Design Issues of Prefetching Strategies for Heterogeneous Software DSM

In Distributed Shared Memory (DSM) systems, one of the major overheads is page fault, because it induces remote memory access latencies. Prefetching strategy permits home nodes sending data to remote nodes in advance, in order to reduce page faults and latencies, as also to improve overall performance; however, it causes accumulated waiting phenomenon, waiting synchronization phenomenon, among others. In heterogeneous environments, considering the fact of different ability among computing hosts, it increases idle time and synchronization time. Therefore, the search for alternatives to reduce these overheads is important. In this paper, we propose a method that utilizes concepts of load balancing in prefetching strategy, where the goal is to reduce idle time and synchronization time. Basically, this approach distributes home pages to appropriate computing nodes and adjusts prefetching strategy’s policies into heterogeneous environments.

[1]  Martin K. Purvis,et al.  Homeless and Home-based Lazy Release Consistency Protocols on Distributed Shared Memory , 2004, ACSC.

[2]  Hu Weiwu,et al.  Dynamic data prefetching in home-based software DSMs , 2001 .

[3]  Kuan-Ching Li,et al.  On the Design and Implementation of an Effective Prefetch Strategy for DSM Systems , 2006, The Journal of Supercomputing.

[4]  Ming Li,et al.  A lock-based cache coherence protocol for scope consistency , 1998, Journal of Computer Science and Technology.

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

[6]  S. Okamoto,et al.  A moving home-based software DSM system , 2003, 2003 IEEE Pacific Rim Conference on Communications Computers and Signal Processing (PACRIM 2003) (Cat. No.03CH37490).

[7]  Ce-Kuen Shieh,et al.  Involving memory resource consideration into workload distribution for software DSM systems , 2004, IEEE International Symposium on Cluster Computing and the Grid, 2004. CCGrid 2004..

[8]  W. Hu,et al.  JIA-JIA : An SVM System Based on A New Cache Coherence Protocol , 1999 .

[9]  Kuan-Ching Li,et al.  On Design of a Prefetching Strategy for DSM System , 2004, International Conference on Parallel and Distributed Processing Techniques and Applications.

[10]  Cho-Li Wang,et al.  Migrating-Home Protocol for Software Distributed Shared Memory , 2002, J. Inf. Sci. Eng..

[11]  Kuan-Ching Li,et al.  On design of agent home scheme for prefetching strategy in DSM systems , 2005, 19th International Conference on Advanced Information Networking and Applications (AINA'05) Volume 1 (AINA papers).

[12]  Weiwu Hu,et al.  A comparison of two strategies of dynamic data prefetching in software DSM , 2001, Proceedings 15th International Parallel and Distributed Processing Symposium. IPDPS 2001.

[13]  Liviu Iftode,et al.  Scope Consistency: A Bridge between Release Consistency and Entry Consistency , 1996, SPAA '96.

[14]  Cho-Li Wang,et al.  A Migrating-Home Protocol for Implementing Scope Consistency Model on a Cluster of Workstations , 1999, PDPTA.

[15]  Kyu Ho Park,et al.  Moving home-based lazy release consistency for shared virtual memory systems , 1999, Proceedings of the 1999 International Conference on Parallel Processing.

[16]  Weisong Shi,et al.  Optimizing Home-Based Software DSM Protocols , 2004, Cluster Computing.

[17]  Alan L. Cox,et al.  Lazy release consistency for software distributed shared memory , 1992, ISCA '92.

[18]  Cho-Li Wang,et al.  A novel adaptive home migration protocol in home-based DSM , 2004, 2004 IEEE International Conference on Cluster Computing (IEEE Cat. No.04EX935).

[19]  Joonwon Lee,et al.  An efficient lock protocol for home-based lazy release consistency , 2001, Proceedings First IEEE/ACM International Symposium on Cluster Computing and the Grid.