Compiler and Software Distributed Shared Memory Support for Irregular Applications

Compiler and Software Distributed Shared Memory Support for Irregular Applications (1997)

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Honghui Lu , Alan L. Cox , Sandhya Dwarkadas , Ramakrishnan Rajamony , Winy Zwaenepoell

Citations:

46 - 3 self

Versions

Version History

Metadata Version 1

User correction supplied by mph

Datum	Value	Source
TITLE	Compiler and Software Distributed Shared Memory Support for Irregular Applications	SVM HeaderParse 0.2
AUTHOR NAME	Honghui Lu	user correction
AUTHOR AFFIL	Department of Electrical and Computer Engg 0 Department of Computer Science; $ Department of Computer Science; Rice University University of Rochester	user correction
AUTHOR NAME	Alan L. Cox	user correction
AUTHOR AFFIL	Department of Electrical and Computer Engg 0 Department of Computer Science; $ Department of Computer Science; Rice University University of Rochester	user correction
AUTHOR NAME	Sandhya Dwarkadas	user correction
AUTHOR AFFIL	Department of Electrical and Computer Engg 0 Department of Computer Science; $ Department of Computer Science; Rice University University of Rochester	user correction
AUTHOR NAME	Ramakrishnan Rajamony	user correction
AUTHOR AFFIL	Department of Electrical and Computer Engg 0 Department of Computer Science; $ Department of Computer Science; Rice University University of Rochester	user correction
AUTHOR NAME	Winy Zwaenepoell	user correction
AUTHOR AFFIL	Department of Electrical and Computer Engg 0 Department of Computer Science; $ Department of Computer Science; Rice University University of Rochester	user correction
ABSTRACT	We investigate the use of a software distributed shared memory (DSM) layer to support irregular computations on distributed memory machines. Software DSM supports irregular computation through demand fetching of data in response to memory access faults. With the addition of a very limited form of compiler support, namely the identification of the section of the indirection array accessed by each processor, many of these on-demand page fetches can be aggregated into a single message, and prefetched prior to the access fault. We have measured the performance of this approach for two irregular applications, moldyn and nbf, using the Tread-Marks DSM system on an 8-processor IBM SP2. We find that it has similar performance to the inspector-executor method supported by the CHAOS run-time library, while requiring much simpler compile-time support. For moldyn, it is up to 23~0 faster than CHAOS, depending on the input problem’s characteristics; and for nbf, it is no worse than 14 % slower. If we include the execution time of the inspector, the software DSM-based approach is always faster than CHAOS. The advantage of this approach increases as the frequency of changes to the indirection array increases. The disadvantage of this approach is the potential for false sharing overhead when the data set is small or has poor spatial locality,	user correction
YEAR	1997	user correction
CITATIONS	20 found	ParsCit 1.0