Influence of performance prediction inaccuracy on task scheduling in grid environment

Authors:
Yuanyuan Zhang;Yasushi Inoguchi
Affiliations:
Graduate School of Information Science;Japan Advanced Institute of Science and Technology, Center for Information Science, Ishikawa, Japan
Venue:
APWeb'05 Proceedings of the 7th Asia-Pacific web conference on Web Technologies Research and Development
Year:
2005

Citing 2
Cited 0

The grid: blueprint for a new computing infrastructure

The grid: blueprint for a new computing infrastructure
The network weather service: a distributed resource performance forecasting service for metacomputing

Future Generation Computer Systems - Special issue on metacomputing

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we study the influence of performance prediction inaccuracy on task scheduling in grid environment from the context of task selection and processor selection, which are two critical phases in task scheduling. Formulas are established for the degree of misprediction, the probability that the predicted values for the performance of tasks and processors reveal different ordering characteristics from their real values. The impacts of different parameters on the degree of misprediction are also investigated extensively. Evaluation results show that an underestimate of performance can result in greater influence on task scheduling compared with an overestimate, while higher heterogeneity results in smaller influence.