Prediction of communication latency over complex network behaviors on SMP clusters

Maxime Martinasso (1) , Jean-François Méhaut (1)

1 APACHE - Parallel algorithms and load sharing

Using MPI as communication interface, one or several applications may introduce complex communication behaviors over the network of a cluster. This effect is increased when nodes of the cluster are multi-processors, and where communications can income or outgo from the same node with a common interval time. Our interest is to understand the effects of complex communication schemes over a network of SMP nodes, and then to identify different network conflicts. Network conflicts stand for a network component (links, NIC, or MPI stack, etc.) shared between communications. This paper describes a set of experiments generating network conflicts and their analysis, in order to accurately predict communication latencies. This analysis is based on cluster of bi-processor with a Myrinet network, using as communication interface a LAM MPI-2 implementation over GM protocol.

Domaines: Autre [cs.OH]

Maxime Martinasso, Jean-François Méhaut. Prediction of communication latency over complex network behaviors on SMP clusters. EPEW'05/WS-FM'05 Proceedings of the 2005 international conference on European Performance Engineering, and Web Services and Formal Methods, international conference on Formal Techniques for Computer Systems and Business Processes, 2005, Versailles, France. ⟨10.1007/11549970_13⟩. ⟨hal-00687244⟩