A novel strategy for building interoperable MPI environment in heterogeneous high performance systems

Abstract

Breakthrough advances in microprocessor technology and efficient power management have altered the course of development of processors with the emergence of multi-core processor technology, in order to bring higher level of processing. The utilization of many-core technology has boosted computing power provided by cluster of workstations or SMPs, providing large computational power at an affordable cost using solely commodity components. Different implementations of message-passing libraries and system softwares (including Operating Systems) are installed in such cluster and multi-cluster computing systems. In order to guarantee correct execution of message-passing parallel applications in a computing environment other than that originally the parallel application was developed, review of the application code is needed. In this paper, a hybrid communication interfacing strategy is proposed, to execute a parallel application in a group of computing nodes belonging to different clusters or multi-clusters (computing systems may be running different operating systems and MPI implementations), interconnected with public or private IP addresses, and responding interchangeably to user execution requests. Experimental results demonstrate the feasibility of this proposed strategy and its effectiveness, through the execution of benchmarking parallel applications.

MPI-Based Heterogeneous Cluster Construction Technology

“…These fostered the development of a parallel software industry, and encouraged development of portable and scalable large-scale parallel applications for HPC programming [1,2]. Since it is easy to acquire the state of resources and processes in MPI, for solving the heterogeneity issues in cluster implementation, the MPI becomes a natural choice on heterogeneous platform [3]. However, popular implementations of MPI most likely support homogenous clusters rather than heterogeneous one.…”

Técnicas de programação e avaliação de desempenho de solvers de sistemas de equações lineares em sistemas computacionais de alto desempenho.

“…Globalmente, precisa ser interpretado como um ambiente MIMD de memória distribuída, já que cada nó possui seu próprio banco de memória. Dessa forma, um melhor desempenho será extraído do sistema se for pensado e programado como um ambiente híbrido, similar ao mencionado em [12].…”