Compiler Reduction of Invalidation Traffic in Virtual Shared Memory Systems
作者： M. F. P. O'Boyle , R. W. Ford , A. P. Nisbet
关键词: Memory ordering 、 Distributed memory 、 Memory management 、 Memory model 、 SPMD 、 Distributed shared memory 、 Extended memory 、 Execution model 、 Data diffusion machine 、 Cache coherence 、 Uniform memory access 、 Shared memory 、 Memory footprint 、 Computer science 、 Interleaved memory 、 Parallel computing 、 Compiler 、 Overlay 、 Memory map 、 Virtual memory
摘要: This paper presents new compiler analysis for the elimination of invalidation traffic in virtual shared memory, using a hybrid distributed coherence scheme. The and acknowledgement messages are removed; this reduces both network latency write fault. It aggressively exploits SPMD execution model uses array section to accurately determine only those instances when is necessary, thus avoiding additional read misses previous schemes. Equations determining precisely what data should be invalidated presented translated into form amenable analysis. Preliminary experimental results on 30 node prototype architecture demonstrate performance attainable
core.ac.uk
springer.com
ed.ac.uk
sci-hub.st 参考文章(10)
Ahmed Louri, Hongki Sung, A Compiler Directed Cache Coherence Scheme with Fast and Parallel Explicit Invalidation. international conference on parallel processing. pp. 2- 9 ,(1992)
A. P. Nisbet, R. W. Ford, Spinning-on-Coherency: A New VSM Optimisation for Write-Invalidate ieee international conference on high performance computing data and analytics. pp. 792- 797 ,(1996) , 10.1007/3-540-61142-8_628
C. J. Skelton, C. Hammer, M. Lopez, M. J. Reeve, P. Townsend, K. F. Wong, EDS: A Parallel Computer System for Advanced Information Processing international conference on parallel architectures and languages europe. pp. 1- 18 ,(1992) , 10.1007/3-540-55599-4_77
H. Cheong, A.V. Veidenbaum, Compiler-directed cache management in multiprocessors IEEE Computer. ,vol. 23, pp. 39- 47 ,(1990) , 10.1109/2.55499
Farnaz Mounes-Toussi, David J. Lilja, Zhiyuan Li, An evaluation of a compiler optimization for improving the performance of a coherence directory Proceedings of the 8th international conference on Supercomputing - ICS '94. pp. 75- 84 ,(1994) , 10.1145/181181.181281
Alvin R. Lebeck, David A. Wood, Dynamic self-invalidation: reducing coherence overhead in shared-memory multiprocessors international symposium on computer architecture. ,vol. 23, pp. 48- 59 ,(1995) , 10.1145/223982.223995
Hoichi Cheong, Life span strategy—a compiler-based approach to cache coherence international conference on supercomputing. pp. 139- 148 ,(1992) , 10.1145/143369.143402
Mark D. Hill, James R. Larus, Steven K. Reinhardt, David A. Wood, Cooperative shared memory: software and hardware for scalable multiprocessors ACM Transactions on Computer Systems. ,vol. 11, pp. 300- 318 ,(1993) , 10.1145/161541.161544
E. Darnell, K. Kennedy, Cache coherence using local knowledge conference on high performance computing (supercomputing). pp. 720- 729 ,(1993) , 10.1145/169627.169821
M. F. P. O'Boyle, R. W. Ford, A. P. Nisbet, Compiler Reduction of Invalidation Traffic in Virtual Shared Memory Systems european conference on parallel processing. pp. 432- 440 ,(1996) , 10.1007/3-540-61626-8_58