Abstract
The design of ScaLAPACK, a scalable software library for performing dense and banded linear algebra computations on distributed memory concurrent computers, is discussed. They are implemented using a block cyclic data distribution, and are built using de facto standard kernels for matrix and vector operations and message passing communication. The details of the implementation of the ScaLAPACK LU factorization routine, and performance and scalability results on the Intel iPSC/680, Touchtone Delta, and Paragon systems are presented.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the IEEE Computer Society Workshop on Future Trends of Distributed Computing Systems|Proc IEEE Computer Soc Workshop Future Trends Distrib Comput |
| Place of Publication | Los Alamitos, CA, United States |
| Publisher | IEEE |
| Pages | 170-177 |
| Number of pages | 7 |
| Publication status | Published - 1995 |
| Event | Proceedings of the 5th IEEE Computer Society Workshop on Future Trends of Distributed Computing Systems - Cheju Island, South Korea Duration: 1 Jul 1995 → … http://dblp.uni-trier.de/db/conf/ftdcs/ftdcs1995.html#ChoiD95http://dblp.uni-trier.de/rec/bibtex/conf/ftdcs/ChoiD95.xmlhttp://dblp.uni-trier.de/rec/bibtex/conf/ftdcs/ChoiD95 |
Conference
| Conference | Proceedings of the 5th IEEE Computer Society Workshop on Future Trends of Distributed Computing Systems |
|---|---|
| City | Cheju Island, South Korea |
| Period | 1/07/95 → … |
| Internet address |