TY - GEN
T1 - LogGOPSC
T2 - 2019 IEEE International Conference on Cluster Computing, CLUSTER 2019
AU - Baicheng, Yan
AU - Yi, Zhou
AU - Limin, Xiao
AU - Jiantong, Huo
AU - Zhaokai, Wang
N1 - Publisher Copyright:
© 2019 IEEE.
PY - 2019/9
Y1 - 2019/9
N2 - Benefits from the simplicity, fastness and accuracy, the LogP model family is widely used to predict the parallel application communication performance, especial for the large-scale parallel application prediction or online prediction. However, this type of methods usually lacks consideration of modeling the network contention effect. This hinders their performance in predicting some real-world parallel applications. We thus propose a new parallel computation model via extending the LogGOPS mode with a parameter C. The additional parameter C is the additional time overhead caused by the network contention and it is predicted by a designed BP neural network. The experimental results show that LogGOPSC is more accurate than LogGOPS when there occur network contentions. Compared to the LogGOPS model, LogGOPSC gains a 93.25% average accuracy improvement for predicting 8mb point-to-point message passing. Furthermore, the average error of predicting two communication patterns is as low as 14.50% on the TianHe-2 HPC system.
AB - Benefits from the simplicity, fastness and accuracy, the LogP model family is widely used to predict the parallel application communication performance, especial for the large-scale parallel application prediction or online prediction. However, this type of methods usually lacks consideration of modeling the network contention effect. This hinders their performance in predicting some real-world parallel applications. We thus propose a new parallel computation model via extending the LogGOPS mode with a parameter C. The additional parameter C is the additional time overhead caused by the network contention and it is predicted by a designed BP neural network. The experimental results show that LogGOPSC is more accurate than LogGOPS when there occur network contentions. Compared to the LogGOPS model, LogGOPSC gains a 93.25% average accuracy improvement for predicting 8mb point-to-point message passing. Furthermore, the average error of predicting two communication patterns is as low as 14.50% on the TianHe-2 HPC system.
KW - Communication Performance Prediction
KW - Network Contention
KW - Parallel Application
UR - https://www.scopus.com/pages/publications/85075269148
U2 - 10.1109/CLUSTER.2019.8891035
DO - 10.1109/CLUSTER.2019.8891035
M3 - 会议稿件
AN - SCOPUS:85075269148
T3 - Proceedings - IEEE International Conference on Cluster Computing, ICCC
BT - Proceedings - 2019 IEEE International Conference on Cluster Computing, CLUSTER 2019
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 23 September 2019 through 26 September 2019
ER -