You do not have permission to edit this page, for the following reason:
You can view and copy the source of this page.
== Abstract ==
This paper details the development and application of a model for predictive performance analysis of a pipelined synchronous wavefront application running on commodity processor cluster systems. The performance model builds on existing work (Cao et al.) by including extensions for modern commodity processor architectures. These extensions, including coarser hardware benchmarking, prove to be essential in countering the effects of modern superscalar processors (e.g. multiple operation pipelines and on-the-fly optimisations), complex memory hierarchies, and the impact of applying modern optimising compilers. The process of application modelling is also extended, combining static source code analysis with run-time profiling results for increased accuracy. The model is validated on several high performance SMP systems and the results show a high predictive accuracy (les 10% error). Additionally, the use of the performance model to speculate on the performance and scalability of this application on a hypothetical cluster with two different problem sizes is demonstrated. It is shown that such speculative techniques can be used to support system procurement, run-time verification and system maintenance and upgrading
== Original document ==
The different versions of the original document can be found in:
* [http://wrap.warwick.ac.uk/30961/7/WRAP-Predictive-performance-analysis-parallel-pipelined-synchronous-wavefront-Jarvis-2020.pdf http://wrap.warwick.ac.uk/30961/7/WRAP-Predictive-performance-analysis-parallel-pipelined-synchronous-wavefront-Jarvis-2020.pdf]
* [http://xplorestaging.ieee.org/ielx5/4100332/4100333/04100394.pdf?arnumber=4100394 http://xplorestaging.ieee.org/ielx5/4100332/4100333/04100394.pdf?arnumber=4100394],
: [http://dx.doi.org/10.1109/clustr.2006.311888 http://dx.doi.org/10.1109/clustr.2006.311888]
* [https://dblp.uni-trier.de/db/conf/cluster/cluster2006.html#MudaligeJSN06 https://dblp.uni-trier.de/db/conf/cluster/cluster2006.html#MudaligeJSN06],
: [http://yadda.icm.edu.pl/yadda/element/bwmeta1.element.ieee-000004100394 http://yadda.icm.edu.pl/yadda/element/bwmeta1.element.ieee-000004100394],
: [http://www.oerc.ox.ac.uk/sites/default/files/uploads/profile-pages/Gihan/cluster2006.pdf http://www.oerc.ox.ac.uk/sites/default/files/uploads/profile-pages/Gihan/cluster2006.pdf],
: [http://dx.doi.org/10.1109/CLUSTR.2006.311888 http://dx.doi.org/10.1109/CLUSTR.2006.311888],
: [http://wrap.warwick.ac.uk/30961 http://wrap.warwick.ac.uk/30961],
: [https://www.researchgate.net/profile/Gihan_Mudalige2/publication/221202182_Predictive_Performance_Analysis_of_a_Parallel_Pipelined_Synchronous_Wavefront_Application_for_Commodity_Processor_Cluster_Systems/links/0c96052c16ce58dd92000000.pdf https://www.researchgate.net/profile/Gihan_Mudalige2/publication/221202182_Predictive_Performance_Analysis_of_a_Parallel_Pipelined_Synchronous_Wavefront_Application_for_Commodity_Processor_Cluster_Systems/links/0c96052c16ce58dd92000000.pdf],
: [https://www.mysciencework.com/publication/show/predictive-performance-analysis-parallel-pipelined-synchronous-wavefront-application-commodity-processor-cluster-system-a6ad881f https://www.mysciencework.com/publication/show/predictive-performance-analysis-parallel-pipelined-synchronous-wavefront-application-commodity-processor-cluster-system-a6ad881f],
: [https://academic.microsoft.com/#/detail/2118932108 https://academic.microsoft.com/#/detail/2118932108]
Return to Spooner et al 2007a.
Published on 01/01/2007
Volume 2007, 2007
DOI: 10.1109/clustr.2006.311888
Licence: CC BY-NC-SA license
Are you one of the authors of this document?