A comparison study of the HLRC-DU protocol versus a HLRC hardware assisted protocol

SVM systems are a cheaper and flexible way to implement the shared memory programming paradigm. Their huge flexibility is due to their software implementation; however, this is also the main responsible of their performance drawbacks with respect to hardware systems. In this paper we compare a pure software HLRC protocol called the HLRC-DU, versus an improved version of the HLRC protocol that uses hardware support to reduce asynchronous communication. Performances of both protocols are compared over a baseline HLRC protocol. Results show that, by on the half of the benchmarks, our protocol performs better than the hardware approach, even more, in some cases our protocol reaches a speedup higher than 22% with respect to the baseline protocol.

[1]  Paul Hudak,et al.  Memory coherence in shared virtual memory systems , 1986, PODC '86.

[2]  John K. Bennett,et al.  Using multicast and multithreading to reduce communication in software DSM systems , 1998, Proceedings 1998 Fourth International Symposium on High-Performance Computer Architecture.

[3]  Ana Pont,et al.  LIDE: a simulation environment for shared virtual memory systems , 2000, CARN.

[4]  Jaswinder Pal Singh,et al.  Application restructuring and performance portability on shared virtual memory and hardware-coherent multiprocessors , 1997, PPOPP '97.

[5]  Ana Pont,et al.  About the sensitivity of the HLRC-DU protocol on diff and page sizes , 2001, 2001 IEEE International Symposium on Performance Analysis of Systems and Software. ISPASS..

[6]  Michael L. Scott,et al.  The effect of network total order, broadcast, and remote-write capability on network-based shared memory computing , 2000, Proceedings Sixth International Symposium on High-Performance Computer Architecture. HPCA-6 (Cat. No.PR00550).

[7]  J.P. Singh,et al.  Using network interface support to avoid asynchronous protocol processing in shared virtual memory systems , 1999, Proceedings of the 26th International Symposium on Computer Architecture (Cat. No.99CB36367).

[8]  Liviu Iftode,et al.  Performance evaluation of two home-based lazy release consistency protocols for shared virtual memory systems , 1996, OSDI '96.

[9]  Angelos Bilas,et al.  The Effects of Communication Parameters on End Performance of Shared Virtual Memory Clusters , 1997, ACM/IEEE SC 1997 Conference (SC'97).

[10]  Kai Li,et al.  IVY: A Shared Virtual Memory System for Parallel Computing , 1988, ICPP.

[11]  Ricardo Bianchini,et al.  Hiding communication latency and coherence overhead in software DSMs , 1996, ASPLOS VII.

[12]  Leigh Stoller,et al.  Making distributed shared memory simple, yet efficient , 1998, Proceedings Third International Workshop on High-Level Parallel Programming Models and Supportive Environments.

[13]  David Kaeli,et al.  Characterizing the dynamic behavior of workload execution in SVM systems , 2004 .

[14]  Liviu Iftode,et al.  Relaxed consistency and coherence granularity in DSM systems: a performance evaluation , 1997, PPOPP '97.

[15]  Liviu Iftode,et al.  Understanding Application Performance on Shared Virtual Memory Systems , 1996, 23rd Annual International Symposium on Computer Architecture (ISCA'96).

[16]  Keshav Pingali,et al.  Proceedings of the tenth ACM SIGPLAN symposium on Principles and practice of parallel programming , 1997, PPoPP 1997.

[17]  L. Iftode,et al.  Design choices in the SHRIMP system: an empirical study , 1998, Proceedings. 25th Annual International Symposium on Computer Architecture (Cat. No.98CB36235).

[18]  Kai Li,et al.  Understanding Application Performance on Shared Virtual Memory Systems , 1996, 23rd Annual International Symposium on Computer Architecture (ISCA'96).