Supercomputers/Systems/Tianhe-1A
China · operational · hpc
Tianhe-1A
Also known as TH-1A
Operated by National Supercomputing Center in Tianjin at National Supercomputing Center in Tianjin .
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
- 2.57 PFlop/s
- Rpeak
- 4.70 PFlop/s
- Rmax ÷ Rpeak
- 54.6%
- Cores
- 186,368
- Power
- 4.04 MW
- Per watt
- 0.6 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 | - NUDT TH-1A | NUDT | - | Reported | nscc-tj.cn |
| Accelerator | NVIDIA Tesla M2050 Accelerators CUDA (Fermi) · 448C · 1.15 GHz · 40 nm · 225 W NVIDIA Tesla M2050 | NVIDIA | 7,168 reported | Reported | nscc-tj.cn |
| CPU | Intel Xeon X5670 CPUs x86-64 · 6C · 2.93 GHz · 32 nm · 95 W Intel Xeon X5670 6C 2.93GHz | Intel | 14,336 reported | Reported | nscc-tj.cn |
| Interconnect | NUDT Arch interconnect Interconnect Hybrid fat tree, optoelectronic · 160 Gb/s bidirectional per link · 1.57 microseconds per hop NUDT proprietary interconnect | NUDT | - | Verified | nscc-tj.cntheregister.com |
| Operating system | Kylin Linux Operating systems Linux distribution · State-endorsed Chinese distribution Kylin Linux | KylinSoft | - | Reported | nscc-tj.cn |
The read
The first Chinese No. 1, and a system that is routinely mis-told. The compute was American: Intel Westmere CPUs and NVIDIA Fermi accelerators. The interconnect was not: NUDT's proprietary Arch fabric was domestically designed and roughly twice the bandwidth of the InfiniBand generation available at the time. Read as a supply chain rather than a spec string, Tianhe-1A is the first visible move in a substitution programme that reached the compute layer six years later. Current operational status has not been independently confirmed since its last public list appearance; we carry the last documented state rather than inferring a retirement.
Timeline
- 2009 Announced source
- 2010-11 List debut at #1 (November 2010) source
- 2010-11 First operational source
- 2015-02 US Entity List addition targets operator NSCC Tianjin over Tianhe-1A's use of US-origin processors 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
National Supercomputing Center in Tianjin
- Location
- Tianjin, China
- Cooling
- Air cooling
- Commissioned
- 2010
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 |
|---|---|---|
| jcst.ict.ac.cn | names system and part | NUDT Arch interconnect |
| theregister.com | names system, part and count | NVIDIA Tesla M2050, Intel Xeon X5670, NUDT Arch interconnect · counts: 7168, 14336 |
Checked 2026-10-18 by pnpm verify. Re-run it and the table changes with the web.
Further reading
Inside a node
Why accelerators exist, what the memory hierarchy costs, and how the CPU and accelerator merged onto one package.
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.