Supercomputers/Systems/Kay
Ireland · decommissioned · hpc
Kay
Operated by ICHEC at ICHEC National HPC Centre (NUI Galway) .
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
- 665.00 TFlop/s
- Rpeak
- -
- Rmax ÷ Rpeak
- -
- Cores
- 13,440
- Power
- -
- Per watt
- - 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 |
|---|---|---|---|---|---|
| Accelerator | NVIDIA Tesla V100 Accelerators CUDA (Volta GV100) · 5,120C · 1.53 GHz · 12 nm · 300 W 2x NVIDIA Tesla V100 16GB PCIe (Volta) per GPU node, 16 GPU nodes, 64 GPUs total (TechCentral's launch-era article instead names Nvidia P100 GPUs; see notes_md) | NVIDIA | 64 reported | Reported | ichec.ie |
| CPU | Intel Xeon Gold 6148 CPUs x86-64 · 20C · 2.4 GHz · 14 nm · 150 W 2x 20-core 2.4GHz Intel Xeon Gold 6148 (Skylake), 336 nodes, 13,440 cores total | Intel | - | Reported | ichec.ietechcentral.ie |
| Interconnect | Omni-Path Architecture Interconnect Fat tree · 100 Gb/s per port · About 1 microsecond 100Gbit OmniPath network adaptor | Cornelis Networks | - | Verified | ichec.ietechcentral.ie |
| Storage | - DDN SFA12k-20 platform, 560 TB capacity, Lustre filesystem | DDN | - | Reported | ichec.ie |
The read
Ireland's former national supercomputer, operated by ICHEC (the Irish Centre for High-End Computing) at NUI Galway from August 2018 until the end of its five-year service life in November 2023, after which the National HPC Service transitioned to renting time on Luxembourg's MeluXina platform per ICHEC's own later announcements. ICHEC's own infrastructure page and the independent trade outlet TechCentral.ie agree on the core cluster: 336 nodes and 13,440 CPU cores over a 100Gbit Intel OmniPath fabric, commissioned in 2018 at NUI Galway. TechCentral's article, written around the system's launch, describes the accelerator partition as Intel Xeon Phi and Nvidia P100 GPUs, while ICHEC's own current infrastructure page describes 16 Intel Xeon Phi nodes and a separate 16-node GPU partition with NVIDIA Tesla V100 PCIe GPUs; this dataset records ICHEC's own V100 description as the operator's current statement of record and notes the P100 discrepancy here rather than resolving it, since it most likely reflects the difference between an early vendor announcement and the final delivered configuration, so the accelerator edge and this row stay reported. Rmax of 665 TFlop/s (0.665 PFlop/s) is ICHEC's own stated Linpack figure; no TOP500 system page with a specific rank could be confirmed from a directly-fetched source, so ranked is left 0 even though ICHEC's own November 2018-era announcements describe that period as adding multiple new Irish TOP500 entries.
Timeline
Site & facility
ICHEC National HPC Centre (NUI Galway)
- Location
- Galway, Ireland
- Commissioned
- 2018
Change history
- 2026-09-24 Kay added (Ireland, decommissioned)
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.
Every source cited on this page was readable, and none of them mentions this system. They are cited at page granularity, normally because no stable deep link could be found. That is a real weakness in this row and it is why the system carries a confidence tier below verified. See how this check works.
| Cited source | Result | Found on the page |
|---|---|---|
| ichec.ie | does not name this system | Intel Xeon Gold 6148, NVIDIA Tesla V100 · counts: 64 |
| techcentral.ie | does not name this system | Intel Xeon Gold 6148 · counts: 64 |
Checked 2026-10-09 by pnpm verify. Re-run it and the table changes with the web.
Further reading
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.
Storage and I/O
Parallel file systems, why checkpointing dominates the write load, and the metadata failure mode nobody expects.