Canada · operational · mixed

Nibi

Also known as Nibi (GPU)

Operated by SHARCNET at University of Waterloo (Nibi) .

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: 7.99 PFlop/s
Rpeak: 15.36 PFlop/s
Rmax ÷ Rpeak: 52.0%
Cores: 6,144
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
Integrator	- Hypertec system architect and prime integrator; TOP500 lists a Hypertec TRIDENT iC610DR-G7	Hypertec Cloud	-	Reported	telconews.ca top500.org
Accelerator	- 288 NVIDIA H100 GPUs	NVIDIA NVDANASDAQ	288 reported	Reported	alliancecan.ca alliancecan.ca
CPU	- Intel Xeon 6972P 96C 2.4GHz (TOP500 GPU entry host CPU)	Intel INTCNASDAQ	-	Reported	top500.org
Interconnect	- Ethernet data centre fabric (400g Ethernet per TOP500), Nokia networking	Nokia	-	Reported	telconews.ca top500.org
Cooling	- Immersion cooling plus direct liquid cooling, deployed by Hypertec	Hypertec Cloud	-	Reported	telconews.ca alliancecan.ca

The read

SHARCNET's national system at the University of Waterloo, run for the Digital Research Alliance of Canada and replacing Graham. The Alliance describes 134,400 CPU cores (the Waterloo news piece says about 140,000 over 700+ nodes), 288 NVIDIA H100 GPUs and more than 25 PB of all-flash storage; Hypertec was the system architect and prime integrator and deployed immersion cooling alongside direct liquid cooling, on an Ethernet fabric from Nokia, with waste heat reused to warm the Lazaridis Quantum-Nano Centre. Full production began on 31 October 2025. TOP500 lists Nibi as separate CPU and GPU entries; the row here carries the GPU entry (Hypertec TRIDENT iC610DR-G7, Xeon 6972P, 400g Ethernet, Rocky Linux 8, 6,144 cores, Rmax 7.985 PFlop/s, Rpeak 15.36 PFlop/s), which ranked 173rd on the June 2025 list, its best rank, and 227th on the June 2026 list. The benchmark run predates the October 2025 production date, so first_operational_date reflects the Alliance's production announcement rather than the TOP500 submission. The TOP500 core count is that partition's, not the 134,400-core whole system. The Alliance wiki page with node-level detail blocked automated access, so per-node counts are not used.

Timeline

2025-06 List debut at #173 (June 2025) source
2025-10 First operational source

Public list appearances

Pointers only: rank and edition, linking to the canonical entry. We do not reproduce list tables. See sourcing policy.

#173: TOP500, June 2025 →

Site & facility

University of Waterloo (Nibi)

Location: Waterloo, Canada
Cooling: Immersion and direct liquid cooling
Commissioned: 2025

uwaterloo.ca alliancecan.ca

Change history

2026-10-02 Added Nibi, SHARCNET's Waterloo H100 research system, 173rd on the June 2025 TOP500

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
alliancecan.ca	names system only	-
top500.org	names system only	-

Checked 2026-10-09 by pnpm verify. Re-run it and the table changes with the web.