Performance Modeling for Cloud Microservice Applications

Jindal, Anshul; Podolskiy, Vladimir; Gerndt, Michael

doi:10.1145/3297663.3310309

Cited by 83 publications

(29 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From the selection of works highlighted in Table 1, some approaches are based on service requirements, so the service can be deployed in nodes with specific hardware characteristics [7], [8], [31]. Other proposals use profiling to determine when to scale services, after identifying workloads and saturation points [9], [32].…”

Section: Related Workmentioning

confidence: 99%

Service Placement for Latency Reduction in the Fog Using Application Profiles

et al. 2021

View full text Add to dashboard Cite

The Cloud-Fog-Internet of Things continuum combines different paradigms to provide connectivity and ubiquity for end-users, while also granting low latency and low jitter to cope with different challenges, including the requirements of latency-sensitive applications, such as virtual/augmented reality and online gaming. This constitutes a complex and dynamic environment with heterogeneous resources that need to be managed or orchestrated, in order to accomplish application requirements for low latency. Common orchestration solutions make placement decisions based only on the resources of the underlying network and the application resource requests; however, using the profiles of applications to make placement decisions has the potential to enhance the final performance perceived by the end-users. This paper proposes the use of application profiles according to their popularity to guide their placement. To corroborate the effectiveness of the use of the profiles, two placement mechanisms are presented, one based on Genetic Algorithms and the other inspired on graph partitions. Simulation results show that it is possible to reduce the latency and jitter of applications via a service placement guided by the profiles. The mechanism based on graph partitions showed better results for all scenarios, followed closely by the Genetic Algorithm in the scenarios with lower load.

show abstract

Section: Related Workmentioning

confidence: 99%

Service Placement for Latency Reduction in the Fog Using Application Profiles

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Measurement of performance metrics under typical or stress test execution conditions, which involve both workload and platform configuration aspects (Menasc'e 2002;Hill, Schmidt, Edmondson & Gokhale 2009;Apte et al 2017;Michael et al 2017;Jindal et al 2019), detection of performance-related issues such as functional problems or violations of performance requirements emerging under certain workload or resource configuration conditions (Briand et al 2005;Zhang et al 2011;Ayala-Rivera et al 2018;Schulz et al 2019) are common objectives of different types of performance testing.…”

Section: Related Workmentioning

confidence: 99%

An autonomous performance testing framework using self-adaptive fuzzy reinforcement learning

et al. 2021

View full text Add to dashboard Cite

Test automation brings the potential to reduce costs and human effort, but several aspects of software testing remain challenging to automate. One such example is automated performance testing to find performance breaking points. Current approaches to tackle automated generation of performance test cases mainly involve using source code or system model analysis or use-case-based techniques. However, source code and system models might not always be available at testing time. On the other hand, if the optimal performance testing policy for the intended objective in a testing process instead could be learned by the testing system, then test automation without advanced performance models could be possible. Furthermore, the learned policy could later be reused for similar software systems under test, thus leading to higher test efficiency. We propose SaFReL, a self-adaptive fuzzy reinforcement learning-based performance testing framework. SaFReL learns the optimal policy to generate performance test cases through an initial learning phase, then reuses it during a transfer learning phase, while keeping the learning running and updating the policy in the long term. Through multiple experiments in a simulated performance testing setup, we demonstrate that our approach generates the target performance test cases for different programs more efficiently than a typical testing process and performs adaptively without access to source code and performance models.

show abstract

“…They have deduced that for CPU intensive workloads, the use of absolute metrics can result in better scaling decisions. Jindal et al [23] addressed the performance modeling of microservices by evaluation of a microservices web application. They identified a microservice's capacity in terms of the number of requests to find the appropriate resources needed for the microservices such that, the system would not violate the performance (response time, latency) requirements.…”

Section: Related Workmentioning

confidence: 99%

From DevOps to NoOps: Is It Worth It?

Jindal

Gerndt

2021

Communications in Computer and Information Science

Self Cite

View full text Add to dashboard Cite

With the rise of the adoption of microservice architecture due to its agility, scalability, and resiliency for building the cloud-based applications and their deployment using containerization, DevOps were in demand for handling the development and operations together. However, nowadays serverless computing offers a new way of developing and deploying cloud-native applications. Serverless computing also called NoOps, offloads management and server configuration (operations work) from the user to the cloud provider and lets the user focus only on the product developments. Hence, there are debates regarding which deployment strategy to use. This research provides a performance comparison of a cloud-native web application along with three different function benchmarks in terms of scalability, reliability, and latency when deployed using DevOps and NoOps deployment strategy. NoOps deployment in this work is achieved using Google Cloud Function and OpenWhisk, while DevOps is achieved using the Kubernetes engine. This research shows that neither of the deployment strategies fits all the scenarios. The experimental results demonstrate that each type of deployment strategy has its advantages under different scenarios. The DevOps deployment strategy has a huge performance advantage (almost 72% lesser 90 percentile response time) for simple web-based requests and requests accessing databases while compute-intensive applications perform better with NoOps deployment. Additionally, NoOps deployment provides better scaling-agility as compared to DevOps.

show abstract

Performance Modeling for Cloud Microservice Applications

Cited by 83 publications

References 7 publications

Service Placement for Latency Reduction in the Fog Using Application Profiles

Service Placement for Latency Reduction in the Fog Using Application Profiles

An autonomous performance testing framework using self-adaptive fuzzy reinforcement learning

From DevOps to NoOps: Is It Worth It?

Contact Info

Product

Resources

About