Supercomputers/Systems/Zaratan
USA · operational · mixed
Zaratan
Also known as UMD Zaratan HPC cluster
Operated by University of Maryland .
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
- -
- Rpeak
- 5.70 PFlop/s
- Rmax ÷ Rpeak
- -
- Cores
- 50,000
- 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 |
|---|---|---|---|---|---|
| Accelerator | - 20 GPU nodes with four NVIDIA A100 40GB GPUs each | NVIDIA | 80 reported | Reported | hpcc.umd.edu |
| Accelerator | - 8 GPU nodes with four NVIDIA H100 80GB GPUs each | NVIDIA | 32 reported | Reported | hpcc.umd.edu |
| CPU | AMD EPYC 7763 (Milan) CPUs x86-64 · 64C · 2.45 GHz · 7 nm · 280 W AMD EPYC 7763 Zen3 64-core CPUs, 360 compute nodes with 128 cores each | AMD | - | Reported | hpcc.umd.edu |
| Interconnect | NVIDIA InfiniBand HDR Interconnect Fat tree or dragonfly+ · 200 Gb/s per port · About 0.6 microseconds HDR/HDR-100 InfiniBand between nodes | NVIDIA | - | Reported | hpcc.umd.edu |
AI datacenter metrics
Figures beyond an accelerator count, each with who stated it. What can and cannot be compared across AI datacenters.
| Metric | Value | As of | Basis | Note | Source |
|---|---|---|---|---|---|
| Accelerators | 112 accelerators | 2026-10 | Operator stated | 80 physical A100 40GB plus 32 H100 80GB per the gres table and node list. | hpcc.umd.edu |
The read
The University of Maryland's flagship HPC cluster, online since August 2022. UMD's page lists 360 compute nodes with 128 cores each (AMD EPYC 7763 Milan), 20 GPU nodes with four A100 40 GB GPUs, 8 GPU nodes with four H100 80 GB GPUs (96 Sapphire Rapids cores each) and 6 large-memory nodes, on HDR/HDR-100 InfiniBand with 2 PB of BeeGFS scratch, and states 5.7 PFlop/s theoretical performance (taken as a peak figure). The gres table counts 32 H100 and 80 physical A100 GPUs; a UMD news article of December 2022 cites about 50,000 cores, 360 liquid-cooled nodes and 80 GPUs.
Timeline
- 2022-08 First operational source
Change history
- 2026-10-07 Added Zaratan (University of Maryland)
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 |
|---|---|---|
| hpcc.umd.edu | could not be fetched | - |
| umdrightnow.umd.edu | names system only | - |
Checked 2026-10-09 by pnpm verify. Re-run it and the table changes with the web.