Trillium
Jump to navigation
Jump to search
| Availability: Spring 2025 |
| Login node: to be determined |
| Globus endpoint: to be determined |
| Data transfer node (rsync, scp, sftp,...): to be determined |
| Portal: to be determined |
Trillium is a large parallel cluster built by Lenovo Canada and hosted by SciNet at the University of Toronto.
Installation and transition[edit]
Due to limits on available power and cooling capacity there will be an interim period in which a significant portion of the old Niagara will be shut down in order to provide power for the new system's acceptance testing and transition. We'll update you when we have a better idea of Trillium's installation schedule.
Storage[edit]
Parallel storage: 29 petabytes, NVMe SSD based storage from VAST Data.
High-performance network[edit]
- Nvidia “NDR” Infiniband network
- 400 Gbit/s network bandwidth for CPU nodes
- 800 Gbit/s network bandwidth for GPU nodes
- Fully non-blocking, meaning every node can talk to every other node at full bandwidth simultaneously.
Node characteristics[edit]
| nodes | cores | available memory | CPU | GPU |
|---|---|---|---|---|
| 1224 | 192 | 768GB DDR5 | 2 x AMD EPYC 9655 (Zen 5) @ 2.6 GHz, 384MB cache L3 | |
| 60 | 96 | 768GB DDR5 | 1 x AMD EPYC 9654 (Zen 4) @ 2.4 GHz, 384MB cache L3 | 4 x NVidia H100 SXM (80 GB memory) |