Cannon2-H200

Also known as Cannon gpu_h200 partition

Operated by Harvard FAS Research Computing at MGHPCC Holyoke .

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: 4.24 PFlop/s
Rpeak: 6.60 PFlop/s
Rmax ÷ Rpeak: 64.2%
Cores: 15,360
Power: 91.61 kW
Per watt: 46.3 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	- 24 Lenovo SD650-N v3 servers	Lenovo 992HKEX	24 reported	Reported	rc.fas.harvard.edu top500.org
Accelerator	NVIDIA H200 SXM Accelerators CUDA (Hopper GH100) · 16,896C · 1.98 GHz · TSMC 4N · 700 W 96 NVIDIA H200 GPUs (4 per node, 141 GB each)	NVIDIA NVDANASDAQ	96 reported	Reported	rc.fas.harvard.edu top500.org
CPU	- Intel Sapphire Rapids, 112 cores per node (TOP500: Xeon Platinum 8480L 56C 2GHz)	Intel INTCNASDAQ	-	Reported	rc.fas.harvard.edu top500.org
Interconnect	NVIDIA InfiniBand NDR (Quantum-2) Interconnect Fat tree or dragonfly+ · 400 Gb/s per port · Under 0.6 microseconds NDR InfiniBand	NVIDIA NVDANASDAQ	-	Reported	rc.fas.harvard.edu top500.org

The read

The H200 partition of Harvard's Cannon cluster, announced by FAS Research Computing on 18 August 2025: 24 Lenovo SD650-N v3 servers with four NVIDIA H200 GPUs (141 GB each) apiece, 96 H200 in total, 112 Sapphire Rapids cores and 1 TB of RAM per node, on NDR InfiniBand. TOP500 lists it as Cannon2-H200 with 15,360 cores (this count appears to include GPU streaming multiprocessors, since 24 nodes carry only 2,688 CPU cores), 4.24 PFlop/s Rmax and 6.60 PFlop/s Rpeak, 302nd in November 2025 and 331st in June 2026. The MGHPCC Holyoke site is attached on the basis that Cannon is housed there per the MGHPCC refresh page; the H200 blog itself does not name the building.

Timeline

2025-08 Announced source
2025-08 First operational source
2025-11 List debut at #302 (November 2025) source

Public list appearances

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

#302: TOP500, November 2025 →

Site & facility

MGHPCC Holyoke

Location: Holyoke, MA, USA

kempnerinstitute.harvard.edu news.harvard.edu

Change history

2026-10-02 Added Cannon2-H200, Harvard's H200 partition, 302nd on the November 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
top500.org	names system, part and count	NVIDIA H200 SXM, NVIDIA InfiniBand NDR (Quantum-2) · counts: 96
rc.fas.harvard.edu	does not name this system	NVIDIA H200 SXM, NVIDIA InfiniBand NDR (Quantum-2) · counts: 96

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