Characterizing Performance of Infrastructure Clouds

Project Details

Project Lead
Paul Marshall 
Project Manager
Paul Marshall 
Project Members
Ali Alzabarah, Dmitry Duplyakin, Yuan Sui, Yuan Sui  
Institution
University of Colorado at Boulder, Computer Science  
Discipline
Computer Science (401) 

Abstract

Infrastructure-as-a-Service (IaaS) clouds typically use system-level virtualization technologies to allow users to deploy custom software stacks, from the operating system upward. Often, the use of virtualization leads to multi-tenancy of the physical hardware where individual resources are shared by multiple users. Furthermore, cloud providers select and configure hardware and software stacks differently. Both multi-tenancy and configuration differences can lead to substantial performance variability. This project will focus on identifying metrics and benchmarks to evaluate the performance of IaaS clouds, furthermore, it will develop necessary tools and frameworks to automate the process of benchmarking and analyzing the performance of IaaS cloud offerings. Automated benchmarking toolkits will enable users to repeatedly evaluate the performance of IaaS clouds as configurations and offerings continually change.

Intellectual Merit

This project will provide a platform for understanding performance variations between different IaaS clouds, giving users the ability to thoroughly evaluate and compare different offerings.

Broader Impacts

Understanding the performance implications of IaaS clouds will provide useful information to users of all disciplines looking to leverage IaaS clouds for their work.

Scale of Use

Initially this project will consist of 3 members using IaaS clouds for development and minimal benchmarking. Most work should typically involve tens of VMs.