DocumentCode
2490192
Title
Performance evaluation of cache depot on CC-NUMA multiprocessors
Author
Hsiao, Hung-Chang ; King, Chung-Ta
Author_Institution
Dept. of Comput. Sci., Nat. Tsing Hua Univ., Hsinchu, Taiwan
fYear
1998
fDate
14-16 Dec 1998
Firstpage
519
Lastpage
526
Abstract
Cache depot is a performance enhancement technique on cache-coherent non-uniform memory access (CC-NUMA) multiprocessors, in which nodes in the system store extra memory blocks on behalf of other nodes. In this way memory requests from a node can be satisfied by nearby depot nodes without going all the way to the home node. This not only reduces memory access latency and network traffic, but also spreads the network load more evenly. We study the design strategy for cache depot that: enhances the network interface of each node to include a depot cache, which stores those extra memory blocks for other nodes; and employs a new multicast routing scheme, which is called the multi-hop worms and works cooperatively with depot caches, to transmit coherence messages. By considering message routing and depot caches together the design concept can be applied even to those CC-NUMA systems that have a non-hierarchical, scalable interconnection network. We have developed an execution-driven simulator to evaluate the effectiveness of the design strategy. Performance results from using four SPLASH-2 benchmarks show that the design strategy improves the performance of the CC-NUMA multiprocessor by 11% to 21%. We have also studied in depth various factors which affect the performance of cache depot
Keywords
cache storage; data integrity; distributed shared memory systems; message passing; multicast communication; performance evaluation; storage management; CC-NUMA multiprocessors; SPLASH-2 benchmarks; cache depot; cache-coherent non-uniform memory access; coherence messages; execution-driven simulator; memory access latency; memory blocks; message routing; multi-hop worms; multicast routing scheme; network interface; network load; network traffic; performance enhancement technique; performance evaluation; scalable interconnection network; Computer science; Computer worms; Delay; Electrical capacitance tomography; Multiprocessor interconnection networks; Read only memory; Routing; Telecommunication traffic;
fLanguage
English
Publisher
ieee
Conference_Titel
Parallel and Distributed Systems, 1998. Proceedings. 1998 International Conference on
Conference_Location
Tainan
ISSN
1521-9097
Print_ISBN
0-8186-8603-0
Type
conf
DOI
10.1109/ICPADS.1998.741127
Filename
741127
Link To Document