Scheduling Policies for Processor Coallocation in Multicluster Systems
Authors:
- Dick H.J. EpemaDEDick H.J. Epema
- 4 June 2007
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 18 (7) , 958-972
Abstract
Building multicluster systems out of multiple, geographically distributed clusters interconnected by high-speed wide-area networks can provide access to a larger computational power and to a wider range of resources. Jobs running on multiclusters and, more generally, in grids, may require (processor) coallocation, i.e., the simultaneous allocation of resources (processors) in different clusters or subsystems of a grid. In this paper, we propose four scheduling policies for processor coallocation in multiclusters, and we assess with simulations their performance under a wide variety of parameter settings. In particular, in our simulations we use synthetic workloads and workloads derived from the logs of actual systems and from runtime measurements. We conclude that although coallocation makes scheduling more difficult and the wide-area communication critically impacts the performance, there is a wide range of realistic applications that may benefit from coallocation. However, unrestricted coallocation is not recommended: Limiting the total job size or the number or the sizes of their components improves performance.Keywords
This publication has 16 references indexed in Scilit:
- An evaluation of the close-to-files processor and data co-allocation policy in multiclusters2005
- Experiences with the KOALA co-allocating scheduler in multiclusters2005
- On Advantages of Grid Computing for Parallel Job Scheduling2003
- Resource co-allocation in computational grids2003
- A Measurement-Based Simulation Study of Processor Co-allocation in Multicluster Systems2003
- Optimizing parallel applications for wide-area clusters2002
- Characteristics of a Large Shared Memory Production Workload2001
- Sensitivity of parallel applications to large differences in bandwidth and latency in two-layer interconnectsFuture Generation Computer Systems, 2001
- The distributed ASCI Supercomputer projectACM SIGOPS Operating Systems Review, 2000
- Wide-area parallel programming using the remote method invocation modelConcurrency: Practice and Experience, 2000