Supercomputers/Systems/Wilkes-3
United Kingdom · operational · mixed
Wilkes-3
Operated by University of Cambridge at University of Cambridge (CSD3) .
Phase 1 seed dataset, compiled by hand. These rows were built from public operator, laboratory and vendor sources. A mechanical second-reader pass has since fetched every cited source: 1092 of 1153 systems have a readable citation that names them, and 40 are genuinely weakly sourced. Every claim carries its source and a confidence tier. Treat anything below verified as a lead, not a citation.
Measured performance
- Rmax
- 2.29 PFlop/s
- Rpeak
- 6.64 PFlop/s
- Rmax ÷ Rpeak
- 34.5%
- Cores
- 26,880
- Power
- 74.26 kW
- Per watt
- 30.8 GF/W
Figures are as last publicly reported for the configuration described below, not a live measurement. Where a system was upgraded in place, the post-upgrade configuration is shown and the earlier one appears in the timeline.
What this machine is made of
One row per supplier relationship. “Supplier at build” is the company that shipped the part at the time; where that company has since been acquired, the parent it rolls up to today is shown beside it. That distinction is what makes ticker-level aggregation possible across a thirty-year dataset.
| Role | Part | Supplier at build | Quantity | Confidence | Source |
|---|---|---|---|---|---|
| Integrator | - Dell PowerEdge XE8545 | Dell Technologies | - | Reported | top500.orgdocs.hpc.cam.ac.uk |
| Accelerator | - NVIDIA A100-SXM-80GB, 4 per node over 90 nodes per Cambridge documentation | NVIDIA | 360 reported | Reported | docs.hpc.cam.ac.uktop500.org |
| CPU | AMD EPYC 7763 (Milan) CPUs x86-64 · 64C · 2.45 GHz · 7 nm · 280 W AMD EPYC 7763 64C 2.45GHz (Cambridge docs: 2x per node) | AMD | - | Reported | top500.orgdocs.hpc.cam.ac.uk |
| Interconnect | NVIDIA InfiniBand HDR Interconnect Fat tree or dragonfly+ · 200 Gb/s per port · About 0.6 microseconds Dual-rail Mellanox HDR200 InfiniBand | NVIDIA | - | Reported | docs.hpc.cam.ac.uktop500.org |
The read
The University of Cambridge's A100 GPU service within CSD3, a Dell PowerEdge XE8545 cluster that placed third on the June 2021 Green500 at 28.14 GFlops/Watt. TOP500 lists 26,880 cores, 2.29 PFlop/s Rmax, 6.64 PFlop/s Rpeak and 74.26 kW (AMD EPYC 7763, A100 80GB, dual-rail HDR200 InfiniBand). Cambridge's documentation currently describes 90 nodes each with two EPYC 7763 and four A100-SXM-80GB on Rocky Linux 8; TOP500's benchmarked configuration may differ from that present node count.
Timeline
Public list appearances
Pointers only: rank and edition, linking to the canonical entry. We do not reproduce list tables. See sourcing policy.
Site & facility
University of Cambridge (CSD3)
- Location
- Cambridge, United Kingdom
- Commissioned
- 2024
Change history
- 2026-10-02 Added Wilkes-3, Cambridge's A100 service, 3rd on the June 2021 Green500
Source check
We fetched this system's own citations and recorded whether each page actually mentions it. This is a corroboration signal, not a fact check, and it is published so you can see how well the sourcing holds up rather than take it on trust.
| Cited source | Result | Found on the page |
|---|---|---|
| docs.hpc.cam.ac.uk | names system and part | AMD EPYC 7763 (Milan), NVIDIA InfiniBand HDR |
| top500.org | names system and part | AMD EPYC 7763 (Milan), NVIDIA InfiniBand HDR |
Checked 2026-10-09 by pnpm verify. Re-run it and the table changes with the web.
Further reading
Operating a supercomputer
Acceptance testing, why "installed" and "in production" are six months apart, and the machine lifecycle.
Inside a node
Latency-optimised versus throughput-optimised processors, NUMA, and the end of the host-to-device copy.
The interconnect
Topologies, why latency and tail behaviour matter more than bandwidth, and how the fabric market consolidated.