Advanced load-testing techniques for a science archive

Abstract

Performance goals for data archive systems need to be established early in the design process to ensure stability and acceptable response throughput. Load testing is one technique used to measure the progress towards these performance goals. Providing resources for load-test planning is critical, and this planning must include feasibility studies, tool analyses, and generation of an overall load-test strategy. This strategy is much different for science data archives than other systems, including commercial websites and high-volume data centers. This paper will provide an overview of the load testing performed on the Spitzer Space Telescope's science archive, which is part of Science Operations System at the Spitzer Science Center (SSC). Methods used for planning and conducting SSC load tests will be presented, and advanced load-testing techniques will be provided to address runtime issues and enhance verification results. This work was performed at the California Institute of Technology under contract to the National Aeronautics and Space Administration.