Climateprediction.net: Design Principles for Public-resource Modeling Research

D. Stainforth, J. Kettleborough, A. Martin, A. Simpson, R. Gillis, A. Akkas, R. Gault, M. Collins, D. Gavaghan, and M. Allen (UK)

Keywords

Grid Computing, Modeling and Simulation, Internet Computing, Parallel and Distributed Systems.

Abstract

Climateprediction.net aims to harness the spare CPU cycles of a million individual users’ PCs to run a massive ensemble of climate simulations using an up-to-date, full resolution, three dimensional atmosphere-ocean climate model. The project has many similarities with other public-resource high-throughput activities but is distinctive in a number of ways including the complexity of the computational task, its duration and system demands, participant interaction, and data volume and analysis procedures. These distinctive aspects, together with the documented experience of other Internet-based distributed computing projects, lead to a number of design challenges. The analysis of these, and the resulting architecture, is presented here. The initial development and testing phase is complete and roll-out will begin shortly. It is expected that the issues raised will be of relevance to an increasing number of similar projects.

Important Links:



Go Back