Keynote Presentation at LT 2015
02/02/2015Our latest research on Load Testing Elasticity and Performance Isolation in Shared Execution Environments was presented at the Fourth International Workshop on Large-Scale Testing (LT 2015), co-located with the 6th International Conference on Performance Engineering (ICPE 2015). The presentation slides are available for download.
The keynote was presented by Samuel Kounev on February 1st as part of the programme of the Fourth International Workshop on Large-Scale Testing (LT 2015), co-located with the 6th International Conference on Performance Engineering (ICPE 2015) .
The conference and workshop was held in Austin, TX, USA at the Hilton Garden Inn Austin Convention Center, January 31st - February 5th, 2015.
Download Presentation Slides: .pdf
Title: Load Testing Elasticity and Performance Isolation in Shared Execution Environments
Abstract: The inability to provide performance guarantees is a major challenge for the widespread adoption of shared execution environments, based on paradigms such as virtualization and cloud computing. Performance is a major distinguishing factor between different service offerings. To make such offerings comparable, novel metrics and techniques are needed allowing to measure and quantify the performance of shared execution environments under load, e.g., public cloud platforms or general virtualized service infrastructures. In this talk, we first discuss the inherent challenges of providing performance guarantees in the presence of highly variable workloads and load spikes. We then present novel metrics and techniques for shared execution environments, specifically taking into account the dynamics of modern service infrastructures. We consider both environments where virtualization is used as a basis for enabling resource sharing, e.g., as in Infrastructure-as-a-Service (IaaS) offerings, as well as multi-tenant Software-as-a-Service (SaaS) applications, where the whole hardware and software stack is shared among different customers. We focus on evaluating two aspects: i) the ability of the system to provision resources in an elastic manner, i.e., system elasticity, ii) the ability of the system to isolate different applications and customers sharing the physical infrastructure in terms of the performance they observe, i.e., performance isolation. We discuss the challenges in measuring and quantifying the mentioned two properties, presenting existing approaches to tackle them. Finally, we discuss open issues and emerging directions for future work.
