Supercomputers/Systems/Facebook DGX-1 Cluster
USA · operational · mixed
Facebook DGX-1 Cluster
Operated by Meta Platforms .
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: 1263 of 1345 systems have a readable citation that names them, and 56 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
- 3.31 PFlop/s
- Rpeak
- 4.90 PFlop/s
- Rmax ÷ Rpeak
- 67.5%
- Cores
- 60,512
- Power
- 349.5 kW
- Per watt
- 9.5 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 | - Self-built cluster of NVIDIA DGX-1 and Relion 2904GT servers as listed by TOP500 | Meta Platforms | - | Reported | top500.org |
| Accelerator | NVIDIA Tesla P100 Accelerators CUDA (Pascal GP100) · 3,584C · 1.48 GHz · 16 nm · 300 W NVIDIA Tesla P100 and Quadro GP100 | NVIDIA | - | Reported | top500.org |
| CPU | - Intel Xeon E5-2698v4 20C 2.2GHz hosts | Intel | - | Reported | top500.org |
| Interconnect | Mellanox InfiniBand EDR Interconnect Fat tree · 100 Gb/s per 4x port · About 0.9 microseconds InfiniBand EDR | NVIDIA | - | Reported | top500.org |
| Operating system | Ubuntu Server Operating systems Linux distribution · Open source, free with paid support tiers Ubuntu | Canonical | - | Reported | top500.org |
The read
Facebook's (now Meta's) self-built cluster of NVIDIA DGX-1 and Relion 2904GT servers as listed by TOP500, with Xeon E5-2698v4 20C 2.2GHz hosts, NVIDIA Tesla P100 and Quadro GP100 GPUs on InfiniBand EDR, running Ubuntu with Intel and CUDA compilers: 60,512 cores at 3.307 PFlop/s Rmax and 4.897 PFlop/s Rpeak on 349.5 kW (power measured at level 2). TOP500 lists the site as Facebook in Santa Clara with manufacturer Self-made. It debuted at rank 32 on the June 2017 list and is still listed at rank 399 on June 2026, which is the basis for the status date. The workload is not stated by TOP500.
Timeline
Public list appearances
Pointers only: rank and edition, linking to the canonical entry. We do not reproduce list tables. See sourcing policy.
Change history
- 2026-10-09 Added Facebook DGX-1 Cluster
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 |
|---|---|---|
| top500.org | names system and part | NVIDIA Tesla P100, Mellanox InfiniBand EDR |
| top500.org | names system and part | NVIDIA Tesla P100, Mellanox InfiniBand EDR, Ubuntu Server |
Checked 2026-10-13 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.