Platform turns research grid into a 'cloud'

By on
Platform turns research grid into a 'cloud'

Compute Cloud launched for Aussie researchers.

The Australian Research Collaboration Service (ARCS) has launched a new platform to facilitate researchers' access to its national computing grid.

Dubbed the ARCS Compute Cloud, the platform aims to reduce the complexity of using ARC's Australian Grid, which networks several Australian supercomputers.

The grid is available to all publicly funded researchers and links more than 1,000 processors from the Victorian Partnership for Advanced Computing (VPAC), Tasmanian Partnership for Advanced Computing (TPAC), eResearch SA (eRSA) and iVEC in WA.

During the past twelve months, the grid has seen around 500,000 CPU hours of use, completing 138,902 jobs for computational chemists, physicists, and other researchers.

Developed by an in-house team of six over the past nine months, the Compute Cloud provides researchers with a single account for accessing all available grid resources, presenting the Australian Grid to the research community as a single system.

It also automates processes like supercomputer location specification, job configuration, scheduling and completion estimates, and integrates with applications such as MrBayes, POVRay, BLAST, R, Octave, BEAST and Underworld.

ARCS associate director Nick Tate expected the Compute Cloud's simple interface to solve a "common research problem", allowing researchers with minimum technical skills to access sophisticated computing systems.

"Ease of use is important," he told iTnews. "Your main job as a researcher is to research, not to work out how to get to the computing resources you need."

The organisation currently is working on making Compute Cloud code available under the open source Apache 2.0 license, so the platform may be implemented overseas.

In consultation with the research community, ARCS also plans to add more interoperable applications to the Compute Cloud such as bioinformatics application mpiBLAST.

Copyright © . All rights reserved.

Most Read Articles

Log In

|  Forgot your password?