TARDIS: Timestamp based Coherence Algorithm for Distributed Shared Memory
作者： Srinivas Devadas , Xiangyao Yu
DOI:
关键词: Sequential consistency 、 Cache 、 Computer science 、 Block (data storage) 、 Parallel computing 、 Coherence (statistics) 、 Shared memory 、 Timestamp 、 Distributed shared memory 、 Memory coherence
摘要: A new memory coherence protocol, Tardis, is proposed. Tardis uses timestamp counters representing logical time as well physical to order operations and enforce sequential consistency in any type of shared system. unique that compared the widely-adopted directory its variants, it completely avoids multicasting only requires O(log N) storage per cache block for an N-core system rather than O(N) sharer information. simpler easier reason about, yet achieves similar performance protocols on a wide range benchmarks run 16, 64 256 cores.
harvard.edu
arxiv.org参考文章(34)
Roberto Bisiani, Andreas Nowatzyk, Mosur Ravishankar, Coherent Shared Memory on a Distributed Memory Machine. international conference on parallel processing. pp. 133- 141 ,(1989)
Sang Lyul Min, Jean-Loup Baer, A Timestamp-based Cache Coherence Scheme. international conference on parallel processing. pp. 23- 32 ,(1989)
Srinivas Devadas, Muralidaran Vijayaraghavan, Xiangyao Yu, A Proof of Correctness for the Tardis Cache Coherence Protocol arXiv: Distributed, Parallel, and Cluster Computing. ,(2015)
Henry Hoffmann, David Wentzlaff, Anant Agarwal, Remote Store Programming High Performance Embedded Architectures and Compilers. pp. 3- 17 ,(2010) , 10.1007/978-3-642-11515-8_3
Sandhya Dwarkadas, Alan L. Cox, Willy Zwaenepoel, Pete Keleher, TreadMarks: distributed shared memory on standard workstations and operating systems usenix winter technical conference. pp. 10- 10 ,(1994)
Censier, Feautrier, A New Solution to Coherence Problems in Multicache Systems IEEE Transactions on Computers. ,vol. 27, pp. 1112- 1118 ,(1978) , 10.1109/TC.1978.1675013
Kourosh Gharachorloo, John L. Hennessy, Anoop Gupta, Two Techniques to Enhance the Performance of Memory Consistency Models. international conference on parallel processing. pp. 355- 364 ,(1991)
Yeong-Chang Maa, Dhiraj K. Pradhan, Dominique Thiebaut, Two economical directory schemes for large-scale cache coherent multiprocessors ACM Sigarch Computer Architecture News. ,vol. 19, pp. 10- ,(1991) , 10.1145/379189.379198
I. Singh, A. Shriraman, W. W. L. Fung, M. O'Connor, T. M. Aamodt, Cache coherence for GPU architectures high-performance computer architecture. pp. 578- 590 ,(2013) , 10.1109/HPCA.2013.6522351
Byn Choi, Rakesh Komuravelli, Hyojin Sung, Robert Smolinski, Nima Honarmand, Sarita V. Adve, Vikram S. Adve, Nicholas P. Carter, Ching-Tsun Chou, DeNovo: Rethinking the Memory Hierarchy for Disciplined Parallelism international conference on parallel architectures and compilation techniques. pp. 155- 166 ,(2011) , 10.1109/PACT.2011.21