Supercomputers/Systems/CM-2
USA · decommissioned · hpc
CM-2
Also known as Connection Machine CM-2 "Capitol"
Operated by NCAR at NCAR Mesa Laboratory (Boulder) .
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
- -
- Rpeak
- 10.0 GFlop/s
- Rmax ÷ Rpeak
- -
- Cores
- 8,192
- 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 | - Thinking Machines Corporation | Thinking Machines | - | Reported | cisl.ucar.edu |
| CPU | Thinking Machines CM-2 processor array CPUs 1-bit SIMD + Weitek FPU · 65,536C 8,192-processor CM-2 configuration ('Capitol') | Thinking Machines | 8,192 reported | Reported | cisl.ucar.eduen.wikipedia.org |
| Interconnect | Thinking Machines CM-2 hypercube network Interconnect Hypercube CM-2 hypercube data network | Thinking Machines | - | Reported | cisl.ucar.eduen.wikipedia.org |
The read
Thinking Machines launched the CM-2 in 1987, adding Weitek floating-point coprocessors to the original CM-1's massively parallel SIMD design (see the tmc-cm2-processor and tmc-cm2-hypercube parts). NCAR's Scientific Computing Division took delivery of an 8,192-processor CM-2, nicknamed 'Capitol', in August 1988 as the operational home of the Center for Applied Parallel Processing (CAPP), a joint initiative with the University of Colorado and NOAA; NCAR's own page dates its in-service period 1 September 1988 to 1 April 1993, when it was replaced by a Cray Y-MP-class successor in NCAR's own lineup (not the LANL CM-5 that this dataset already carries as cm-5, a different institution's later and larger machine from the same vendor). Disambiguated from Thinking Machines' other Connection Machine models by name, per this dataset's site-collision rule. This is a single-source installation record: NCAR's own history page is the only source found that confirms Capitol's specific dates and configuration, so the tier stays reported despite Wikipedia independently confirming the CM-2's general architecture.
Site & facility
NCAR Mesa Laboratory (Boulder)
- Location
- Boulder, CO, USA
- Commissioned
- 1986
Change history
- 2026-09-24 CM-2 added
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 |
|---|---|---|
| en.wikipedia.org | names system, part and count | Thinking Machines CM-2 processor array, Thinking Machines CM-2 hypercube network · counts: 8192 |
| cisl.ucar.edu | names system, part and count | Thinking Machines CM-2 processor array, Thinking Machines CM-2 hypercube network · counts: 8192 |
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.
Operating a supercomputer
Acceptance testing, why "installed" and "in production" are six months apart, and the machine lifecycle.