|
| This Article | ||
| ||
| Share | ||
| Bibliographic References | ||
| Add to: | ||
 Digg  Furl  Spurl  Blink  Simpy  Google  Del.icio.us  Y!MyWeb | ||
| Search | ||
| ||
| ASCII Text | x | ||
| Vitus J. Leung, Esther M. Arkin, Michael A. Bender, David Bunde, Jeanette Johnston, Alok Lal, Joseph S. B. Mitchell, Cynthia Phillips, Steven S. Seiden, "Processor Allocation on Cplant: Achieving General Processor Locality Using One-Dimensional Allocation Strategies," 2012 IEEE International Conference on Cluster Computing, pp. 296, Fourth IEEE International Conference on Cluster Computing (CLUSTER'02), 2002. | |||
| BibTex | x | ||
| @article{ 10.1109/CLUSTR.2002.1137758, author = {Vitus J. Leung and Esther M. Arkin and Michael A. Bender and David Bunde and Jeanette Johnston and Alok Lal and Joseph S. B. Mitchell and Cynthia Phillips and Steven S. Seiden}, title = {Processor Allocation on Cplant: Achieving General Processor Locality Using One-Dimensional Allocation Strategies}, journal ={2012 IEEE International Conference on Cluster Computing}, volume = {0}, year = {2002}, isbn = {0-7695-1745-5}, pages = {296}, doi = {http://doi.ieeecomputersociety.org/10.1109/CLUSTR.2002.1137758}, publisher = {IEEE Computer Society}, address = {Los Alamitos, CA, USA}, } | |||
| RefWorks Procite/RefMan/Endnote | x | ||
| TY - CONF JO - 2012 IEEE International Conference on Cluster Computing TI - Processor Allocation on Cplant: Achieving General Processor Locality Using One-Dimensional Allocation Strategies SN - 0-7695-1745-5 SP EP A1 - Vitus J. Leung, A1 - Esther M. Arkin, A1 - Michael A. Bender, A1 - David Bunde, A1 - Jeanette Johnston, A1 - Alok Lal, A1 - Joseph S. B. Mitchell, A1 - Cynthia Phillips, A1 - Steven S. Seiden, PY - 2002 KW - null VL - 0 JA - 2012 IEEE International Conference on Cluster Computing ER - | |||
The Computational Plant or Cplant is a commodity-based supercomputer under development at Sandia National Laboratories. This paper describes resource-allocation strategies to achieve processor locality for parallel jobs in Cplant and other supercomputers. Users of Cplant and other Sandia supercomputers submit parallel jobs to a job queue. When a job is scheduled to run, it is assigned to a set of processors. To obtain maximum throughput, jobs should be allocated to localized clusters of processors to minimize communication costs and to avoid bandwidth contention caused by overlapping jobs.
This paper introduces new allocation strategies and performance metrics based on space-filling curves and one dimensional allocation strategies. These algorithms are general and simple. Preliminary simulations and Cplant experiments indicate that both space-filling curves and one-dimensional packing improve processor locality compared to the sorted free list strategy previously used on Cplant. These new allocation strategies are implemented in the new release of the Cplant System Software, Version 2.0, phased into the Cplant systems 1 at Sandia by May 2002.