Loading content…
Loading content…
The High Performance Computing (HPC) resources at TU Dublin are provided by the Digital Futures Research Hub and operated by the School of Computer Science. These resources support computational research across TU Dublin.
The HPC cluster gives access to significant computational capability beyond personal computing devices. For appropriate research, the resources additionally provide an intermediate step to using the School’s High-Performance Computing (HPC) cluster.
The following resources are available as listed in the current HPC inventory:
| Node | Specifications |
|---|---|
| adapt-master | PowerEdge R730xd, 2 x Intel(R) Xeon(R) CPU E5-2620 v4 @ 2.10GHz (32 cores total), 62GB RAM, Storage: 278.9G,50.9T |
| Node | Specifications |
|---|---|
| ADAPT-01 | PowerEdge R730, NVIDIA GeForce RTX 2080 Ti [11GB], NVIDIA GeForce RTX 2080 Ti [11GB], 2 x Intel(R) Xeon(R) CPU E5-2623 v4 @ 2.60GHz (16 cores total), 503GB RAM, Storage: 3.6T |
| ADAPT-02 | PowerEdge R730, 8 CPU cores, 512GB Ram, 2 x RTX 2080TI |
| ADAPT-03 | PowerEdge R730, NVIDIA GeForce RTX 4070 [11GB], 2 x Intel(R) Xeon(R) CPU E5-2623 v4 @ 2.60GHz (16 cores total), 503GB RAM, Storage: 3.6T |
| ADAPT-CLIN | Precision 7920 Rack, 8 CPU cores, 128 GB Ram, 2 x RTX Quadro 8000 graphics cards |
| AIRC-01 | PowerEdge R740, Tesla K80 [11GB], Tesla K80 [11GB], Tesla K80 [11GB], Tesla K80 [11GB], 2 x Intel(R) Xeon(R) Silver 4110 CPU @ 2.10GHz (32 cores total), 109GB RAM, Storage: 111.3G |
| AIRC-12 | PowerEdge R730, 128 GB Ram, 2 x RTX 2080 TI |
| ml-01 | Precision 7920 Rack, 128 GB Ram, 2 x RTX Quadro 8000 graphics cards |
| ml-12 | NVIDIA RTX A5500 [23GB], NVIDIA RTX A5500 [23GB], 2 x Intel(R) Xeon(R) Gold 6234 CPU @ 3.30GHz (32 cores total), 125GB RAM, Storage: 465.8G |
| ml-13 | PowerEdge Server, 117GB RAM, 16 CPU cores, 2x NVIDIA GPUs |
| socs-7960-01 | Precision 7960 Rack, NVIDIA RTX 6000 Ada Generation [47GB], 2 x Intel(R) Xeon(R) Silver 4410Y (48 cores total), 503GB RAM, Storage: 1.8T |
| socs-7960-02 | Precision 7960 Rack, NVIDIA RTX 6000 Ada Generation [47GB], 2 x Intel(R) Xeon(R) Silver 4410Y (48 cores total), 503GB RAM, Storage: 1.8T |
| socs-7960-13 | Precision 7960 Rack, NVIDIA RTX 6000 Ada Generation [47GB], 2 x Intel(R) Xeon(R) Silver 4410Y (48 cores total), 503GB RAM, Storage: 1.8T |
| socs-7960-14 | Precision 7960 Rack, NVIDIA RTX 6000 Ada Generation [47GB], 2 x Intel(R) Xeon(R) Silver 4410Y (48 cores total), 503GB RAM, Storage: 1.8T |
| Resource | Specifications |
|---|---|
| CPU RACK | 68-Node Rackmount CPU Cluster: Dual Xeon Gold 6148, 192GB RAM, 480GB SSD, Omni-Path Interconnect |
As of March 2025, all the GPU nodes (except AIRC-01 TBD) & G2 GPU nodes are standardized to use:
In nearly all cases you should install your own software environment using Python’s built-in venv module or similar. See the SLURM usage guide for more information.
Each node has access to shared directories under /home.
Previously there was a separate folder /data for your data files. This has now been merged with the home directory structure. Folders that were previously in /data can now be found in your home directory — typically the data is now located in your_home/from_data
For issues with HPC services, a ticketing system will be implemented shortly.