The maximal utilization of processor co-allocation in multicluster systems

Bucur, Anca; Epema, Dick

doi:10.1109/ipdps.2003.1213154

Cited by 24 publications

(25 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…One of the most important findings is that when the slowdown of jobs due to the wide-area communication is less than or equal to 1:25, it pays to use co-allocation. In [5], we consider the maximal utilization, i.e., the utilization at which the system becomes saturated, as a metric for assessing the performance of processor coallocation. …”

Section: Related Workmentioning

confidence: 99%

A Dynamic Co-allocation Service in Multicluster Systems

Sinaga

Mohamed

Epema

2005

Job Scheduling Strategies for Parallel Processing

View full text Add to dashboard Cite

show abstract

Section: Related Workmentioning

confidence: 99%

A Dynamic Co-allocation Service in Multicluster Systems

Sinaga

Mohamed

Epema

2005

Job Scheduling Strategies for Parallel Processing

View full text Add to dashboard Cite

show abstract

“…Basically, the CF policy tries to reduce the overhead of waiting in multiple clusters for input files to become available in the right locations. We have performed comparison experiments of CF with the Worst-Fit (WF) jobplacement policy, which is the default policy of KOALA, whose aim is to keep the loads in the clusters balanced [7]. The results showed that the combination of the CF policy and file replication is very beneficial when jobs have large input files (i.e., larger than 2GB).…”

Section: Introductionmentioning

confidence: 99%

Communication-Aware Job Placement Policies for the KOALA Grid Scheduler

Sonmez¹,

Mohamed²,

Epema³

2006

2006 Second IEEE International Conference on E-Science and Grid Computing (E-Science'06)

View full text Add to dashboard Cite

show abstract

“…As grid technologies gain in popularity, separate clusters are increasingly being interconnected to create multicluster computing architectures for the processing of scientific and commercial applications [1][2] [5]. These constituent clusters may be located within a single organization or across different geographical sites [3] [6].…”

Section: Introductionmentioning

confidence: 99%

Performance-Aware Load Balancing for Multiclusters

Jarvis

Bacigalupo

et al. 2004

Parallel and Distributed Processing and Applications

View full text Add to dashboard Cite

Abstract. In a multicluster architecture, where jobs can be submitted through each constituent cluster, the job arrival rates in individual clusters may be uneven and the load therefore needs to be balanced among clusters. In this paper we investigate load balancing for two types of jobs, namely non-QoS and QoSdemanding jobs and as a result, two performance-specific load balancing strategies (called ORT and OMR) are developed. The ORT strategy is used to obtain the optimised mean response time for non-QoS jobs and the OMR strategy is used to achieve the optimised mean miss rate for QoS-demanding jobs. The ORT and OMR strategies are mathematically modelled combining queuing network theory to establish sets of optimisation equations. Numerical solutions are developed to solve these optimisation equations, and a so called fair workload level is determined for each cluster. When the current workload in a cluster reaches this pre-calculated fair workload level, the jobs subsequently submitted to the cluster are transferred to other clusters for execution. The effectiveness of both strategies is demonstrated through theoretical analysis and experimental verification. The results show that the proposed load balancing mechanisms bring about considerable performance gains for both job types, while the job transfer frequency among clusters is considerably reduced. This has a number of advantages, in particular in the case where scheduling jobs to remote resources involves the transfer of large executable and data files.

show abstract

The maximal utilization of processor co-allocation in multicluster systems

Abstract: Abstract

Cited by 24 publications

References 16 publications

A Dynamic Co-allocation Service in Multicluster Systems

A Dynamic Co-allocation Service in Multicluster Systems

Communication-Aware Job Placement Policies for the KOALA Grid Scheduler

Performance-Aware Load Balancing for Multiclusters

Contact Info

Product

Resources

About