Date of Award

3-21-2013

Document Type

Thesis

Degree Name

Master of Science

Department

Department of Electrical and Computer Engineering

First Advisor

Kenneth M. Hopkinson, PhD.

Abstract

This thesis researches cloud computing workload characteristics and synthetic workload generation. A heuristic presented in the work guides the process of workload trace characterization and synthetic workload generation. Analysis of a cloud trace provides insight into client request behaviors and statistical parameters. A versatile workload generation tool creates client connections, controls request rates, defines number of jobs, produces tasks within each job, and manages task durations. The test system consists of multiple clients creating workloads and a server receiving request, all contained within a virtual machine environment. Statistical analysis verifies the synthetic workload experimental results are consistent with real workload behaviors and characteristics.

AFIT Designator

AFIT-ENG-13-M-11

DTIC Accession Number

ADA582340

Download

Share

COinS