TY - GEN
T1 - Compiler and runtime support for out-of-core HPF programs
AU - Thakur, Rajeev
AU - Bordawekar, Rajesh
AU - Choudhary, Alok
N1 - Funding Information:
Wewould like to thank Geoffrey Fox, Ken Kennedy, Chuck Koelbel, Ravi Ponnusamy and Joel %ltz for many enlightening discussions. We also thank our compiler group at Syracuse University for their help with the basic infrastructure of the HPF compiler. Thk work was sponsored in part by NSF Young Investigator Award CCR-9357840 with a matching grant from Intel SSD, and also by ARPA under contract no. DABT63-91-C-O028. The content of the information does not necessarily reflect the position or policy of the Government and no official endorsement should be inferred. Thk research was performed in part using the Intel Touchstone Delta System operated by Caltech on behalf of the Concurrent Supercomputing Consortium. Access this facility was provided by CRPC.
Publisher Copyright:
© 1994 ACM.
PY - 1994/7/16
Y1 - 1994/7/16
N2 - This paper describes the design of a compiler which can translate out-of-core programs written in a data parallel language like HPF. Such a compiler is required for compiIing large scale scientific applications, such as the Grand Chalienge applications, which deal with enormous quantities of data. We propose a framework by which a compiler together with appropriate runtime support can translate an out-of-core HPF program to a message passing node program with explicit parallel I/O. We describe the basic model of the compiler and the various transformations made by the compiler. We also discuss the runtime routines used by the comrriler for I/O and communication. In order to minimize I/O. The runtirne support system can reuse data already fetched into memory. The working of the compiler is illustrated using two out-of-core applications, namely a Laplace equation solver and LU Decomposition, together with performance results on the Intel Touchstone Delta.
AB - This paper describes the design of a compiler which can translate out-of-core programs written in a data parallel language like HPF. Such a compiler is required for compiIing large scale scientific applications, such as the Grand Chalienge applications, which deal with enormous quantities of data. We propose a framework by which a compiler together with appropriate runtime support can translate an out-of-core HPF program to a message passing node program with explicit parallel I/O. We describe the basic model of the compiler and the various transformations made by the compiler. We also discuss the runtime routines used by the comrriler for I/O and communication. In order to minimize I/O. The runtirne support system can reuse data already fetched into memory. The working of the compiler is illustrated using two out-of-core applications, namely a Laplace equation solver and LU Decomposition, together with performance results on the Intel Touchstone Delta.
UR - http://www.scopus.com/inward/record.url?scp=85017252965&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85017252965&partnerID=8YFLogxK
U2 - 10.1145/181181.181571
DO - 10.1145/181181.181571
M3 - Conference contribution
AN - SCOPUS:85017252965
T3 - Proceedings of the International Conference on Supercomputing
SP - 382
EP - 391
BT - Proceedings of the 8th International Conference on Supercomputing, ICS 1994
PB - Association for Computing Machinery
T2 - 8th International Conference on Supercomputing, ICS 1994
Y2 - 11 July 1994 through 15 July 1994
ER -