Design for Applications

Design for Applications2016-11-10T13:22:28+00:00

We work closely with commercial ISV’s, open-source and in-house applications, providing benchmarking and ensuring the optimal performance of end-user applications. Staying close to user applications allows our engineers to design cluster systems which are highly customised to specific application requirements, and to tune and optimise software and system performance.

Many of our team have a background in applied scientific research, and practical experience in a wide range of HPC applications. We can offer help and guidance in the following areas of application.

  • Manufacturing/CAE
  • Computational Fluid Dynamics (CFD)
  • Mathematics
  • Chemistry
  • Physics & Astronomy
  • Bio-Informatics
  • Climatology, Geophysics & Exploration
  • Financial Modelling

Designing for Your Applications

The best selection and combination of hardware components always depends on how your particular end-user application code works. Some of the most important considerations are :

  • Does it perform more floating point or more integer operations?
  • Does it use large amounts of data?
  • How is data managed and used within the application?
  • Does it access this data in a predictable or a random fashion?
  • Can the code easily be parallelised?
  • Is it more efficient to use a fine or a coarse grained parallelism?

If you use well-known scientific application codes, the chances are one of our in-house or consultant application specialists have previous experience running those codes on a cluster. We can then advise you on the best cluster design for your applications.

If we are not familiar with your particular application code, but you can tell us how your code works, how you read your input data, which computational algorithms are used, etc., then we can often predict reasonably well what combination of hardware components will work best.

Often we advise to first benchmark your application code on a real cluster. We can then provide you with an account on one of our test clusters, or we can benchmark your code for you.