Evaluating The Impact Of Cloud-Based Microservices Architecture

On Application Performance
Ganesh Chowdary Desina
California State University Fresno
Computer Science Dept.
Fresno, California

optimization strategies for enhancing performance in

ABSTRACT: cloud-primarily based microservices environments.
Monitoring and troubleshooting performance issues in
The study assess the impact of cloud-based a dispensed microservices environment poses precise
microservices architectures on application performance. demanding situations.
Several aspects of performance evaluation are discussed,
including response time, throughput, scalability, and It discusses the function of monitoring gear,
reliability. This article examines the advantages and logging, and metrics in identifying performance
challenges of adopting a cloud-based approach. It bottlenecks and facilitating powerful troubleshooting.
explores potential bottlenecks and issues in a In end, this paper aims to provide a complete evaluation
microservices architecture and presents optimization of the effect of cloud-based totally microservices
techniques. architecture on application overall performance. By
reading the advantages, demanding situations,
With the help of case studies and empirical performance assessment strategies, and optimization
studies, it compares cloud-based microservices techniques, it assists organizations in making informed
architectures with traditional monolithic architectures. selections regarding their architectural picks and overall
Furthermore, the paper examines the challenges of performance optimization strategies.
monitoring and troubleshooting distributed
microservices. In conclusion, it emphasizes the Background Knowledge:
importance of planning, designing, and testing during
the adoption of cloud-based microservices. Cloud-based microservices architecture is a
software development approach that emerged from the
convergence of service-oriented architecture, cloud
Introduction: computing, and containerization technologies. It offers
a way to design applications as a collection of small,
In this paper, we observe the performance loosely coupled services running in a cloud
effect of cloud-based microservices architectures. environment. The origins of this technology can be
Microservices' advantages must be evaluated if you traced back to the early 2000s with the advent of
want to ensure most advantageous overall performance service-oriented architecture and the subsequent
and user pleasure, and this evaluation is important. The advancements in cloud computing and containerization.
paper gives an overview of the microservices
architecture and its key traits. These benefits make The rise of DevOps practices and the popularity
microservices structure a popular desire for companies of continuous deployment further emphasized the need
in search of agility and scalability in their programs. for more lightweight and agile architectures. Industry
pioneers like Netflix played a significant role in
Next, the paper goes through the assessment of popularizing microservices architecture by developing
performance in a cloud-based totally microservices and open-sourcing tools and frameworks. The
architecture. It discusses performance metrics along community actively contributed to the development of
with reaction time, throughput, scalability, and best practices and patterns for building and deploying
reliability. Furthermore, the paper addresses microservices-based applications in the cloud. Overall,
 cloud-based microservices architecture represents an evaluation techniques, findings, and conclusions.
evolutionary approach to address scalability, flexibility, Taken notes and organize the extracted information for
and continuous deployment challenges faced by further analysis.
traditional monolithic applications in a cloud
computing context.
Related Work:
Several important research that has advanced
Method of Gathering Literature our understanding of performance evaluation in
microservices architectures are included in "Evaluating
I used the terms "microservices," "cloud the Impact of Cloud-based Microservices Architecture
computing," and "performance evaluating" in my on Application Performance." In their systematic
search query to find literature on the issue of mapping study, Pezzè and Spina identified research
"evaluating the impact of cloud-based microservices gaps and gave an overview of performance analysis
architecture on application performance." I searched methods. In their case study, Barbosa et al. provided
for studies that highlighted the usage of microservices insights on performance and scalability by analyzing
technology to assess website usability. As suggested, the operation of a real-world microservices-based
I mostly searched the IEEE Xplore and ACM Digital application. Martini et al. proposed a framework for
Library databases. performance testing, addressing the challenges specific
to microservices.
I clearly articulated the specific aspects of cloud-based Performance benchmarking techniques were examined
microservices architecture and application by Charfi et al., who identified methodologies and best
performance that you want to investigate. This will practices. In their evaluation of the performance of
help you narrow down your search and focus on microservices architectures in the cloud, Macedo et al.
relevant literature. focused on scalability, throughput, and response time.
Collectively, these studies provide the framework to
Utilized academic databases: Access reputable evaluate the impact of cloud-based microservices
academic databases such as IEEE Xplore, ACM architecture on application performance.
Digital Library, Google Scholar, or specific computer
science repositories. Use relevant keywords and
combinations, such as "cloud-based microservices,"
"application performance," "evaluation," and A) Performance Analysis of
"impact," to conduct comprehensive searches.
microservices:
Reviewed relevant conferences and journals: Identify
The papers discussed revolve around the
conferences, symposiums, and journals that regularly
concept of microservices architecture and share a
publish research related to cloud computing,
common focus on evaluating and analyzing its impact
microservices, and performance evaluation. Examples
on application development and performance.
include the International Conference on Cloud
Microservices architecture involves breaking down
Computing (ICCC), the International Conference on
complex applications into smaller, independent
Microservices (Microservices), and the IEEE
services, emphasizing modularity, scalability, and fault
Transactions on Cloud Computing.
isolation.
Explored related work in cited references attention to
A key aspect among these papers is the evaluation and
the references cited in the papers you find, as they can
analysis of microservices architecture. Researchers
lead you to additional relevant sources that may not
employ various methodologies and techniques to gather
have appeared in your initial search.
empirical data, metrics, and insights to support their
evaluations. The papers aim to assess the impact of
Read full papers and extract key information about the
microservices architecture on different aspects,
full papers of the selected publications and thoroughly
including performance, quality assurance, and
read them. Extract relevant information, including
architectural comparison. They delve into performance
research methodologies, performance metrics,
metrics, scalability patterns, resource consumption, and
testing strategies to understand the implications of One recurring theme in this paper 5 is the
microservices on application performance. evaluation of performance. Researchers analyze and
assess performance metrics such as response time,
Application performance is a shared focus across these throughput, latency, scalability, and resource
papers. Researchers aim to understand and improve the utilization. By studying these metrics, they gain insights
performance of applications built using microservices into how microservices architecture influences
architecture. They explore various performance application performance and identify areas for
metrics, investigate scalability patterns, analyze improvement.
resource utilization, and propose testing strategies to
assess the performance implications of microservices Scalability and elasticity are key benefits highlighted in
architecture. By gaining insights into application the paper 4 They emphasize the dynamic scaling
performance, researchers aim to optimize the design capabilities of cloud-based microservices, allowing
and implementation of microservices-based systems. them to handle varying workload demands effectively.
Techniques and approaches for achieving scalability
Challenges and perspectives associated with and elasticity are explored, contributing to the
microservices architecture are acknowledged in these understanding of how microservices architecture can
papers. They recognize the complexities arising from adapt to changing resource needs.
service dependencies, communication overhead, and
coordination in microservices-based systems. The resilience and fault tolerance aspects of cloud-
Researchers provide perspectives on addressing these based microservices are also addressed. Researchers
challenges and offer insights into best practices and discuss strategies for handling failures, implementing
potential solutions. By identifying and addressing the fault tolerance mechanisms, and ensuring high
challenges, they aim to facilitate the successful adoption availability in microservices-based architectures. By
and optimization of microservices architecture. exploring these topics, they contribute to the
development of robust and reliable systems that can
The practical relevance of these papers is another withstand failures and maintain service continuity.
common factor. They aim to bridge the gap between
research and practice by presenting real-world case Containerization technologies, such as Docker, and
studies, industry perspectives, and practical container orchestration platforms like Kubernetes,
implications. By providing valuable insights and receive significant attention in several paper 5 and paper
lessons learned, these papers guide organizations in 6. They explore the benefits and challenges associated
adopting and optimizing microservices architecture. with using containers and orchestration in deploying
The researchers aim to provide practical guidance and and managing microservices in the cloud. These
inform decision-making processes for organizations discussions shed light on the practical aspects of
venturing into microservices-based development. containerization and orchestration and their impact on
microservices architecture.
In summary, these papers collectively contribute to the
body of knowledge on microservices architecture by Another common thread in the papers is the inclusion
evaluating its impact on application development and of real-world case studies and empirical evaluations.
performance. They emphasize the importance of These studies provide practical insights into the
modularity, scalability, and fault isolation in implementation and performance of cloud-based
microservices architecture. Through empirical analysis, microservices in various domains, such as healthcare, e-
researchers gain insights into application performance, commerce, and IoT. By presenting real-world
address challenges, and provide practical guidance for scenarios, researchers bridge the gap between theory
organizations. The findings and perspectives presented and practice, offering valuable lessons and experiences
in these papers support informed decision-making and that can inform future deployments.
optimization of microservices-based systems.
The challenges associated with cloud-based
microservices, such as service discovery,
communication, monitoring, and security, are also
B) Uses of Cloud based microservices: addressed in the papers. Researchers propose solutions
and best practices to tackle these challenges and
improve the overall efficiency and reliability of comparisons with performance benchmarks or industry
microservices-based systems. By sharing their findings, standards, it gives measurable information on how the
they contribute to a better understanding of the microservices design manages a range of workloads.
complexities and potential solutions in deploying and
managing microservices architecture. Experimental Evaluation: This paper uses
experimental evaluation to assess the performance of
In summary, the papers 4, paper 5, paper 6, paper 7 microservices architectures in cloud environments.
collectively contribute to the body of knowledge They design and implement testbeds or utilize existing
surrounding cloud-based microservices architecture. platforms to conduct experiments with representative
They offer insights, evaluations, and recommendations workloads. Performance metrics such as response time,
for practitioners and researchers interested in adopting throughput, scalability, and resource utilization are
and optimizing microservices in the cloud computing measured and analyzed.
environment. By exploring various aspects such as
performance, scalability, fault tolerance, and resilience,
these papers help advance the understanding and
Paper 2:
implementation of microservices architecture in real-
Profiling: Profiling methods are used to
world scenarios.
investigate how microservices-based applications
perform during runtime. Resource usage patterns,
Metrics used performance bottlenecks, and areas for optimization can
all be found using profiling tools and methodologies.
Researchers learn more about the performance
Paper 1: characteristics of distinct microservices and how they
interact by examining measures like CPU usage,
Combining case studies with performance
memory consumption, and network latency. By using
testing and profiling research approaches could produce
profiling, performance-critical components can be
useful findings when analyzing how cloud-based
optimized, and the design can be fine-tuned.
microservices architecture impacts application
performance. Integrating these two approaches enables
a thorough evaluation of performance analysis and real- Concurrent users: Users concurrently
world implementations. connected to the machine are said to be concurrent
users. It determines the machine's capacity to handle
Case Studies: Case studies analyze how actual several requests at once and evaluates the machine's
cloud-based microservices architecture scalability and overall performance in low-load
implementations impact the performance of scenarios.
applications. Researchers can collect information on
performance before and after the implementation by
working with companies that are currently using Researchers may achieve an in-depth
microservices. Specific industries, applications, or use evaluation of the effect of cloud-based microservices
cases may be the focus of the case studies. They provide architecture on application performance by integrating
thorough insights into the real-world effects of case studies with performance testing and profiling
microservices architecture on performance, including approaches. While performance testing and profiling
scalability, improvements in performance, and provide quantitative and qualitative data on indicators
transitional difficulties. of performance and optimization opportunities, case
studies offer real-world context and useful insights. The
Performance testing: For evaluating the performance benefits, challenges, and potential improvements of the
of microservices-based applications, performance microservices architecture in terms of application
testing involves modeling realistic workloads. To performance are better understood according to this
evaluate how the application operates under various integrated approach.
loads and stresses, load testing, stress testing, and
endurance testing can be done. Parameters like reaction
time, throughput, and scalability are measured with the
use of performance testing. In addition to allowing Paper 3:
 To evaluate an application's or device's general
Results:
performance characteristics, performance testing
Dataset for Request 1:
requires evaluating a variety of metrics. These metrics
provide quantifiable information on response times,
Microservice Microservice B Microservice C
throughput, effective resource use, and other
A
performance-related factors. Some of the metrics used
Response Response time = Response time =
in performance testing are Response Time, Throughput,
time = {80,89,72,85} {120,122,118,1
Error rate.
{50,55,48,52} 24}
Response Time: In this paper Response time is used to CPU CPU CPU
measure how quickly a device reacts to user input or Utilization= Utilization= Utilization=
transactions. It comprises the time spent processing, {30,32,29,31} {28,26,30,27} {31,33,30,32}
communicating through networks, and accessing Memory Memory Memory
databases. An important information that directly Utilization= Utilization= Utilization=
affects the user experience and device performance is {100,102,98,9 {150,148,152,1 {180,178,182,1
response time. 5} 55} 85}

Throughput: The number of transactions, requests, or

activities that a system can handle in a specific amount Microservice A: Microservice B: Microservice C:
of time is known as throughput. It implies the Average Average Average
effectiveness and processing power of the device. Due Response time: Response time: Response time:
to the machine's ability to handle a greater workload, 50 ms 80 ms 120 ms
higher throughput values imply higher overall Average CPU Average CPU Average CPU
performance. utilization: 20% utilization: 30% utilization: 35%
Average Memory Average Average
Error Rate: During performance auditing, the utilization: 100 Memory Memory
percentage of unsuccessful or inaccurate transactions is MB utilization: 150 utilization: 180
measured by the error rate. It aids in determining the MB MB
device's stability, error-handling capacity, and
toughness under pressure. A lower error rate indicates
improved efficiency and dependability. Dataset for Request 2:

Microservice A Microservice B Microservice C

Evidences:
Response time = Response time = Response time =
{70, 75, 68, 72} {60, 58, 63, 62} {110,108,113,
Groupings References 115}
Performance analysis of [1] [2] [3] CPU CPU CPU
Micro services Utilization= Utilization= Utilization=
{30,32,29,31} {28,26,30,27} {31,33,30,32}
Uses of cloud based [4] [5] [6] [7]
Memory Memory Memory
micro services
Utilization= Utilization= Utilization=
Metrics [8] [9] [10] {120, 118, 122, {130, 132, 128, {160, 162, 158,
125} 135} 165}

Microservice A: Microservice Microservice

Average B: C:
Response time: Average Average
60 ms Response time: Response time:
70 ms 110 ms
 Average CPU Average CPU Average CPU Key Takeways:
utilization: 25% utilization: 28% utilization: 32%
Average Average Average A) Performance Metrics
Memory Memory Memory In paper A layered framework for root cause
utilization: 120 utilization: 130 utilization: 160 diagnosis of microservices A variety of performance
MB MB MB metrics are used by various articles to assess how the
microservices architecture affects application
performance. These metrics include response time,
throughput, scalability, resource utilization, and
latency. Response time measures the amount of time it
takes to respond to a request. Throughput measures how
Dataset: many requests are processed per unit of time.
Researchers can evaluate the effectiveness, scalability,
Micro Micro Micro and responsiveness of microservices systems by
assessing these parameters.
service service service
A B C B) Scalability and Elasticity
Response
Time 50 89 120 The papers Efficiency analysis of provisioning
CPU of micro services: When compared to conventional
Utilization 30 28 33 monolithic systems, microservices design allows better
scalability and elasticity. By breaking down an
Memory application into smaller services, it is possible for each
Utilization 102 148 180 service to be independently scaled based on demand.
Organizations may distribute resources more
effectively and manage shifting workloads more
skillfully because to this flexibility. The usage of load
Chart Title balancing, auto scaling, and horizontal scaling, among
400
other scalability patterns, are discussed in papers as
ways to obtain the best performance possible in
microservices systems.
200

C) Dependencies:
0
In the paper “Performance evaluation of
Micro service A Micro service B Micro service C
microservices architectures” using containers Service
Response Time CPU Utilization The interactions between microservices can have a big
Memory Utilization
impact on how well an application performs. To prevent
performance bottlenecks and latency problems, papers
stress the significance of knowing and controlling
The graph is for the dataset 1 and the first elements in service dependencies. Asynchronous communication,
the list. event-driven architectures, and distributed caching are
just a few of the strategies being investigated to lessen
the effects of service dependencies and boost overall
application performance.
 Conclusions: Spain, 2020, pp. 1-6, doi:
10.23919/CISTI49556.2020.9141168.
The papers provided focus on evaluating the
[4] S. Haselböck, R. Weinreich, G. Buchgeher, Deci-
impact of cloud-based microservices architecture on
sion guidance models for microservices: service
application performance. They highlight the benefits
discovery and fault tolerance, in: Proceedings of the
and challenges of adopting microservices, such as
Fifth European Conference on the Engineer- ing of
improved modularity, scalability, and fault isolation.
Computer-Based Systems, 2017, pp. 1–10.
The evaluation of performance metrics, including
response time, throughput, and resource utilization,
[5] H. Khazaei, C. Barna, N. Beigi-Mohammadi and M.
helps understand the impact of microservices on
Litoiu, "Efficiency Analysis of Provisioning
application performance.
Microservices," 2016 IEEE International Conference
on Cloud Computing Technology and Science
Scalability and elasticity are key advantages of
(CloudCom), Luxembourg, Luxembourg, 2016, pp.
microservices architecture, enabling dynamic scaling to
261-268, doi: 10.1109/CloudCom.2016.0051.
handle varying workload demands. Containerization
technologies like Docker and orchestration platforms
like Kubernetes are explored for deploying and
[6] M. Villamizar et al., "Evaluating the monolithic and
managing microservices in the cloud. The papers also
the microservice architecture pattern to deploy web
address fault tolerance and resilience, proposing
applications in the cloud," 2015 10th Computing
mechanisms to handle failures and ensure high
Colombian Conference (10CCC), Bogota, Colombia,
availability. Real-world case studies demonstrate the
2015, pp. 583-590, doi:
practical implementation and performance of
10.1109/ColumbianCC.2015.7333476.
microservices in domains like healthcare and IoT.
Challenges related to service discovery,
[7] A. S. Kondratyev, K. A. Aksyonov, N. A. Buravova
communication, monitoring, and security are
and O. P. Aksyonova, "Cloud-Based Microservices to
acknowledged, with proposed solutions and best
Decision Support," 2018 Tenth International
practices. Overall, the papers contribute valuable
Conference on Ubiquitous and Future Networks
insights, evaluations, and recommendations for
(ICUFN), Prague, Czech Republic, 2018, pp. 389-394,
practitioners and researchers interested in adopting and
doi: 10.1109/ICUFN.2018.8437015.
optimizing cloud-based microservices architecture.
[8] C. Esposito, A. Castiglione, C. -A. Tudorica and F.
References Pop, "Security and privacy for cloud-based data
management in the health network service chain: a
[1] Luo S, Xu H, Ye K, Xu G, Zhang L, He J, Yang microservice approach," in IEEE Communications
G and Xu C. Erms: Efficient Resource Management for Magazine, vol. 55, no. 9, pp. 102-108, Sept. 2017, doi:
Shared Microservices with SLA Guarantees. 10.1109/MCOM.2017.1700089.
Proceedings of the 28th ACM International Conference
on Architectural Support for Programming Languages [9] L. Abdollahi Vayghan, M. A. Saied, M. Toeroe and
and Operating Systems, Volume 1. (62-77) F. Khendek, "Microservice Based Architecture:
Towards High-Availability for Stateful Applications
[2] M. Amaral, J. Polo, D. Carrera, I. Mohomed, M. with Kubernetes," 2019 IEEE 19th International
Unuvar and M. Steinder, "Performance Evaluation of Conference on Software Quality, Reliability and
Microservices Architectures Using Containers," 2015 Security (QRS), Sofia, Bulgaria, 2019, pp. 176-185, doi:
IEEE 14th International Symposium on Network 10.1109/QRS.2019.00034.
Computing and Applications, Cambridge, MA, USA,
2015, pp. 27-34, doi: 10.1109/NCA.2015.49. [10] M. Ma, W. Lin, D. Pan and P. Wang, "Self-
Adaptive Root Cause Diagnosis for Large-Scale
[3] K. Castro et al., "Performance Evaluation of the Microservice Architecture," in IEEE Transactions on
Virtualization Environment of a Microservices-Based Services Computing, vol. 15, no. 3, pp. 1399-1410, 1
Payroll System," 2020 15th Iberian Conference on May-June 2022, doi: 10.1109/TSC.2020.2993251.
Information Systems and Technologies (CISTI), Seville,
[11] A. Bento et al., "A layered framework for root
cause diagnosis of microservices," 2021 IEEE 20th
International Symposium on Network Computing and
Applications (NCA), Boston, MA, USA, 2021, pp. 1-8,
doi: 10.1109/NCA53618.2021.9685494.

[12] T. Porrmann, R. Essmann and A. W. Colombo,

"Development of an event-oriented, cloud-based
SCADA system using a microservice architecture under
the RAMI4.0 specification: Lessons learned," IECON
2017 - 43rd Annual Conference of the IEEE Industrial
Electronics Society, Beijing, China, 2017, pp. 3441-
3448, doi: 10.1109/IECON.2017.8216583.

[13] Brunnert et al., A. (2015). Performance-oriented

devops: A research agenda. Technical Report: SPEC-
RG-2015- 01, SPEC RG DevOps Performance Working
Group.

[14] N. V. D. Shahir Daya et al., "Microservices from

Theory to Practice: Creating Applications in IBM
Bluemix Using the Microservices Approach" in , IBM
Redbooks, pp. 6-10, 2015.
[15] S. Newman, Building Microservices, Sebastopol,
California, USA:O'Reilly Media, Inc., 2015.