As datasets increase in size the need for more memory in your available compute hardware becomes more pressing. Applications such as R, Stata, python and MATLAB are often used to process these datasets and can require significant amounts of memory – to load the dataset, process it and generate results.
Welcome to the first instalment in a series of “Hints and Tips” from our expert research software engineer and research infrastructure engineers with the aim of making your research that little bit easier to do! In this article Ian Cottam looks at how to transfer your files between the various Research IT computational resources if you use Apple macOS or Linux.
Are you a Humanities researcher? Would you like to be able to repeat a simple computing task several times but don’t want to use your own machine to do this? By using the UoM computational shared facilities (CSF) you can offload your jobs to the high performance computing (HPC) facility. This will allow you to save space and resources on your own machine to work on something else.
Find out how a research group in the School of MACE are using our HPC resources in conjunction with those at The Hartree Centre, to help Unilever tackle business challenges and to develop new and better products.
We have recently made some significant changes to our computationally intensive research (CIR) platforms, DPSF and CSF2. Find out what has changed and how this will affect your research!
The CSF (Computational Shared Facility) and DPSF (Data Processing Shared Facility), the University’s flagship HPC systems for compute and high memory work, have been steadily growing over the past 7 years. Approximately 4 million pounds have been invested resulting in over 10,000 CPU cores, over 1PB of scratch and 60TB of RAM being available.
The Computational Shared Facility (CSF) has now officially come of age – it has reached an amazing 10,000 CPU cores – with around 4000 cores of Haswell / Broadwell and a Skylake procurement coming up in January.