PERFORMANCE PREDICTION VIA MODELING: A CASE STUDY OF THE ORNL CRAY XT4 UPGRADE
Abstract
We present predictive performance models of two of the petascale applications, S3D and GTC, from the DOE Office of Science workload. We outline the development of these models and demonstrate their validation on an Opteron/Infiniband cluster and the pre-upgrade ORNL Jaguar system (Cray XT3/XT4). Given the high accuracy of the full application models, we predict the performance of the Jaguar system after the upgrade of its nodes, and subsequently compare this to the actual performance of the upgraded system. We then analyze the performance of the system based on the models to quantify bottlenecks and potential optimizations. Finally, the models are used to quantify the benefits of alternative node allocation strategies, and to quantify performance degradation resulting from inter-process competition for network resources.
References
-
F. Petrini , D.J. Kerbyson and S. Pakin , The case of the missing supercomputer performance: achieving optimal performance on the 8, 192 processors of ASCI Q , Proc. IEEE/ACM SuperComputing (SC03) ( 2003 ) . Google Scholar - Oak Ridge National Laboratory, Jaguar System. http://www.nccs.gov/computing-resources/jaguar/ . Google Scholar
-
T. Simon , S. Cable and M. Mahmoodi , Application Scalability and Performance on Multicore Architectures , HPCMP Users Group Conference ( 2007 ) . Google Scholar -
K.J. Barker , Experiences in Scaling Scientific Applications on Current-generation Quad-core Processors , Proc. Workshop on Large-Scale Parallel Processing (LSPP) ( FL , Miami , 2008 ) . Google Scholar - Parallel Processing Letters 18(4), 453 (2008), DOI: 10.1142/S012962640800351X. Link, Google Scholar
- Journal of Physics: Conference Series 16, 1 (2005), DOI: 10.1088/1742-6596/16/1/001. Crossref, ISI, Google Scholar
- Journal of Physics: Conference Series 16, 65 (2005), DOI: 10.1088/1742-6596/16/1/009. Crossref, ISI, Google Scholar
- Cray Users Groug (CUG) (2007). Google Scholar
-
D.J. Kerbyson , A Look at Application Performance Sensitivity to the Bandwidth and Latency of Infiniband Networks , Proc. Workshop on Communication Architectures for Clusters (CAC) ( Greece , 2006 ) . Google Scholar -
P.H. Worley , Comparison of Cray XT3 and XT4 Scalability , Proc Cray User's Group (CUG) ( 2007 ) . Google Scholar -
S. R. Alam , Cray XT4: An Early Evaluation for Petascle Scientific Simulation , Proc. IEEE/ACM Supercomputing (SC07) ( 2007 ) . Google Scholar -
K. J. Barker , Performance Modeling in Action: Performance Prediction and Optimization of the Jaguar System during Upgrade , Proc. Workshop on Large-Scale Parallel Processing (LSPP) ( 2009 ) . Google Scholar


