Analysis of remote execution models for grid middleware

ABSTRACT

Grid computing applications, and distributed applications in general, often experience performance deterioration due to the latencies inherent in the execution of remote operations. Here we analyze three approaches for reducing latencies: an asynchronous model which executes operations in a thread to hide the remote latency of an operation, a bulk model which bundles multiple operations together in a single remote operation, and a pipelining model which executes remote operations in a pipeline-parallel mode. We analyze the performance, parameters and technical requirements of each model, and identify general properties which can help determine which model is the most suitable.Our results show that, depending on the use case scenario, any of the three models can offer the best performance, and we conclude by presenting a hybrid model that combines all three approaches, potentially providing the benefit of each.

REFERENCES

• 1: William Allcock , John Bresnahan , Rajkumar Kettimuthu , Joseph Link, The Globus eXtensible Input/Output System (XIO): A Protocol Independent IO System for the Grid, Proceedings of the 19th IEEE International Parallel and Distributed Processing Symposium (IPDPS'05) - Workshop 4, p.179.1, April 04-08, 2005 [doi>10.1109/IPDPS.2005.429]
• 2: R. Badia, J. Labarta, R. Sirvent, J. Pérez, J. Cela, and R. Grima. Programming Grid Applications with GRID Superscalar. Journal of Grid Computing, 1(2):151--170, 2003.
• 3: Rüdiger Berlich , Marcel Kunze , Kilian Schwarz, Grid computing in Europe: from research to deployment, Proceedings of the 2005 Australasian workshop on Grid computing and e-research, p.21-27, January 01, 2005, Newcastle, New South Wales, Australia
• 4: T. Goodale, S. Jha, H. Kaiser, T. Kielmann, P. Kleijer, G. von Laszewski, C. Lee, A. Merzky, H. Rajic, and J. Shalf. SAGA: A Simple API for Grid Applications - High-Level Application Programming on the Grid. Computational Methods in Science and Technology: special issue "Grid Applications: New Challenges for Computational Methods", 8(2), 2005.
• 5: S. Hirmer. Enabling bulk operations in dynamically changing grid environments. Master's thesis, University of Karlsruhe (TH), August 2006.
• 6: S. Hirmer, H. Kaiser, A. Merzky, A. Hutanu, and G. Allen. Seamless Integration of Generic Bulk Operations in Grid Applications. In Accepted for GADA '06: Proceedings of the 2006 International Symposium on Grid computing, high-performance and Distributed Applications, August 2006.
• 7: A. Merzky and S. Jha. A Requirements Analysis for a Simple API for Grid Applications. Technical report, Global Grid Forum, May 2006. GFD.71.
• 8: A. Merzky and S. Jha. Simple API for Grid Applications - Use Case Document. Technical report, Global Grid Forum, March 2006. GFD.70.
• 9: Jim Waldo , Geoff Wyant , Ann Wollrath , Samuel C. Kendall, A Note on Distributed Computing, Selected Presentations and Invited Papers Second International Workshop on Mobile Object Systems - Towards the Programmable Internet, p.49-64, July 08-09, 1996
• 10: E. Walker, R. Floyd, and P. Neves. Asynchronous remote operation execution in distributed systems. Proceedings of the 10th International Conference on Distributed Computing Systems Distributed Computing Systems, pages 253--259, 1990.

INDEX TERMS

Primary Classification:
  C. Computer Systems Organization
  C.1 PROCESSOR ARCHITECTURES
      C.1.4 Parallel Architectures
          Subjects: Distributed architectures

Additional Classification:
  C. Computer Systems Organization
  C.2 COMPUTER-COMMUNICATION NETWORKS
      C.2.4 Distributed Systems
          Subjects: Distributed applications
  C.4 PERFORMANCE OF SYSTEMS
      Subjects: Modeling techniques

General Terms:
Performance