Supercomputers/Systems/Sapphire (ERDC)
USA · operational · hpc
Sapphire (ERDC)
Also known as ERDC Cray XT3
Operated by ERDC DSRC at ERDC DSRC (Vicksburg) .
Researched with AI assistance, checked by scripts. These rows were researched by AI agents reading public operator, laboratory and vendor sources, appended by scripts and checked by automated validation; no subject-matter expert has reviewed them (how this site is made). A mechanical second-reader pass has since fetched every cited source: 1350 of 1540 systems have a readable citation that names them, and 64 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
- 16.98 TFlop/s
- Rpeak
- 21.30 TFlop/s
- Rmax ÷ Rpeak
- 79.7%
- Cores
- 4,096
- 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 | - Cray XT3, Red Storm architecture, 4,096 Opteron cores at ERDC | Cray HPENYSE
rolls up to Hewlett Packard Enterprise | - | Reported | top500.orgnextgov.com |
| CPU | - AMD Opteron 2.6 GHz, 4,096 cores | AMD | 4,096 reported | Reported | top500.orgnextgov.com |
| Interconnect | - Cray XT3 internal interconnect | Cray HPENYSE
rolls up to Hewlett Packard Enterprise | - | Reported | top500.orgnextgov.com |
The read
Cray XT3 at the US Army Engineer Research and Development Center's DoD supercomputing centre in Vicksburg, Mississippi, built on the Red Storm design that Cray and Sandia developed together (see red-storm and the PSC machine bigben). A February 2005 Nextgov report on Cray's contract announcement describes more than 4,000 AMD Opteron processors and more than 21 trillion calculations per second, as part of a $23 million award that would more than triple the centre's capability; the centre's other systems at the time included a 64-processor Cray X1 and a Cray T3E, and delivery was planned for summer 2005. The $23 million covers the whole Cray award and is not a price for this machine alone. TOP500 lists 4,096 cores of 2.6 GHz Opteron, 16.98 TFlop/s Rmax and 21.30 TFlop/s peak under UNICOS, ranked 14th in November 2005, 20th in June 2006 and 26th in November 2006, its last list. The HPC Challenge site separately records an HPL run of 16.9752 TFlop/s on 4,096 processes for an ERDC XT3 running UNICOS/lc 1.2. No retirement date was found, so the status is carried at its last TOP500 month.
Timeline
- 2005 First operational source
- 2005-01 Announced source
- 2005-11 List debut at #14 (November 2005) source
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
- 2026-10-17 Sapphire (ERDC) added from its source pages
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 |
|---|---|---|
| nextgov.com | does not name this system | - |
| top500.org | names system only | - |
Checked 2026-10-18 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.