TY - JOUR
T1 - Simulation-based performance prediction for large parallel machines
AU - Zheng, Gengbin
AU - Wilmarth, Terry
AU - Jagadishprasad, Praveen
AU - Kalé, Laxmikant V.
N1 - This work was supported by a National Science Foundation grant NSF NGS #0103645.
PY - 2005/6
Y1 - 2005/6
N2 - We present a performance prediction environment for large scale computers such as the Blue Gene machine. It consists of a parallel simulator, BigSim, for predicting performance of machines with a very large number of processors, and BigNetSim, which incorporates a pluggable module of a detailed contention-based network model. The simulators provide the ability to make performance predictions for very large machines such as Blue Gene/L. We illustrate the utility of our simulators using validation and prediction studies of several applications using smaller numbers of processors for simulations.
AB - We present a performance prediction environment for large scale computers such as the Blue Gene machine. It consists of a parallel simulator, BigSim, for predicting performance of machines with a very large number of processors, and BigNetSim, which incorporates a pluggable module of a detailed contention-based network model. The simulators provide the ability to make performance predictions for very large machines such as Blue Gene/L. We illustrate the utility of our simulators using validation and prediction studies of several applications using smaller numbers of processors for simulations.
KW - Adaptive MPI
KW - CHARMH
KW - Computation modeling
KW - Large parallel machines
KW - Simulation-based performance prediction
UR - http://www.scopus.com/inward/record.url?scp=23944486115&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=23944486115&partnerID=8YFLogxK
U2 - 10.1007/s10766-005-3582-6
DO - 10.1007/s10766-005-3582-6
M3 - Article
AN - SCOPUS:23944486115
SN - 0885-7458
VL - 33
SP - 183
EP - 207
JO - International Journal of Parallel Programming
JF - International Journal of Parallel Programming
IS - 2-3
ER -