General Information

Overview

The Research Computing Services section is tasked with supporting those ICT requirements of researchers across the University which are beyond what is expected to be provided by their faculty. It does this by providing a range of physical resources together with advice, training and consulting in the most effective means of using these resources to further their research needs. Where this results in changed methods or major productivity improvements this may be referred to as "eResearch".

Services Provided

The physical resources managed by the group made available as a number of services. These are grouped as Computing Services, Data Services and Visualization Services. We also offer Grid services, but these are based on making use of the other services in a different context, rather than being physically distinct, although some of the physical resources are labelled with the name Grid. A full list of services, who can access them, and their availability can be found at http://www.hpc.unimelb.edu.au/RCS-services.html.

Computing Services

Our current services are based on a number of cluster systems. These consist of a collection of single or dual cpu computers which are managed as a single unit. They have a "head node" which users can log onto and prepare jobs which are then submitted to a queueing system which allocates them to an available "worker node" which then runs the jobs. The main differences between the systems is how the nodes are allocated, although they also differ in the actual hardware. In addition, we have a small number of different machines for testing and development purposes.

Data Services

We provide two distinct data services. For users of our computing services, we have a single large data storage node (~10TB) which is shared across all our clusters, and in addition we have 2 dedicated Grid storage nodes where we can store sharable/shared data for use in collaborative projects using Grid based authentication and authorization to control access.

Visualization Services

In contrast to the previous services which are accessed remotely, for visualization we have a small lab (the Computer Visualization Facility or CVF) where users can come and use workstations with the latest graphics hardware and software. Supporting this we have a small renderfarm for animation jobs and can access additional resources for large rendering jobs. We also have an AccessGrid node, which is designed to support collaborative video-conferencing, but whose display capabilities, including passive stereo projection and multi-screen projection, can be used for visualization.

Grid Services

Grid Computing and DataGrid are emerging paradigms for consolidating disparate resources to support distributed collaborative research. They are based on a layer of "Middleware" (software) which shares a common authentication and authorization structure and allows location independent operations such as reading and writing files. This software is not really "production ready" but we are working with those groups who are already using it in order to bring it into production.