Fabrice Huet

A High Performance Java Middleware with a Real Application

Previous experiments with high-performance Java were initially disappointing. After several years of optimization, this paper investigates the current suitability of such object-oriented middleware for High-Performance and Grid programming. Using a middleware offering high level abstractions (ProActive), we have replaced the standard Java RMI layer with the optimized Ibis RMI interface. Ibis is a grid programming environment featuring efficient communications. Using a 3D electromagnetic application (an object-oriented time domain finite volume solver for 3D Maxwell equations) we have first conducted benchmarks on single clusters, including comparisons with the same application in Fortran MPI. Finally, Grid experiments have been conducted simultaneously on up to 5 different clusters. Overall, the paper reports extremely promising results. For instance, a speed up of 12 on 16 machines (vs. 13.8 for Fortran), a speedup of 100 on 150 machines on a Grid.

An Efficient Framework for Running Applications on Clusters, Grids, and Clouds

Since the appearance of distributed computing technology, there has been a significant effort in designing and building the infrastructure needed to tackle the challenges raised by complex scientific applications that require massive computational resources. This increases the awareness to harness the power and flexibility of Clouds that have recently emerged as an alternative to data centers or private clusters. We describe in this chapter an efficient high-level Grid and Cloud framework that allows a smooth transition from clusters and Grids to Clouds. The main lever is the ability to move application infrastructure-specific information away from the code and manage them in a deployment file. An application can thus easily run on a cluster, a grid, or a cloud, or any mix of them without modification.

An adaptative mechanism for communicating with mobile objects

As middlewares are more and more spanning over larger networks, towards world-wide grid computing, communicating with mobile objects in a scalable and reliable manner becomes an important and difficult issue. This paper proposes a parametric protocol for such communications. Based on Time To Live (TTL) forwarders, and Time To Update (TTU) mobile objects, it allows to get the best of those two well-known techniques (forwarders, servers). As a parametric protocol, it can be adapted to various network conditions.