User-Centric Performance Analysis of Market-Based Cluster Batch Schedulers

This paper presents a performance analysis of market-based batch schedulers for clusters of workstations. In contrast to previous work, we use user-centric performance metrics as the basis for system evaluation. Each user is modeled as having a utility function for each job which measures value deli...

Full description

Saved in:
Bibliographic Details
Published inCluster Computing and the Grid (CCGrid 2002): 2nd IEEE International Symposium p. 30
Main Authors Chun, B.N., Culler, D.E.
Format Conference Proceeding
LanguageEnglish
Published IEEE 2002
Subjects
Aggregates
Clustering algorithms
Computer science
Cost accounting
Feedback
Performance analysis
Processor scheduling
Resource management
Scheduling algorithm
Time measurement
Online AccessGet full text
ISBN9780769515823
0769515827
DOI10.1109/CCGRID.2002.1017109

Cover

More Information
Summary:This paper presents a performance analysis of market-based batch schedulers for clusters of workstations. In contrast to previous work, we use user-centric performance metrics as the basis for system evaluation. Each user is modeled as having a utility function for each job which measures value delivered to the user as function of execution time. Summing over all utility functions in the workload, we use aggregate utility as a measure of overall value delivered to users. With aggregate utility as the performance metric, simulations are used to quantify the performance of both market-based and traditional batch scheduling algorithms under a variety of synthetic work-loads. Results show that an auction-based batch scheduling algorithm improves performance by a factor of up to 2-5x for sequential workloads and up to 14x for highly parallel workloads compared to traditional scheduling algorithms.
ISBN:9780769515823
0769515827
DOI:10.1109/CCGRID.2002.1017109