Supercomputers/Systems/NCEP Gyre
USA · operational · hpc
NCEP Gyre
Also known as Gyre
Operated by NOAA .
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 1581 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
- 705.90 TFlop/s
- Rpeak
- 776.09 TFlop/s
- Rmax ÷ Rpeak
- 91.0%
- Cores
- 37,312
- Power
- 775.04 kW
- Per watt
- 0.9 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.
How it compares
Ranked against the other machines in this dataset that report the same figure: those that first came into service within two years of NCEP Gyre, and those in USA. Computed from our tables; a metric is ranked only when at least eight peers report it.
| Metric | Value | Against its era | Against its country |
|---|---|---|---|
| Rmax | 705.90 TFlop/s | #90 of 227 · top 39% | #195 of 332 · bottom 42% |
| Energy efficiency | 0.9 GF/W | #116 of 186 · bottom 39% | #101 of 162 · bottom 39% |
| Rmax ÷ Rpeak | 91.0 % | #24 of 226 · top 10% | #15 of 331 · top 4% |
| Power draw | 775 kW | #67 of 188 · top 35% | #72 of 171 · top 42% |
Every metric is ranked largest-first, so a high power-draw rank means a big power draw, not a good one. Rmax ÷ Rpeak uses HPL-measured results only. Rankings move as machines are added; they describe this dataset, not the world.
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 | - IBM iDataPlex DX360M4 | IBM | - | Reported | top500.orgnextplatform.com |
| CPU | Intel Xeon E5-2670 CPUs x86-64 · 8C · 2.6 GHz · 32 nm · 115 W Intel Xeon E5-2670 8C 2.6GHz | Intel | - | Reported | top500.org |
| Interconnect | Mellanox InfiniBand FDR Interconnect Fat tree · 56 Gb/s per 4x port · About 1 microsecond Infiniband FDR (56 Gb/s) | Mellanox NVDANASDAQ
rolls up to NVIDIA | - | Reported | top500.orgnextplatform.com |
| Operating system | Red Hat Enterprise Linux Operating systems Linux distribution · Open source, commercial support subscription Red Hat Enterprise Linux | Red Hat | - | Reported | nextplatform.com |
Engineering, funding and access
What NCEP Gyre runs on, who paid for it and how people get time on it. Each line carries who said it and when.
Hardware
| Item | Detail | As of | Basis | Source |
|---|---|---|---|---|
| Interconnect and topology | 56 Gb/s FDR InfiniBand | 2016-01 | Press report | nextplatform.com |
Software
| Item | Detail | As of | Basis | Source |
|---|---|---|---|---|
| Job scheduler | IBM LSF | 2016-01 | Press report | nextplatform.com |
| Programming and software stack | Red Hat Enterprise Linux | 2016-01 | Press report | nextplatform.com |
Funding
| Item | Detail | As of | Basis | Source |
|---|---|---|---|---|
| Procurement | IBM iDataPlex under the ten-year WCOSS contract worth USD 502 million, awarded March 2011; the Orlando backup twin of TideThe contract value covers the whole WCOSS program, not Gyre alone. | 2016-01 | Press report | nextplatform.com |
The read
The Orlando backup-center twin of Tide (ncep-tide) in NOAA's first-phase operational weather computing: an IBM iDataPlex DX360M4 cluster on 56 Gb/s FDR InfiniBand running Red Hat Enterprise Linux with IBM LSF, per Next Platform. TOP500 gives Gyre exactly the same figures as Tide (37,312 cores, 705.9 TFlop/s Rmax, 776.09 TFlop/s Rpeak, 775.04 kW) and lists it one rank below Tide on every list from June 2015 (rank 104) to November 2017 (rank 380). Next Platform reports the Tide and Gyre clusters processing 3.5 billion observations a day with a 99.9 percent uptime goal for running models; it also says each cluster grew to about 830 TFlop/s after IBM's January 2015 node additions, a figure TOP500 does not reflect. The Cray XC40 and Dell systems NOAA added later are recorded as noaa-luna, noaa-surge, noaa-mars and noaa-venus. Tier is reported because TOP500 is the only source for the benchmark figures.
Timeline
- 2013-07 First operational source
- 2015-01 IBM added iDataPlex nodes, taking the WCOSS iDataPlex capacity to about 830 teraflops of aggregate compute source
- 2015-11 List debut at #127 (November 2015) source
Public list appearances
Pointers only: rank and edition, linking to the canonical entry. We do not reproduce list tables. See sourcing policy.
Change history
- 2026-10-17 NCEP Gyre added from its source pages
Nearby in scale
The machines closest to NCEP Gyre by Rmax, smaller to larger.
- NCEP Tide · USA, operational · 0.71 PFlop/s
- BEM · Poland, operational · 0.7 PFlop/s
- Dammam (2017) · Saudi Arabia, operational · 0.69 PFlop/s
- Owens · USA, decommissioned · 0.71 PFlop/s
- Blue Brain IV · Switzerland, decommissioned · 0.72 PFlop/s
- Avoca · Australia, decommissioned · 0.72 PFlop/s
Related systems
Same operator
- NCEP Tide · USA, operational, 2013
- NOAA Luna · USA, operational, 2015
- NOAA Surge · USA, operational, 2015
- Hera · USA, operational, 2019
- Gaea · USA, operational, 2023
- NOAA FSL HPTi Xeon Cluster · USA, operational, 2002
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 |
|---|---|---|
| nextplatform.com | names system and part | Intel Xeon E5-2670, Mellanox InfiniBand FDR |
| top500.org | names system and part | Intel Xeon E5-2670, Mellanox InfiniBand FDR |
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.