Parts/Storage/Scalable Disk Array
Storage · SDA · 1993
Scalable Disk Array
Disk storage nodes plugged directly into the CM-5's own data network, NFS-mountable as the Scalable File System (SFS). Predates Lustre and GPFS by a decade.
Why this part
Thinking Machines' own October 1993 Scale 5 brochure describes a processing-intensive configuration with 12 disk storage nodes, 176 GB of capacity and 132 MB/s of transfer, scaling to a file-intensive configuration with 48 disk storage nodes, 704 GB and 528 MB/s. There is no separate metadata server or object layer here, the pattern later filesystems would formalize; storage nodes are simply more nodes on the same fat tree the compute nodes use, which is both the design's simplicity and its ceiling.
Systems using this part
- Systems
- 2
- Rmax underneath
- 59.7 GFlop/s
- Units recorded
- -
| System | Role | Country | Units | System Rmax | Supplied by | Rank |
|---|---|---|---|---|---|---|
| CM-5/1024 1993-06 | Storage | USA | - | 59.7 GFlop/s | Thinking Machines | #1 → |
| Littlebear (NCAR CM-5) 1993-04 | Storage | USA | - | - | Thinking Machines | - |
“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.