidis_Final.pdf Accessed 15 Feb 2025\u003c/li\u003e\n\u003cli\u003ePatil R, Dudeja H, Modi C. Designing in-VM-assisted lightweight agent-based malware detection framework for securing virtual machines in cloud computing. \u003cem\u003eInt. J. Inf. Secur.\u003c/em\u003e 2020;19:147\u0026ndash;162.\u003c/li\u003e\n\u003cli\u003eAlyas T, Ali S, Khan HU, Samad A, Alissa K, Saleem MA. Container performance and vulnerability management for container security using Docker engine. Shafiq M, editor. Security and communication networks; 2022. p.1\u0026ndash;11. Available at http://dx.doi.org/10.1155/2022/6819002 Accessed 25 Feb 2025\u003c/li\u003e\n\u003cli\u003eBabatunde DJ. Evaluating the Impact of Security Measures on Performance of secure Web Applications hosted on Virtualized Platforms. Available at https://repository.uel.ac.uk/download/860cd9c0802ea8ba5d31f5df9\n1e3bdbfb919eee114bac2e69ee2a03f9cb7612f/20753811/John%2520Babatunde.pdf 2015. Accessed 15 April 2025\u003c/li\u003e\n\u003cli\u003eWalid R, Joshi KP, Choi SG. Comparison of attribute-based encryption schemes in securing healthcare systems. Sci Rep. 2024;14:7147. Available at https://doi.org/10.1038/s41598-024-57692-w Accessed 15 March 2025\u003c/li\u003e\n\u003cli\u003eFang Z, Ma H, Chen G, Hartmann S. A group genetic algorithm for energy-efficient resource allocation in container-based clouds with heterogeneous physical machines. In: Liu T, Webb G, Yue L, Wang D., editors. Advances in Artificial Intelligence. AI 2023. Lecture Notes in Computer Science. Singapore: Springer; 2023. p.14472. https://doi.org/10.1007/978-981-99-8391-9_36 Accessed 15 Feb 2025\u003c/li\u003e\n\u003cli\u003eFelter W, Ferreira A, Rajamony R, Rubio J. An updated performance comparison of virtual machines and Linux containers. IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS); 2015. p.171-172. https://ieeexplore.ieee.org/document/7095802 Accessed 30 April 2025\u003c/li\u003e\n\u003cli\u003eSultan S, Ahmad I, Dimitriou T. Container security: Issues, challenges, and the road ahead. IEEE Access. 2019;7:52976\u0026ndash;52996 https://doi.org/10.1109/ACCESS.2019.2911732 \u003c/li\u003e\n\u003cli\u003eMalhotra R, Bansal A, Kessentini M. Deployment and performance monitoring of docker based federated learning framework for software defect prediction. Cluster Comput. 2024;27:6039\u0026ndash;6057. https://doi.org/10.1007/s10586-024-04266-0 \u003c/li\u003e\n\u003cli\u003eMartin A, Raponi S, Combe T, Di Pietro R. Docker ecosystem \u0026ndash; Vulnerability Analysis. Computer Communications. 2018;122:30\u0026ndash;43.\u003c/li\u003e\n\u003cli\u003eSu JM. WebHOLE: Developing a web-based hands-on learning environment to assist beginners in learning web application security. Educ Inf Technol. 2024; 29:6579\u0026ndash;6610.\u003c/li\u003e\n\u003cli\u003eIwendi DC, Kryvinska DN, Awan I-U, editors. Recent Advancements in Cloud Security using Performance Technologies and Techniques; 2022. SpringerOpen; 2022.Available at https://www.springeropen.com/collections/raicsuptat Accessed 20 May 2025\u003c/li\u003e\n\u003cli\u003eRandal A. The ideal versus the real. ACM Computing Surveys. 2020;53(1):1\u0026ndash;31 https://doi.org/10.1145/3365199.\u003c/li\u003e\n\u003cli\u003eManco F et al. My VM is kighter (and safer) than your container. SOSP 2017. Proceedings of the 26th ACM Symposium on Operating Systems Principles; 2017 October 28\u0026ndash;31; Shanghai, China. ACM Press; 2017. p.218\u0026ndash;233. \u003c/li\u003e\n\u003cli\u003eLi Z, Kihl M, Lu Q, Andersson JA. Performance overhead comparison between hypervisor and container based virtualization. Proceedings of the IEEE 31st International Conference on Advanced Information Networking and Applications; 2017 March 27-29; Taipei, Taiwan, IEEE; 2017. p. 955-962. http://dx.doi.org/10.1109/aina.2017.79. \u003c/li\u003e\n\u003cli\u003eKaur P, Josan JK, Neeru N. Performance analysis of Docker containerization and virtualization. In: Bindhu V, Tavares JMRS, Du KL, editors. Proceedings of Third International Conference on Communication, Computing and Electronics Systems; 2022. Lecture Notes in Electrical Engineering; Singapore: Springer; 2022. 844; p.863\u0026ndash;877 http://dx.doi.org/10.1007/978-981-16-8862-1_56. \u003c/li\u003e\n\u003cli\u003eAbuabdo A, Al-Sharif ZA. Virtualization vs. Containerization: Towards a multithreaded performance evaluation approach. AICCSA 2019. Proceedings of the IEEE/ACS 16th International Conference on Computer Systems and Applications. 2019 Nov. 3-7; Abu Dhabi, United Arab Emirates, p. 1-6. http://dx.doi.org/10.1109/aiccsa47632.2019.9035233\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Cloud computing, Virtual Machine, Containers, Web Applications, Performance, Security","lastPublishedDoi":"10.21203/rs.3.rs-7368391/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7368391/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis paper explores the trade-off between security and performance of web applications hosted on Virtual Machine based environments and container-based environments. This paper also investigates the comparison between container and Virtual Machine performance on several aspects. With the increasing adoption of Virtual Machines and container technologies for web applications, understanding the significance of security measures becomes crucial. A series of tests were conducted on Virtual Machine based and container-based web applications, implementing encryption, and network isolation.\u003c/p\u003e\u003cp\u003eThis paper measures the key performance indicators, such as response time, CPU utilization, Memory utilization, and network utilization. The results indicate a 40% increase in the response time of a container than a Virtual Machine in the case where security measures have been applied to the environment. In the analysis of secured and unsecured VMs, all the metrics show a difference of less than 10%. However, in the case of containers, there is a significant difference in the performance ranging from 28\u0026ndash;85% across the board.\u003c/p\u003e\u003cp\u003eThese findings highlight the need for optimized security strategies that mitigate performance degradation. Future works will focus on exploring more advanced cloud setups and additional security measures. This paper provides foundational analysis to balance security and performance in the design and deployment of modern web applications.\u003c/p\u003e","manuscriptTitle":"Evaluation of Impact of Security Measures on Container-based and VM-based Application Performance in Cloud","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-23 02:35:36","doi":"10.21203/rs.3.rs-7368391/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"ccb93c3e-8018-492a-8cb6-7661dfed119b","owner":[],"postedDate":"September 23rd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-11-17T05:23:18+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-23 02:35:36","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7368391","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7368391","identity":"rs-7368391","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}