Supercomputers/Systems/Cannon (FASRC)
USA · upgraded · hpc
Cannon (FASRC)
Also known as Cannon (2019); FASRC Cannon
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
- 2.08 PFlop/s
- Rpeak
- 2.98 PFlop/s
- Rmax ÷ Rpeak
- 69.8%
- Cores
- 32,160
- 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 | - 670 Lenovo SD650 NeXtScale water-cooled nodes | Lenovo | 670 reported | Reported | top500.orgmghpcc.org |
| Accelerator | NVIDIA Tesla V100 Accelerators CUDA (Volta GV100) · 5,120C · 1.53 GHz · 12 nm · 300 W 16 SR670 servers with four NVIDIA V100 GPUs each | NVIDIA | 64 reported | Reported | mghpcc.org |
| CPU | - Intel Xeon Platinum 8268 24C 2.9GHz (Cascade Lake), 48 cores per node | Intel | - | Reported | top500.orgmghpcc.org |
| Interconnect | NVIDIA InfiniBand HDR Interconnect Fat tree or dragonfly+ · 200 Gb/s per port · About 0.6 microseconds HDR100 InfiniBand single fat tree with 200 Gbps core | NVIDIA | - | Reported | mghpcc.orgtop500.org |
The read
Harvard's Lenovo water-cooled Cannon cluster, refreshed in 2019 to 670 SD650 NeXtScale nodes with Intel Xeon Platinum 8268 'Cascade Lake' processors (48 cores per node, about 30,000 cores per MGHPCC) on a single fat-tree HDR100 InfiniBand fabric, plus 16 SR670 servers with four V100 GPUs each. TOP500 lists 32,160 cores at 2.08 PFlop/s Rmax and 2.98 Rpeak, debuting 144th in November 2019 and last appearing at 485th in November 2023. FASRC has since added GPU partitions (see Cannon2-H200), so the status is upgraded rather than a snapshot of the 2019 build; the CPU node fleet's present state was not confirmed.
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
Change history
- 2026-10-02 Added Cannon, Harvard's Lenovo water-cooled cluster, 144th on the November 2019 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 |
|---|---|---|
| mghpcc.org | names system and part | NVIDIA InfiniBand HDR |
| top500.org | names system and part | 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.