TY - GEN
T1 - On simulating hierarchical clusters for performance of ising spin systems
AU - Muthukrishnan, Gayathri
AU - Santos, Eunice E.
PY - 2004
Y1 - 2004
N2 - In this paper, we discuss the simulation of hierarchical clusters on a homogeneous set of machines, and the efficient implementation of the parallel Ising model on such clusters. A hierarchical cluster can be considered as a cluster of homogeneous nodes which are partitioned into multiple supernodes such that communication across homogenous clusters are represented by a supernode topological network. The results of implementing the algorithm that uses layouts optimized for the cluster are compared with the theoretical results obtained on a modified version of the LogP model that represents such tiered networking, and we show that the simulation of the hierarchy realistically predicts the theoretical results. We show that the model for simulating the hierarchical clusters can be easily extended to simulate many other kinds of networks for performance analysis and also show that the simulation can be used to implement many other data parallel applications on such clusters.
AB - In this paper, we discuss the simulation of hierarchical clusters on a homogeneous set of machines, and the efficient implementation of the parallel Ising model on such clusters. A hierarchical cluster can be considered as a cluster of homogeneous nodes which are partitioned into multiple supernodes such that communication across homogenous clusters are represented by a supernode topological network. The results of implementing the algorithm that uses layouts optimized for the cluster are compared with the theoretical results obtained on a modified version of the LogP model that represents such tiered networking, and we show that the simulation of the hierarchy realistically predicts the theoretical results. We show that the model for simulating the hierarchical clusters can be easily extended to simulate many other kinds of networks for performance analysis and also show that the simulation can be used to implement many other data parallel applications on such clusters.
KW - Hierarchical clusters
KW - Ising model
KW - Parallel & network modeling
KW - Performance prediction & evaluation
KW - Simulation
UR - http://www.scopus.com/inward/record.url?scp=12344287419&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=12344287419&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:12344287419
SN - 1932415262
SN - 9781932415261
T3 - Proceedings of the International Conference on Parallel and Distributed Processing Techniques and Applications, PDPTA'04
SP - 429
EP - 434
BT - Proceedings of the International Conference on Parallel and Distributed Processing Techniques and Applications, PDPTA'04
A2 - Arabnia, H.R.
T2 - Proceedings of the International Conference on Parallel and Distributed Processing Techniques and Applications, PDPTA'04
Y2 - 21 June 2004 through 24 June 2004
ER -