Burstiness-aware service level planning for enterprise application clouds

Anas A Youssef,Diwakar Krishnamurthy

doi:10.1186/s13677-017-0087-y

Abstract

Enterprise applications are being increasingly deployed on cloud infrastructures. Often, a cloud service provider (SP) enters into a Service Level Agreement (SLA) with a cloud subscriber, which specifies performance requirements for the subscriber’s applications. An SP needs systematic Service Level Planning (SLP) tools that can help estimate the resources needed and hence the cost incurred to satisfy their customers’ SLAs. Enterprise applications typically experience bursty workloads and the impact of such bursts needs to be considered during SLP exercises. Unfortunately, most existing approaches do not consider workload burstiness. We propose a Resource Allocation Planning (RAP) technique, which allows an SP to identify a time varying allocation plan of resources to applications that satisfies bursts. Extensive simulation results show that the proposed RAP variants can identify resource allocation plans that satisfy SLAs without exhaustively generating all possible plans. Furthermore, the results show that RAP can permit SPs to more accurately determine the capacity required for meeting specified SLAs compared to other competing techniques especially for bursty workloads.

Highlights

Enterprises are beginning to increasingly rely on cloud computing systems for hosting their applications
In the first experiment four different applications are subjected to a non-bursty workload, i.e., exponential session arrivals, over a planning horizon of 4 h with a resource allocation interval of 1 h
The maximum number of available web server instances per resource allocation interval is set to 7. These settings allow the exhaustive enumeration of all possible resource allocation plans for these applications

Summary

Introduction

Enterprises are beginning to increasingly rely on cloud computing systems for hosting their applications. We consider a performance-aware cloud system that is capable of providing performance guarantees to its subscribers In such an environment, a cloud Service Provider (SP) enters into a Service Level Agreement (SLA) with a subscriber prior to deploying the subscriber’s applications. In contrast to a traditional cloud system, In a performance-aware cloud system, the SP needs systematic Service Level Planning (SLP) tools to guide the process of formulating performance SLAs with subscribers and translating them to application resource allocations. An SP should consider the application’s workload as well as the workload of other applications on the cloud to determine whether there is adequate capacity to satisfy this performance requirement. Our work focuses on SLP tools that can help SPs undertake such exercises

Methods

Results

Conclusion