In this webinar, NeSI's Alexander Pletzer will walk through some of the key questions that can help determine whether GPUs are among the best tool(s) for your project's needs. These questions include:

• what approaches are currently available to offload computations to GPU hardware
• what speedup can be realistically expected compared to a CPU
• what code features could be used to determine whether a program could benefit from running on a GPU

He'll also share examples from NeSI's own discovery process in this area and the insights we gained from our pre-launch testing of the A100s. Working closely with researchers who contributed their code to help us test, we evaluated the A100s' performance for a broad range of applications, including machine learning to a finite difference code, a quasi-equilibrium tropical circulation model, and an application from marketing research.

You can expect to leave the webinar with a better understanding of how GPUs can benefit your research.

Background

GPU computing is the use of a GPU (graphics processing unit) as a co-processor to accelerate CPUs for general-purpose scientific and engineering computing. They promise to significantly accelerate compute-intensive research and migrating code from CPU to GPUs can often bring huge rewards.

NeSI recently launched new NVIDIA A100 GPUs for researchers using machine learning applications. These new resources complement NeSI's existing set of Pascal P100 GPUs (which are available to any project with a Mahuika allocation).

Who Should Attend

Postgraduates, established researchers, support staff, and anyone in between are encouraged to attend, particularly if they are interested in using advanced GPU resources to support their work.

More Information

If you have any questions or would like more information about this session, please email training@nesi.org.nz.

Additionally, consider subscribing to the NeSI training mailing list to stay updated about future training opportunities.