Training on DEISA HPC Infrastructure
Traditionally, scientific groups provide training to their own researchers in an informal manner: the group builds up expertise in their field over time, and this knowledge is naturally transferred to the research staff working in the group.
Unfortunately, this approach is no longer appropriate for world-leading HPC where supercomputers are evolving on a very short timescale (every few years). It is now not possible for researchers to keep up-to-date with all the technological developments in addition to their scientific research. Crucial information on how best to utilise modern supercomputers lies with a small number of HPC experts scattered across the national centres. The activities described in WP5 will coordinate this pool of expertise to target a number of specific applications areas. However, it is in Europe’s long-term interest also to transfer this knowledge to the user community as a whole. This is best achieved through a coordinated training programme.
The training programme is designed directly to complement the Operations and Applications enabling activities; we target current and future users of the DEISA infrastructure and specifically those requiring extreme parallel scaling. The objectives are to enable new DEISA users to learn about the benefits of the novel features of the infrastructure and services as maintained by the Operations team; to enhance the flow of information between the service providers and our users by having face-to-face contact at the training sessions; to make existing users aware of the advanced tools and techniques that they can use to take full advantage of the supercomputing platforms within DEISA; to allow experienced DEISA users to share their experiences via invited presentations.
This section provides introductory training
material covering the DEISA infrastructure and which is also based on the training sessions held in the past.
You find material about advanced training courses on specific topics of direct relevance to users. Topics for advanced training include: Parallel I/O; Advanced Message-Passing Features; Efficient Use of Multi-Core Processors; New Programming Models and Languages.
