Accession Number:

ADA461275

Title:

Data Mining Meets Performance Evaluation: Fast Algorithms for Modeling Bursty Traffic

Descriptive Note:

Conference paper

Corporate Author:

CARNEGIE-MELLON UNIV PITTSBURGH PA SCHOOL OF COMPUTER SCIENCE

Report Date:

2001-04-01

Pagination or Media Count:

18.0

Abstract:

Network, web, and disk IO traffic are usually bursty, self-similar 9, 3, 5, 6 and therefore can not be modeled adequately with Poisson arrivals9. However, we do want to model these types of traffic and to generate realistic traces, because of obvious applications for disk scheduling, network management, web server design. Previous models like fractional Brownian motion, ARFIMA etc tried to capture the burstiness . However the proposed models either require too many parameters to fit andor require prohibitively large quadratic time to generate large traces. We propose a simple, parsimonious method, the b-model , which solves both problems It requires just one parameter b, and it can easily generate large traces. In addition, it has many more attractive properties aWith our proposed estimation algorithm, it requires just a single pass over the actual trace to estimate b. For example, a one-day-long disk trace in milliseconds contains about 86Mb data points and requires about 3 minutes for model fitting and 5 minutes for generation. b The resulting synthetic traces are very realistic our experiments on real disk and web traces show that our synthetic traces match the real ones very well in terms of queuing behavior.

Subject Categories:

  • Computer Hardware
  • Computer Systems
  • Computer Programming and Software

Distribution Statement:

APPROVED FOR PUBLIC RELEASE