Supercomputers/Systems/Phoenix (PACE)
USA · operational · mixed
Phoenix (PACE)
Also known as Georgia Tech Phoenix
Operated by Georgia Institute of Technology at Georgia Tech Coda Data Center .
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
- 1.84 PFlop/s
- Rpeak
- 2.69 PFlop/s
- Rmax ÷ Rpeak
- 68.4%
- Cores
- 31,104
- 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 | - Penguin Computing (TOP500 manufacturer for the 2020 submission) | Penguin Computing | - | Reported | top500.org |
| Accelerator | NVIDIA Tesla V100 Accelerators CUDA (Volta GV100) · 5,120C · 1.53 GHz · 12 nm · 300 W 74 GPU nodes with 2x NVIDIA Tesla V100 (quantity derived as 74 x 2) | NVIDIA | 148 reported | Reported | bpb-us-e1.wpmucdn.comtop500.org |
| Accelerator | NVIDIA H100 SXM5 Accelerators CUDA (Hopper GH100) · 16,896C · 1.98 GHz · TSMC 4N · 700 W 4 GPU nodes with 8x NVIDIA H100 (quantity derived as 4 x 8) | NVIDIA | 32 reported | Reported | bpb-us-e1.wpmucdn.com |
| Accelerator | NVIDIA H200 SXM Accelerators CUDA (Hopper GH100) · 16,896C · 1.98 GHz · TSMC 4N · 700 W 7 GPU nodes with 8x NVIDIA H200 (quantity derived as 7 x 8) | NVIDIA | 56 reported | Reported | bpb-us-e1.wpmucdn.com |
| CPU | - Intel Xeon Gold 6226 12C 2.7GHz (TOP500 submission) | Intel | - | Reported | top500.org |
| Interconnect | NVIDIA InfiniBand HDR Interconnect Fat tree or dragonfly+ · 200 Gb/s per port · About 0.6 microseconds Mellanox HDR InfiniBand, HDR100 links to each compute node and HDR200 uplinks | NVIDIA | - | Verified | bpb-us-e1.wpmucdn.comtop500.org |
| Storage | Lustre Storage Parallel filesystem, separate metadata and object servers DDN SFA 18K running Lustre, multiple petabytes | DDN | - | Reported | bpb-us-e1.wpmucdn.com |
The read
Georgia Tech's main PACE research cluster, deployed in fall 2020 in the Coda data center and grown since. Georgia Tech's facilities document (revision 4.3, May 2025) lists 36,328 cores across 1,403 nodes plus 74 GPU nodes with 2x Tesla V100, 37 nodes with 4x RTX 6000, 10 L40S nodes, 4 nodes with 8x H100 and 7 nodes with 8x H200, on HDR InfiniBand (HDR100 to each node) and DDN SFA 18K Lustre storage. The HPL figures are the November 2020 TOP500 submission (Penguin Computing, Xeon Gold 6226 12C, Tesla V100, HDR100; 31,104 cores; rank 277), which both TOP500 and Georgia Tech state as 1.84 PFlop/s, so the tier is verified for that figure. Phoenix stayed on the list through November 2022 (rank 463). Power was not stated in the sources fetched.
Timeline
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 Phoenix (PACE) (Georgia Tech)
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 |
|---|---|---|
| bpb-us-e1.wpmucdn.com | does not name this system | · counts: 32, 56 |
| top500.org | names system, part and count | NVIDIA Tesla V100, NVIDIA InfiniBand HDR · counts: 32 |
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.