Run-Time Support for Adaptive Load Balancing

作者: Milind A. Bhandarkar , Robert K. Brunner , Laxmikant V. Kalé

DOI: 10.1007/3-540-45591-4_159

关键词:

摘要: Many parallel scientific applications have dynamic and irregular computational structure. However, most such exhibit persistence of load communication This allows us to embed measurement-based automatic balancing framework in run-time systems languages that are used build applications. In this paper, we describe a built for the Converse [4] teroperable runtime system. is composed mechanisms recording application performance data, mechanism object migration, interfaces plug-in strategy objects. terfaces objects allow easy implementation novel strategies could use characteristics on entire machine, or only local neighborhood. We present few synthetic benchmark also impact an actual application.

参考文章(7)
Raymond Namyst, Jean-François Méhaut, PM2: Parallel Multithreaded Machine. A Computing Environment for Distributed Architectures. parallel computing. pp. 279- 285 ,(1995)
Jim Pruyne, Miron Livny, Parallel Processing on Dynamic Resources with CARMI job scheduling strategies for parallel processing. pp. 259- 278 ,(1995) , 10.1007/3-540-60153-8_33
L.V. Kale, M. Bhandarkar, N. Jagathesan, S. Krishnan, J. Yelon, Converse: an interoperable framework for parallel programming international conference on parallel processing. pp. 212- 217 ,(1996) , 10.1109/IPPS.1996.508060
Jos'e Nagib, Jos'e Nagib Cotrim, Adam Beguelin, Bruce Lowekamp, Erik Seligman, Mike Starkey, Peter Stephan, Dome: Parallel Programming in a Heterogeneous Multi-User Environment Carnegie Mellon University. ,(1995) , 10.21236/ADA295491
George Karypis, Vipin Kumar, A Fast and High Quality Multilevel Scheme for Partitioning Irregular Graphs SIAM Journal on Scientific Computing. ,vol. 20, pp. 359- 392 ,(1998) , 10.1137/S1064827595287997
Nick Doulas, Balkrishna Ramkumar, Efficient Task Migration for Message Driven Parallel Execution of Nonshared Memory Architectures international conference on parallel processing. ,vol. 2, pp. 170- 173 ,(1994) , 10.1109/ICPP.1994.104
R.K. Brunner, L.V. Kale, Adapting to load on workstation clusters symposium on frontiers of massively parallel computation. pp. 106- 112 ,(1999) , 10.1109/FMPC.1999.750590