Companies/Thinking Machines
Cambridge, MA, USA · defunct
Thinking Machines
Built the Connection Machine series, including the CM-5 that topped the first published HPL list in June 1993. Filed for bankruptcy in 1994; no surviving corporate parent.
Systems this company appears in
- Systems
- 5
- Aggregate Rmax
- 120.5 GFlop/s
- Aggregate power
- 0 kW
- First appearance
- 1988-09
- Most recent
- 1993-06
- Concentration
- 50%
Aggregate Rmax is the sum of the systems this company supplies any part of. It is not revenue and not attributable share: an integrator and a memory vendor in the same machine both count its full FLOPS here. Attribution weighted by role is Phase 4.
Role breakdown
| Role | Systems | Rmax in those systems |
|---|---|---|
| Integrator | 5 | 120.5 GFlop/s |
| CPU | 5 | 120.5 GFlop/s |
| Interconnect | 4 | 120.5 GFlop/s |
| Storage | 2 | 59.7 GFlop/s |
| Cooling | 1 | 59.7 GFlop/s |
| Operating system | 4 | 120.5 GFlop/s |
Appearances
| System | Supplied as | Shipped under | Rmax | Rank |
|---|---|---|---|---|
| CM-5/1024 USA · 1993-06 | CPU, Integrator, Interconnect, Operating system, Storage, Cooling | Thinking Machines | 59.7 GFlop/s | #1 June 1993 → |
| Littlebear (NCAR CM-5) USA · 1993-04 | CPU, Integrator, Operating system, Storage | Thinking Machines | - | unranked |
| FROSTBURG USA · 1993 | Integrator, CPU, Interconnect, Operating system | Thinking Machines | 30.4 GFlop/s | #4 June 1993 → |
| NCSA CM-5/512 USA · 1993 | Integrator, CPU, Interconnect, Operating system | Thinking Machines | 30.4 GFlop/s | #3 June 1993 → |
| CM-2 USA · 1988-09 | CPU, Interconnect, Integrator | Thinking Machines | - | unranked |
Parts supplied
| Part | Category | Generation | Released |
|---|---|---|---|
| Thinking Machines CM-5 fat tree | Interconnect | Connection Machine | 1993 |
| Thinking Machines CM-5 vector node | CPUs | Connection Machine | 1993 |
| Scalable Disk Array | Storage | SDA | 1993 |
| Thinking Machines CMost | Operating systems | CMost | 1991 |
| Thinking Machines CM-2 processor array | CPUs | Connection Machine CM-2 | 1987 |
| Thinking Machines CM-2 hypercube network | Interconnect | Connection Machine CM-2 | 1987 |
Further reading
Inside a node
Latency-optimised versus throughput-optimised processors, NUMA, and the end of the host-to-device copy.
Operating a supercomputer
Acceptance testing, why "installed" and "in production" are six months apart, and the machine lifecycle.
The interconnect
Topologies, why latency and tail behaviour matter more than bandwidth, and how the fabric market consolidated.