Supercomputers/Systems/Beacon (NICS)
USA · operational · hpc
Beacon (NICS)
Also known as Beacon
Operated by NICS .
Researched with AI assistance, checked by scripts. These rows were researched by AI agents reading public operator, laboratory and vendor sources, appended by scripts and checked by automated validation; no subject-matter expert has reviewed them (how this site is made). A mechanical second-reader pass has since fetched every cited source: 1350 of 1540 systems have a readable citation that names them, and 64 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
- 110.50 TFlop/s
- Rpeak
- 157.55 TFlop/s
- Rmax ÷ Rpeak
- 70.1%
- Cores
- 9,216
- Power
- 45.11 kW
- Per watt
- 2.4 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 | - Appro GreenBlade GB824M (Cray CS300-AC) | Appro HPENYSE
rolls up to Hewlett Packard Enterprise | - | Reported | top500.orgcug.org |
| Accelerator | - Intel Xeon Phi 5110P, four per compute node, 48 compute nodes | Intel | 192 reported | Reported | top500.orgcug.org |
| CPU | Intel Xeon E5-2670 CPUs x86-64 · 8C · 2.6 GHz · 32 nm · 115 W Xeon E5-2670 8C 2.6GHz, 768 conventional cores | Intel | 96 estimated | Reported | top500.orgcug.org |
| Interconnect | Mellanox InfiniBand FDR Interconnect Fat tree · 56 Gb/s per 4x port · About 1 microsecond FDR InfiniBand | Mellanox NVDANASDAQ
rolls up to NVIDIA | - | Reported | top500.orgcug.org |
The read
Xeon Phi test cluster of the University of Tennessee Application Acceleration Center of Excellence at NICS, funded by an NSF STCI grant and the State of Tennessee. The NICS paper at CUG 2013 says it started as a 16-node, two-MIC-per-node cluster in Aug 2012 and that the production machine came in Apr 2013: 48 compute nodes, 6 I/O nodes, 2 login nodes, each compute node with Xeon E5-2670 CPUs, 256 GB RAM, four Xeon Phi 5110P cards, 960 GB of SSD and FDR InfiniBand; 768 conventional cores plus 11,520 coprocessor cores, over 210 TF combined. It topped the Nov 2012 Green500 at 2.449 GFLOP/W per the paper (press releases quote 2.499). TOP500 lists it as an Appro GreenBlade GB824M (the paper calls it a Cray CS300-AC) with 9,216 cores, Rmax 110.5 TF, Rpeak 157.55 TF and 45.11 kW, rank 253 in Nov 2012 and 398 in Jun 2013. Source discrepancy: TOP500 shows the 48-node configuration already in Nov 2012, while the paper dates production deployment to Apr 2013. The site is not recorded because the fetched pages do not state where Beacon was housed.
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
- 2026-10-17 Beacon (NICS) added from its source pages
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 |
|---|---|---|
| cug.org | does not name this system | · counts: 96 |
| top500.org | names system, part and count | Intel Xeon E5-2670, Mellanox InfiniBand FDR · counts: 96 |
Checked 2026-10-18 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.