Supercomputers/Systems/Rice (Purdue)
USA · decommissioned · hpc
Rice (Purdue)
Also known as Rice
Operated by Purdue University at Purdue RCAC (West Lafayette) .
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 1581 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
- 407.14 TFlop/s
- Rpeak
- 549.12 TFlop/s
- Rmax ÷ Rpeak
- 74.1%
- Cores
- 13,200
- Power
- 726 kW
- Per watt
- 0.6 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.
How it compares
Ranked against the other machines in this dataset that report the same figure: those that first came into service within two years of Rice (Purdue), and those in USA. Computed from our tables; a metric is ranked only when at least eight peers report it.
| Metric | Value | Against its era | Against its country |
|---|---|---|---|
| Rmax | 407.14 TFlop/s | #172 of 230 · bottom 26% | #213 of 332 · bottom 37% |
| Energy efficiency | 0.6 GF/W | #168 of 180 · bottom 8% | #125 of 162 · bottom 24% |
| Rmax ÷ Rpeak | 74.1 % | #123 of 226 · bottom 47% | #170 of 331 · bottom 50% |
| Power draw | 726 kW | #73 of 182 · top 40% | #77 of 171 · top 45% |
Every metric is ranked largest-first, so a high power-draw rank means a big power draw, not a good one. Rmax ÷ Rpeak uses HPL-measured results only. Rankings move as machines are added; they describe this dataset, not the world.
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 | - 576 HP compute nodes (Cluster Platform DL60) | Hewlett Packard Enterprise | 576 reported | Reported | top500.orgtop500.org |
| CPU | - Intel Xeon E5-2660v3 10C 2.6GHz | Intel | - | Reported | top500.orgrcac.purdue.edu |
| Interconnect | Mellanox InfiniBand FDR Interconnect Fat tree · 56 Gb/s per 4x port · About 1 microsecond Mellanox 56 Gb/s FDR InfiniBand | Mellanox NVDANASDAQ
rolls up to NVIDIA | - | Reported | top500.orgrcac.purdue.edu |
| Storage | DDN EXAScaler Storage Lustre appliance Lustre parallel file system on DDN SFA12KX EXAScaler | DDN | - | Reported | top500.org |
The read
Purdue's Rice community cluster, named for the computer scientist John R. Rice and the eighth cluster built under the university's Community Cluster Program in eight years. TOP500 News reports that more than 100 staff and volunteers assembled 576 HP compute nodes (two 10-core Xeon E5 per node, 64 GB each) on Friday 8 May, on Mellanox 56 Gb/s FDR InfiniBand with a Lustre file system on DDN SFA12KX EXAScaler storage; an atmospheric-science doctoral student was running storm simulations before it was fully unboxed. The RCAC page gives April 2015 as the installation date and 29 January 2021 as retirement, so the 8 May build day and the RCAC install date are not the same event, and only the year is recorded for the start. TOP500 lists 13,200 cores of Xeon E5-2660v3 10C 2.6 GHz, 407.14 TFlop/s Rmax, 549.12 TFlop/s Rpeak and 726 kW; 576 nodes of 20 cores would be 11,520, and the dataset records TOP500's count. Listed from June 2015 (rank 164) to November 2016 (rank 430). Tier is reported.
Timeline
- 2015 First operational source
- 2016-11 List debut at #430 (November 2016) source
- 2021-01-29 Decommissioned source
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
- 2026-10-17 Rice (Purdue) added from its source pages
Nearby in scale
The machines closest to Rice (Purdue) by Rmax, smaller to larger.
- CNIC Era · China, operational · 0.41 PFlop/s
- Yemoja · Brazil, operational · 0.41 PFlop/s
- Vilje · Norway, decommissioned · 0.4 PFlop/s
- Triolith · Sweden, decommissioned · 0.41 PFlop/s
- Computerome 1 · Denmark, operational · 0.41 PFlop/s
- NETL HPCEE · USA, operational · 0.41 PFlop/s
Related systems
Same site
- Conte · USA, decommissioned, 2013
- Carter (Purdue) · USA, decommissioned, 2011
- Bell (Purdue) · USA, operational, 2020
- Anvil · USA, operational, 2022
- Gautschi AI · USA, operational, 2024
Same operator
- Brown (Purdue) · USA, decommissioned, 2017
- Gilbreth (Purdue) · USA, operational
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 |
|---|---|---|
| rcac.purdue.edu | names system and part | Mellanox InfiniBand FDR |
| top500.org | names system and part | Mellanox InfiniBand FDR |
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.