Paper Info
Title
Store-Ordered Streaming of Shared Memory
Abstract
Coherence misses in shared-memory multiprocessors account for a substantial fraction of execution time in many important scientific and commercial workloads. Memory streaming provides a promising solution to the coherence miss bottleneck because it improves memory level parallelism and lookahead while using on-chip resources efficiently. We observe that the order in which shared data are consumed by one processor is correlated to the order in which they were produced by another. We investigate this phenomenon and demonstrate that it can be exploited to send Store-ORDered Streams (SORDS) of shared data from producers to consumers, thereby eliminating coherent read misses. Using a trace-driven analysis of all user and OS memory references in a cache-coherent distributed shared-memory multiprocessor, we show that SORDS based memory streaming can eliminate between 36% and 100% of all coherent read misses in scientific workloads and between 23% and 48%in online transaction processing workloads.
Year
DOI
Venue
2005
10.1109/PACT.2005.37
IEEE PACT
Keywords
Field
DocType
online transaction processing workloads,shared data,memory level parallelism,store-ordered streams,shared-memory multiprocessors,coherent read,shared memory,scientific workloads,store-ordered streaming,commercial workloads,os memory reference,shared-memory multiprocessor,cache coherence,distributed shared memory,resource allocation,transaction processing,chip
Interleaved memory,Uniform memory access,Shared memory,Computer science,Parallel computing,Cache-only memory architecture,Distributed memory,Data diffusion machine,Real-time computing,Non-uniform memory access,Distributed shared memory
Conference
ISBN
Citations 
PageRank 
0-7695-2429-X
9
0.50
References 
Authors
27
7
Name
Order
Citations
PageRank
Thomas F. Wenisch12112105.25
Stephen Somogyi216012.72
Nikolaos Hardavellas322620.82
Jangwoo Kim444735.38
Chris Gniady529720.47
Anastasia Ailamaki64178349.12
Babak Falsafi74147242.18