A self-scaling and self-configuring benchmark for Web servers (extended abstract)

  • Authors:

  • Stephen Manley;Margo Seltzer;Michael Courage

  • Affiliations:

  • Network Appliance, 2770 San Tomas Expressway, Santa Clara, CA;Harvard University, 29 Oxford Street, Cambridge, MA;Microsoft Corporation, One Microsoft Way, Redmond, WA

  • Venue:
  • SIGMETRICS '98/PERFORMANCE '98 Proceedings of the 1998 ACM SIGMETRICS joint international conference on Measurement and modeling of computer systems
  • Year:
  • 1998

Quantified Score

Hi-index 0.00

Visualization

Abstract

World Wide Web clients and servers have become some of the most important applications in our computing base, and we need realistic and meaningful ways of measuring their performance. Current server benchmarks do not capture the wide variation that we see in servers and are not accurate in their characterization of web traffic. In this paper, we present a self-configuring, scalable benchmark that generates a server benchmark load based on actual server loads. In contrast to other web benchmarks, our benchmark focuses on request latency instead of focusing exclusively on throughput sensitive metrics. We present our new benchmark hBench:Web, and demonstrate how it accurately models the load of an actual server. The benchmark can also be used to assess how continued growth or changes in the workload will affect future performance. Using existing log histories, we now that these predictions are sufficiently realistic to provide insight into tomorrow's Web performance.