Scalability resilience framework using application-level fault injection for cloud-based software services

Amro Al-Said Ahmad,Peter Andras

doi:10.1186/s13677-021-00277-z

Abstract

This paper presents an investigation into the effect of faults on the scalability resilience of cloud-based software services. The study introduces an experimental framework using the Application-Level Fault Injection (ALFI) to investigate how the faults at the application level affect the scalability resilience and behaviour of cloud-based software services. Previous studies on scalability analysis of cloud-based software services provide a baseline of the scalability behaviour of such services, allowing to conduct in-depth scalability investigation of these services. Experimental analysis on the EC2 cloud using a real-world cloud-based software service is used to demonstrate the framework, considering delay latency of software faults with two varied settings and two demand scenarios. The experimental approach is explained in detail. Here we simulate delay latency injection with two different times, 800 and 1600 ms, and compare the results with the baseline data. The results show that the proposed approach allows a fair assessment of the fault scenario’s impact on the cloud software service’s scalability resilience. We explain the use of the methodology to determine the impact of injected faults on the scalability behaviour and resilience of cloud-based software services.

Highlights

As cloud-based software services have become more popular and dependable, evaluating the performance of such services is more critical than before
The second scenario is a stepped increase and decreases, again with a set peak level of workload; with this scenario, we Application-level fault injection for scalability resilience This paper aims to establish a framework for measuring the scalability resilience of cloud-based software services and to investigate the effect of runtime fault injection at the application level on the scalability behaviour of such services
This paper presents an experimental analysis of the impact of fault injection on the scalability resilience of cloud-based software services

Summary

Introduction

As cloud-based software services have become more popular and dependable, evaluating the performance of such services is more critical than before. Previous research studies [1, 2] have focused on measuring the scalability performance of such services to collect the right measurements and set up specific metrics such as technical evaluation metrics and infrastructure-monitoring metrics. These metrics are important to set a baseline for the scalability performance behaviour of these services. Fault injection has been utilized to analyse the dependability and reliability of cloud-based software systems [10–12]

Objectives

Findings

Discussion

Conclusion