Statistical scalability analysis of communication operations in distributed applications
PPoPP '01 Proceedings of the eighth ACM SIGPLAN symposium on Principles and practices of parallel programming
IPDPS '02 Proceedings of the 16th International Parallel and Distributed Processing Symposium
TAU: A Portable Parallel Program Analysis Environment for pC++
CONPAR 94 - VAPP VI Proceedings of the Third Joint International Conference on Vector and Parallel Processing: Parallel Processing
Scalable analysis techniques for microprocessor performance counter metrics
Proceedings of the 2002 ACM/IEEE conference on Supercomputing
Evaluating similarity-based trace reduction techniques for scalable performance analysis
Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis
Scalable event trace visualization
Euro-Par'09 Proceedings of the 2009 international conference on Parallel processing
Trace profiling: Scalable event tracing on high-end parallel systems
Parallel Computing
Concurrency and Computation: Practice & Experience
Traces generation to simulate large-scale distributed applications
Proceedings of the Winter Simulation Conference
Hi-index | 0.00 |
Often parallel scientific applications are instrumented and traces are collected and analyzed to identify processes with performance problems or operations that cause delays in program execution. The execution of instrumented codes may generate large amounts of performance data, and the collection, storage, and analysis of such traces are time and space demanding. To address this problem, this paper presents an efficient, systematic, multi-step methodology, based on hierarchical clustering, for analysis of communication traces of parallel scientific applications. The methodology is used to discover potential communication performance problems of three applications: TRACE, REMO, and SWEEP3D.