compute at leeds. note that Leeds' uses cluster to mean "lab" and research compute when they mean "cluster". (I mean technically, I guess we have clusters of urninals in many gent's bathrooms, I still wouldn't write cluster on the door)
Main compute resources are the local ARC, regional
Resources for
Comparing different clusters is non-trivial. This table just counts the number of GPUs - this is normally what ML-ey graphics-ey people use. We note that server-side GPU is expensive pared to (gamer grade) local (under desk GPUs). Local is also much easier to debug (trying to debug something run with a scheduler is a nightmare). (a list of national HPC facilities).
| institution | cluster | GPU hardware | GPU count |
| York | Viking | 8 x v100 | 8 |
| Ireland/ICHEC | Kay | 32 x v100 | 32 |
| Leeds | ARC 3/4 | 4 x k80, 24 x P100, 12 x V100 | 40 |
| Bede (Tier 2) | NICE-19 | 80 x V100 (IBM Power9) | 80 |
| Cirrus (Tier 2) | Cirrus | 152 x v100 | 152 |
| Bristol | BlueCrystal 3/4 | 76 x k20, 65 x p100 | 141 |
| Jade | v1 | 176 x V100 | 176 |
| Edinburgh | Eddie + James and Charles | eddie ( 44 x k80, 80 x titan X) + james (59 mixed) | 183 |
| Birmingham | Baskerville (July 21) | 184 x a100 | 184 |
| Cambridge (Tier 2) | Wilkes2 | 360 x p100 | 360 |
| KAUST | Ibext | 32 x 2080ti, 32 x 1080ti, 20 x P100, 4 x p6000, 272xv100 | 340 |
| Jade 2 (Tier 2) | v2 | 504 x V100 | 504 |
| UMass | MGHPCC | unspecified mix | 1092 |