Supercomputers/Systems/TX-GAIA
USA · operational · mixed
TX-GAIA
Also known as Green AI Accelerator
Operated by MIT Lincoln Laboratory at MIT LLSC Holyoke .
Phase 1 seed dataset, compiled by hand. These rows were built from public operator, laboratory and vendor sources. A mechanical second-reader pass has since fetched every cited source: 1092 of 1153 systems have a readable citation that names them, and 40 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
- 5.16 PFlop/s
- Rpeak
- 6.98 PFlop/s
- Rmax ÷ Rpeak
- 73.9%
- Cores
- 88,400
- Power
- 1.44 MW
- Per watt
- 3.6 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 | - HPE Apollo 2000 system | Hewlett Packard Enterprise | - | Verified | top500.orgdeveloper.nvidia.com |
| Accelerator | NVIDIA Tesla V100 Accelerators CUDA (Volta GV100) · 5,120C · 1.53 GHz · 12 nm · 300 W 896 NVIDIA Tesla V100 Tensor Core GPUs (2 per node) | NVIDIA | 896 reported | Reported | developer.nvidia.com |
| CPU | - Intel Xeon Gold 6248 20C 2.5GHz (nearly 900 processors) | Intel | - | Reported | top500.orgnews.mit.edu |
| Interconnect | Omni-Path Architecture Interconnect Fat tree · 100 Gb/s per port · About 1 microsecond Intel Omni-Path | Intel | - | Reported | top500.org |
The read
MIT Lincoln Laboratory Supercomputing Center's HPE Apollo 2000 AI supercomputer, announced on 27 September 2019 and housed in an HPE EcoPOD modular data center at a hydroelectrically powered site in Holyoke, Massachusetts. MIT and NVIDIA describe nearly 900 Intel Xeon processors and 896 NVIDIA V100 GPUs with a stated 100 peak AI petaflops, which MIT called the most powerful AI supercomputer at any university at the time. MIT's launch article gave 3.9 PFlop/s Rmax (June 2019) and later 4.725 PFlop/s; TOP500's current system page lists 5.16 PFlop/s Rmax, 6.98 PFlop/s Rpeak, 88,400 cores and 1,436 kW, and those figures are used here. TOP500 shows the system on the list continuously from June 2019 (rank 51) through June 2026 (rank 299). The tier is reported because the fetched sources give differing Rmax values.
Timeline
- 2019-09 Announced source
- 2019-09 First operational source
- 2019-11 List debut at #42 (November 2019) 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
- 2026-10-01 Added TX-GAIA (MIT Lincoln Laboratory)
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 |
|---|---|---|
| news.mit.edu | names system only | - |
| top500.org | names system and part | NVIDIA Tesla V100, Omni-Path Architecture |
Checked 2026-10-09 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.