Download:
|
by Eyal De Lara, Y. Charlie Hu, Honghui Lu, Alan L. Cox, Willy Zwaenepoel
http://www.cs.rice.edu/~willy/papers/lcr00b.ps.gz
Add To MetaCart
Abstract:
Abstract. In this paper, we examine the causes and eects of contention for shared data access in parallel programs running on a software distributed shared memory (DSM) system. Specically, we experiment on two widely-used, page-based protocols, Princeton's home-based lazy release consistency (HLRC) and TreadMarks. For most of our programs, these protocols were equally aected by latency increases caused by contention and achieved similar performance. Where they dier signicantly, HLRC's ability to manually eliminate load imbalance was the largest factor accounting for the dierence. To quantify the eects of contention we either modied the application to eliminate the cause of the contention or modied the underlying protocol to eciently handle it. Overall, we nd that contention has profound eects on performance: eliminating contention reduced execution time by 64 % in the most extreme case, even at the relatively modest scale of 32 nodes that we consider in this paper. 1
Citations
|
705
|
SPLASH: Stanford Parallel Applications for Shared Memory
– Singh, Weber, et al.
- 1992
|
|
637
|
Memory consistency and event ordering in scalable shared-memory multiprocessors
– Gharachorloo, Lenoski, et al.
- 1990
|
|
477
|
TreadMarks: Distributed shared memory on standard workstations and operating systems
– Keleher, Dwarkadas, et al.
- 1994
|
|
462
|
The NAS Parallel Benchmarks
– Bailey, Barton, et al.
- 1991
|
|
422
|
Lazy release consistency for software distributed shared memory
– Keleher, Cox, et al.
- 1992
|
|
202
|
Shasta: A Low Overhead, Software-Only Approach for Supporting Fine-Grain Shared Memory
– Scales, Gharachorloo, et al.
- 1996
|
|
137
|
Performance evaluation of two home-based lazy release consistency protocols for shared memory virtual memory systems
– Zhou, Iftode, et al.
- 1996
|
|
118
|
CASHMERE2L: Software Coherent Shared Memory on a Clustered Remote-Write Network
– Stets, Dwarkadas, et al.
- 1997
|
|
105
|
A unified formalization of four shared-memory models
– Adve, Aggarwal
- 1993
|
|
63
|
Performance evaluation of the orca shared object system
– Bal, Bhoedjang, et al.
- 1998
|
|
45
|
Home-based svm protocols for smp clusters: design and performance
– Samanta, Bilas, et al.
- 1998
|
