Tinkercliffs

Also known as TinkerCliffs

Operated by Virginia Tech at Virginia Tech Steger Hall HPC 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: -
Rpeak: -
Rmax ÷ Rpeak: -
Cores: 41,600
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	- 14 nodes with eight NVIDIA A100 80 GB each (112 A100 as stated by ARC)	NVIDIA NVDANASDAQ	112 reported	Reported	docs.arc.vt.edu
Accelerator	NVIDIA H200 SXM Accelerators CUDA (Hopper GH100) · 16,896C · 1.98 GHz · TSMC 4N · 700 W 6 nodes with eight NVIDIA H200 141 GB each (48 H200 as stated by ARC)	NVIDIA NVDANASDAQ	48 reported	Reported	docs.arc.vt.edu
CPU	- 288 base compute nodes with AMD EPYC 7702 Zen 2 (128 cores per node; quantity is nodes)	AMD AMDNASDAQ	288 reported	Reported	docs.arc.vt.edu
CPU	AMD EPYC 7742 (Rome) CPUs x86-64 · 64C · 2.25 GHz · 7 nm · 225 W 14 A100 GPU nodes with AMD EPYC 7742 (quantity is nodes; 10 of them DGX A100)	AMD AMDNASDAQ	14 reported	Reported	docs.arc.vt.edu
CPU	- 16 nodes with Intel Xeon Platinum 9242 (96 cores per node; quantity is nodes)	Intel INTCNASDAQ	16 reported	Reported	docs.arc.vt.edu
Interconnect	NVIDIA InfiniBand HDR Interconnect Fat tree or dragonfly+ · 200 Gb/s per port · About 0.6 microseconds HDR InfiniBand, HDR-100 to 8x HDR-200 depending on node type	NVIDIA NVDANASDAQ	-	Reported	docs.arc.vt.edu
Storage	IBM Spectrum Scale ESS (GPFS) Storage Parallel filesystem (GPFS), erasure-coded declustered RAID · 250 PB on Summit (Alpine) · About 2.5 TB/s IBM ESS GPFS project storage	IBM IBMNYSE	-	Reported	docs.arc.vt.edu
Storage	- VAST NVMe high-performance file system	VAST Data	-	Reported	docs.arc.vt.edu

The read

Virginia Tech Advanced Research Computing's flagship CPU and GPU cluster in the Steger Hall HPC data center, online since summer 2020. The ARC documentation lists 332 nodes, 41,600 CPU cores and 130 TB of RAM: 288 AMD EPYC 7702 nodes (128 cores, 256 GB), 8 high-memory EPYC 7702 nodes (1 TB), 16 Intel Xeon Platinum 9242 nodes, 14 GPU nodes with eight NVIDIA A100 80 GB each (10 of them DGX A100), and 6 nodes with eight NVIDIA H200 141 GB each, for 112 A100 and 48 H200 GPUs. Interconnect is HDR InfiniBand (HDR-100 to 8x HDR-200 depending on node type); storage is IBM ESS GPFS plus VAST NVMe. Both sources are VT pages, so the tier is reported; no HPL result is published.

Timeline

2020 First operational source

Site & facility

Virginia Tech Steger Hall HPC Data Center

Location: Blacksburg, VA, USA

arc.vt.edu docs.arc.vt.edu

Change history

2026-10-02 Added Tinkercliffs (Virginia 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
arc.vt.edu	names system only	-
docs.arc.vt.edu	names system, part and count	AMD EPYC 7742 (Rome), NVIDIA H200 SXM · counts: 14, 48

Checked 2026-10-09 by pnpm verify. Re-run it and the table changes with the web.