Parts/Operating systems/Hitachi HI-UX/MPP
Why this part
Built on a Mach 3.0 microkernel rather than a monolithic Unix, which is what let Hitachi split the SR2201 the way it did: a thin, low-overhead layer on the 1,024 or 2,048 compute nodes and a fuller environment on the I/O side, without maintaining two unrelated kernels. It shipped under two different Hitachi customers within the same year, the University of Tokyo's SR2201/1024 and the University of Tsukuba's CP-PACS, which is the clearest sign that the SR2201 was a genuine product line sold to multiple buyers rather than a one-off machine built for a single physics department.
Systems using this part
- Systems
- 7
- Rmax underneath
- 5.70 TFlop/s
- Units recorded
- -
| System | Role | Country | Units | System Rmax | Supplied by | Rank |
|---|---|---|---|---|---|---|
| UTokyo SR8000/MPP 2001 | Operating system | Japan | - | 1.71 TFlop/s | Hitachi | #5 → |
| KEK SR8000-F1/100 2000-03 | Operating system | Japan | - | 917.0 GFlop/s | Hitachi | #6 → |
| ISSP SR8000-F1/60 2000-03 | Operating system | Japan | - | 577.0 GFlop/s | Hitachi | #17 → |
| HLRB SR8000-F1 2000 | Operating system | Germany | - | 1.03 TFlop/s | Hitachi | #5 → |
| UTokyo SR8000/128 1999 | Operating system | Japan | - | 873.0 GFlop/s | Hitachi | #5 → |
| CP-PACS 1996-10 | Operating system | Japan | - | 368.2 GFlop/s | Hitachi | #1 → |
| Hitachi SR2201/1024 1996 | Operating system | Japan | - | 220.4 GFlop/s | Hitachi | #1 → |
“Units recorded” sums only the edges where a public source states a quantity. It is a floor, not a total, and should never be read as installed base.