XLG Nodes¶
We have 2 Dell PowerEdge XE7745 nodes with Nvidia Lovelace L40s 48GB cards for GPU jobs.
Accessing the XLG nodes
One XLG node is allocated to the Apini cluster, purchased by Digital Environment Research Institute. The other XLG node is restricted to users within the Science and Engineering (S&E) faculty only.
GPU high memory
The XLG nodes have more RAM available than standard GPU nodes, supporting
up to 240G RAM per GPU (16 tasks per GPU). Authorised users may use the
--mem-per-cpu=15G submission parameter to request larger amounts of RAM
on these nodes.
| XLG | Dell PowerEdge XE7745 |
|---|---|
| Processor | 2 x 64 Core AMD Zen EPYC 9555 |
| Cores/Node | 128 |
| RAM | 2.2TB |
| Accessible RAM | ~2.2TB |
| TMP Size | 5.6TB |
| Interconnect | 100Gb Ethernet |
| GPU | 8 x NVIDIA Lovelace L40s |
| GPU architecture | Lovelace |
| Form Factor | PCIe |
| Tensor Cores | 568 4th Generation |
| CUDA Cores | 18,176 |
| GPU Memory | 48GiB per GPU |
| CUDA Compute | 8.9 (CUDA version 11.8 or greater required) |