Team grand-large

Overall Objectives
Scientific Foundations
Application Domains
New Results
Other Grants and Activities

Section: New Results

Large Scale Peer to Peer Performance Evaluations

Participant : Serge Petiton.

Large Scale Grid Computing

Recent progress has made possible to construct high performance distributed computing environments, such as computational grids and cluster of clusters, which provide access to large scale heterogeneous computational resources. Exploration of novel algorithms and evaluation of performance is a strategic research for the future of computational grid scientific computing for many important applications [113] . We adapted [87] an explicit restarted Lanczos algorithm on a world-wide heterogeneous grid platform. This method computes one or few eigenpairs of a large sparse real symmetric matrix. We take the specificities of computational resources into account and deal with communications over the Internet by means of techniques such as out-of-core and data persistence. We also show that a restarted algorithm and the combination of several paradigms of parallelism are interesting in this context. We perform many experimentations using several parameters related to the Lanczos method and the configuration of the platform. Depending on the number of computed Ritz eigenpairs, the results underline how critical the choice of the dimension of the working subspace is. Moreover, the size of platform has to be scaled to the order of the eigenproblem because of communications over the Internet.

High Performance Cluster Computing

Grid computing focuses on making use of a very large amount of resources from a large-scale computing environment. It intends to deliver high-performance computing over distributed platforms for computation and data-intensive applications. We propose [132] an effective parallel hybrid asynchronous method to solve large sparse linear systems by the use of a Grid Computing platform Grid5000. This hybrid method combines a parallel GMRES(m) (Generalized Minimum RESidual) algorithm with the Least Square method that needs some eigenvalues obtained from a parallel Arnoldi algorithm. All of these algorithms run on the different processors of the platform Grid5000. Grid5000, a 5000 CPUs nation-wide infrastructure for research in Grid computing, is designed to provide a scientific tool for computing. We discuss the performances of this hybrid method deployed on Grid5000, and compare these performances with those on the IBM SP series supercomputers.

Large Scale Power aware Computing

Energy conservation is a dynamic topic of research in High Performance Computing and Cluster Computing. Power-aware computing for heterogeneous world-wide Grid is a new track of research. We have studied and evaluated the impact of the heterogeneity of the computing nodes of a Grid platform on the energy consumption. We propose to take advantage of the slack-time caused by the heterogeneity in order to save energy with no significant loss of performance by using Dynamic Voltage Scaling (DVS) in a distributed eigensolver [88] . We show that using DVS only during the slack-time does not penalize the performances but it does not provide significant energy savings. If DVS is applied to all the execution, we get important global and local energy savings (respectively up to 9% and 20%) without a significant rise of the wall-clock times.


Logo Inria