Experiences with predicting resource performance on-line in computational grid settings
Author:
Open Access
- 1 March 2003
- journal article
- Published by Association for Computing Machinery (ACM) in ACM SIGMETRICS Performance Evaluation Review
- Vol. 30 (4) , 41-49
Abstract
In this paper, we describe methods for predicting the performance of Computational Grid resources (machines, networks, storage systems, etc.) using computationally inexpensive statistical techniques. The predictions generated in this manner are intended to support adaptive application scheduling in Grid settings, and on-line fault detection. Wedescribe a mixture-of-experts approach to non-parametric, univariate time-series forecasting, and detail the effectiveness of the approach using example data gathered from "production" (i.e. non-experimental) Computational Grid installations.Keywords
This publication has 16 references indexed in Scilit:
- High-performance remote access to climate simulation data2001
- Numerical libraries and the grid2001
- The GrADS Project: Software Support for High-Level Grid Application DevelopmentThe International Journal of High Performance Computing Applications, 2001
- Writing programs that run EveryWare on the Computational GridIEEE Transactions on Parallel and Distributed Systems, 2001
- On the constancy of internet path properties2001
- The network weather service: a distributed resource performance forecasting service for metacomputingFuture Generation Computer Systems, 1999
- Using pathchar to estimate Internet link characteristics1999
- Application level scheduling of gene sequence comparison on metacomputers1998
- Application-level scheduling on distributed heterogeneous networks1996
- Wide area traffic: the failure of Poisson modelingIEEE/ACM Transactions on Networking, 1995