Supercomputers/Systems/SciPhi-XVI
USA · operational · hpc
SciPhi-XVI
Also known as Jefferson Lab 16p
Operated by Jefferson Lab at Jefferson Lab Newport News .
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
- 425.87 TFlop/s
- Rpeak
- 702.87 TFlop/s
- Rmax ÷ Rpeak
- 60.6%
- Cores
- 16,896
- Power
- 111 kW
- Per watt
- 3.8 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 | - KOI Computers cluster of 264 Xeon Phi nodes | Koi Computers | - | Reported | top500.orgjlab.org |
| CPU | - Intel Xeon Phi 7230 64C 1.3GHz Knights Landing, one per node, 264 nodes | Intel | 264 reported | Verified | jlab.orgtop500.org |
| Interconnect | Omni-Path Architecture Interconnect Fat tree · 100 Gb/s per port · About 1 microsecond Intel Omni-Path Architecture, 100 Gb/s | Cornelis Networks | - | Verified | jlab.orgtop500.org |
The read
SciPhi-XVI is Jefferson Lab's lattice QCD cluster of 264 Intel Xeon Phi (Knights Landing) nodes with 64 cores each (16,896 cores) on Intel Omni-Path, built by KOI Computers. Jefferson Lab says it was installed in mid-August 2016, deployed on 30 September with 200 nodes, grew by 64 nodes in late October and benchmarked 425.8 TFlop/s on 256 nodes. The lab's release says 397th on the November 2016 list while the TOP500 list page shows 398th, which is the value used here. TOP500 gives 0.70 PFlop/s Rpeak and 111 kW. It appeared only on that one list.
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 SciPhi-XVI, a mid-list TOP500 system (2016-11 398)
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 |
|---|---|---|
| top500.org | names system and part | Omni-Path Architecture |
| jlab.org | names system and part | Omni-Path Architecture |
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.