Supercomputers/Systems/Betty
USA · operational · mixed
Betty
Also known as Betty Holberton
Operated by University of Pennsylvania at Betty Data Center (Collegeville) .
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
- 8.49 PFlop/s
- Rpeak
- 9.46 PFlop/s
- Rmax ÷ Rpeak
- 89.7%
- Cores
- 40,176
- 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 | - NVIDIA DGX SuperPOD with DGX B200 systems | NVIDIA | - | Verified | parcc.upenn.edutop500.org |
| Accelerator | NVIDIA B200 SXM Accelerators CUDA (Blackwell) · TSMC 4NP · 1,000 W NVIDIA B200 GPUs, 8 per node on 31 nodes (derived from the node count) | NVIDIA | 248 reported | Reported | parcc.upenn.edutop500.org |
| CPU | Intel Xeon Platinum 8570 CPUs x86-64 · 56C · 2.1 GHz · Intel 7 Intel Xeon Platinum 8570, 2 per DGX B200 node, 31 GPU nodes (derived from the node count) | Intel | 62 reported | Verified | parcc.upenn.edutop500.org |
| Interconnect | NVIDIA InfiniBand NDR (Quantum-2) Interconnect Fat tree or dragonfly+ · 400 Gb/s per port · Under 0.6 microseconds NDR400 InfiniBand compute fabric | NVIDIA | - | Verified | parcc.upenn.edutop500.org |
| Storage | VAST Data Platform Storage Disaggregated shared-everything VAST external storage | VAST Data | - | Reported | parcc.upenn.edu |
The read
The University of Pennsylvania's first university-wide HPC and AI cluster, run by the Penn Advanced Research Computing Center (PARCC) and named for ENIAC programmer Frances Elizabeth "Betty" Snyder Holberton; TOP500 lists it under the site name BETTY-HOLBERTON. PARCC's system page lists an NVIDIA DGX B200 SuperPOD of 31 nodes with eight B200 GPUs and two Xeon Platinum 8570 CPUs each on NDR400 InfiniBand, plus 64 AMD EPYC 9374F standard CPU nodes and 10 large-memory CPU nodes, with VAST and Ceph external storage. Penn Today dates the launch to May 2025 and puts it 189th on the TOP500; TOP500 lists 40,176 cores, Rmax 8.49 PFlop/s and Rpeak 9.46 PFlop/s, with a debut at rank 189 in November 2025 and 212th in June 2026. HPCwire's relay of Daily Pennsylvanian reporting says the machine sits in a data center in Collegeville about 30 miles from campus and that the current iteration needs 1 MW of power. PARCC and TOP500 agree on the DGX B200, Xeon 8570 and NDR400 hardware, so the tier is verified. TOP500 shows RHEL while PARCC lists Ubuntu 24.04 Pro / DGX OS, so no OS edge is recorded.
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
- 2026-10-07 Added Betty (upenn)
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 |
|---|---|---|
| parcc.upenn.edu | names system and part | Intel Xeon Platinum 8570, NVIDIA B200 SXM, NVIDIA InfiniBand NDR (Quantum-2), VAST Data Platform |
| top500.org | names system and part | Intel Xeon Platinum 8570, NVIDIA B200 SXM, NVIDIA InfiniBand NDR (Quantum-2) |
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.