Challenges and Opportunities on Parallel/Distributed Programming for Large-scale: From Multi-core to Clouds

Abstract

This talk will (modestly) attend to address challenges and opportunities ahead of us for making parallel programming as efficient, as productive, as effective as possible, on both Multi-cores and large-scale infrastructures such as Grids and Clouds. We draw on our experience at simplifying the programming of applications that are distributed on Local Area Network (LAN), on cluster of workstations, or GRIDs, and of course, Clouds. We will discuss a specific approach, Network On Ship, to cope seamlessly with both distributed and shared-memory multi-core machines. The point will be illustrated with ProActive an Open Source library for parallel, distributed, and concurrent computing, allowing to showcase interactive and graphical GUI and tools. Benchmarks on platforms such as Grid 5000, together with standardization and collaboration with Chinese partners will also be reported.