iCSF - interactive Computational Shared Facility (aka Incline)

Current System Configuration

High-memory Nodes (256 GB RAM) – recommended for most users
Six nodes with 256 GB RAM and 2 x 14-core Intel Broadwell CPU. Running Scientific Linux 7.9. Access via

incline256.itservices.manchester.ac.uk

If you receive a security key error on login, please follow these instructions.

Standard Nodes (64 GB RAM) – older nodes
There are eight standard nodes, each containing 64 GB RAM and 2 x 6-core Intel Xeon Sandybridge CPUs. Running Scientific Linux 7.9. Access via

incline.itservices.manchester.ac.uk

Please note that these nodes run an older version of Linux (Scientific Linux 6) and so it is not possible to update some application on these nodes (e.g., R-Studio). We recommend using the high-memory (256GB) nodes by default now – they are running a newer version of Linux (Scientific Linux 7) and so have newer version of applications available.

Super-high-memory Node (2 TB RAM).
One node with 2 TB (~ 2000 GB) RAM! 4 x 10-core Intel Xeon Westmere EX CPUs (hyper-threading enabled giving 80 virtual cores). Running Scientific Linux 7.9.

  • Priority use of this node is given to the contributor, but “spare” time is available to all iCSF contributors and users.
incline2000.itservices.manchester.ac.uk
GPus
The iCSF has no GPU hardware. If you need GPUs you will need to use the CSF3 which has Nvidia V100s and A100s and a range of popular GPU/Machine Learning/Deep Learning software.

Last modified on June 15, 2022 at 3:25 pm by Pen Richardson