|
All-Pairs Abstraction
The All-Pairs abstraction computes the Cartesian product of two sets, generating a matrix where each cell M[i,j] contains the output of the function F on objects A[i] and B[j]. You provide two sets of data files and a function F that computes on them. You may optionally provide additional parameters to control the actual computation(e.g. compute only part of the matrix). The abstraction then executes the computation in parallel, automatically handling load balancing, data movements, fault tolerance, and so on.
For More InformationPublicationsHarnessing Parallelism in Multicore Clusters with the All-Pairs, Wavefront, and Makeflow Abstractions, Journal of Cluster Computing, 13(3), pages 243-256, September, 2010. DOI: 10.1007/s10586-010-0134-7
All-Pairs: An Abstraction for Data Intensive Computing on Campus Grids, IEEE Transactions on Parallel and Distributed Systems, 21(1), pages 33-46, January, 2010. DOI: 10.1109/TPDS.2009.49
Exploiting Locality with QThreads for Portable Parallel Performance, Ph.D. Thesis, University of Notre Dame, November, 2009.
Harnessing Parallelism in Multicore Clusters with the All-Pairs and Wavefront Abstractions, IEEE High Performance Distributed Computing, pages 1-10, June, 2009. DOI: 10.1145/1551609.1551613
All-Pairs: An Abstraction for Data Intensive Cloud Computing, IEEE International Parallel and Distributed Processing Symposium (IPDPS), pages 1-11, April, 2008. DOI: 10.1109/IPDPS.2008.4536311
Poster: All-Pairs: An Abstraction for Data Intensive Computing, IEEE/ACM Grid Computing, September, 2007.
| ||||||||||||||||||||||||||||||||||||||||