Experiments with a Software Component Enabling NetSolve with Direct Communications in a Non-Intrusive and Incremental Way

The paper presents a software component that enables NetSolve with direct communications between servers in a non-intrusive and incremental way. Non-intrusiveness means that the software component is supplementary, working on top of the original system, which does not change at all. Increment means that the software component does not have to be installed on all computers to enable applications with the new feature. It can be done incrementally, step by step, and the new feature can be enabled in part, with the completeness dependent on how many nodes have been upgraded with the software component. The paper describes the design and implementation of the software component. The paper also reports on experiments with three typical scientific NetSolve applications having different communication structures: (i) protein tertiary structure prediction, (ii) image processing using sequential algorithms, and (iii) the matrix chain product. The presented experimental results show that the performance of these grid applications can be easily and significantly improved by using the proposed supplementary software component.

[1]  Tomoyuki Hiroyasu,et al.  Parallel Simulated Annealing using Genetic Crossover , 2000 .

[2]  Peng Zhao,et al.  A Non-intrusive and Incremental Approach to Enabling Direct Communications in RPC-Based Grid Programming Systems , 2006, International Conference on Computational Science.

[3]  Henri Casanova,et al.  NetSovle: A Network Server for Solving Computational Science Problems , 1996, Proceedings of the 1996 ACM/IEEE Conference on Supercomputing.

[4]  Jack Dongarra,et al.  Users' Guide to NetSolve v1.4.1 , 2002 .

[5]  Alexey L. Lastovetsky,et al.  SmartNetSolve: high-level programming system for high performance grid computing , 2006, Proceedings 20th IEEE International Parallel & Distributed Processing Symposium.

[6]  Emmanuel Jeannot,et al.  Improving the GridRPC model with data persistence and redistribution , 2004, Third International Symposium on Parallel and Distributed Computing/Third International Workshop on Algorithms, Models and Tools for Parallel Computing on Heterogeneous Networks.

[7]  Satoshi Matsuoka,et al.  Ninf-G: A Reference Implementation of RPC-based Programming Middleware for Grid Computing , 2003, Journal of Grid Computing.

[8]  Keqin Li.  Fast and scalable parallel algorithms for matrix chain product and matrix powers on distributed memory systems , 2001, Proceedings 15th International Parallel and Distributed Processing Symposium. IPDPS 2001.

[9]  L.P. Cordella,et al.  Mapping schemes for sequential image processing algorithms , 1995, Proceedings of Conference on Computer Architectures for Machine Perception.

[10]  Francine Berman,et al.  Middleware for the use of storage in communication , 2002, Parallel Comput..

[11]  Henri Casanova,et al.  Overview of GridRPC: A Remote Procedure Call API for Grid Computing , 2002, GRID.

[12]  Linda M. Wills,et al.  Exposing data-level parallelism in sequential image processing algorithms , 2002, Ninth Working Conference on Reverse Engineering, 2002. Proceedings..

[13]  Y. Okamoto,et al.  Implementation of protein tertiary structure prediction system with NetSolve , 2004, Proceedings. Seventh International Conference on High Performance Computing and Grid in Asia Pacific Region, 2004..

[14]  M.I.T. Press,et al.  The International Journal of Supercomputer Applications and High Performance Computing— , 1994 .

[15]  Henri Casanova,et al.  Innovations of the NetSolve Grid Computing System , 2002, Concurr. Comput. Pract. Exp..

[16]  Keqin Li,et al.  Fast and Scalable Parallel Algorithms for Matrix Chain Product and Matrix Powers on Reconfigurable Pipelined Optical Buses , 2002, J. Inf. Sci. Eng..

[17]  Henri Casanova,et al.  Netsolve: a Network-Enabled Server for Solving Computational Science Problems , 1997, Int. J. High Perform. Comput. Appl..