Supercomputers/Systems/Laconia
USA · operational · hpc
Laconia
Also known as MSU Laconia
Operated by Michigan State University .
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
- 535.90 TFlop/s
- Rpeak
- 754.71 TFlop/s
- Rmax ÷ Rpeak
- 71.0%
- Cores
- 15,296
- Power
- 198.86 kW
- Per watt
- 2.7 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 Laconia, 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 | 535.90 TFlop/s | #180 of 242 · bottom 27% | #201 of 332 · bottom 41% |
| Energy efficiency | 2.7 GF/W | #64 of 176 · top 36% | #65 of 162 · top 40% |
| Rmax ÷ Rpeak | 71.0 % | #129 of 238 · bottom 47% | #196 of 331 · bottom 42% |
| Power draw | 199 kW | #145 of 178 · bottom 20% | #128 of 171 · bottom 26% |
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 | - Lenovo NeXtScale nx360 M5 | Lenovo | - | Reported | top500.orgdeveloper.nvidia.com |
| Accelerator | - 200 dual-GPU NVIDIA Tesla K80 accelerators in 50 nodes | NVIDIA | 200 reported | Reported | developer.nvidia.comtop500.org |
| CPU | Intel Xeon E5-2680 v4 CPUs x86-64 · 14C · 2.4 GHz · 14 nm · 120 W two Intel Xeon E5-2680v4 14-core 2.4GHz CPUs per node | Intel | - | Reported | developer.nvidia.comtop500.org |
| Interconnect | - InfiniBand EDR/FDR | Mellanox NVDANASDAQ
rolls up to NVIDIA | - | Reported | developer.nvidia.comtop500.org |
The read
Michigan State University's $3 million GPU-accelerated cluster built by its Institute for Cyber-Enabled Research with Lenovo, Intel, NVIDIA and Mellanox. NVIDIA's blog describes 320 nodes in seven racks with two 14-core Xeon E5-2680v4 2.4 GHz CPUs and 128 GB each, plus 50 Lenovo NeXtScale nx360 nodes with four K80 GPUs each (200 dual-GPU K80 boards), 240 GB of local SSD scratch and EDR/FDR InfiniBand. It reports 535.9 TFlop/s Linpack, rank 205 on the June 2016 TOP500 and an expected rank of 30 on the Green500; TOP500 adds 15,296 cores, 754.71 TFlop/s Rpeak and 198.86 kW measured at power level 3. Uses cited include GPU photon tracking for the IceCube neutrino detector, face-search neural networks for biometrics, and molecular simulation for drug design. Listed from June 2016 (rank 206) to June 2017 (rank 381). Tier is reported.
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 Laconia added from its source pages
Nearby in scale
The machines closest to Laconia by Rmax, smaller to larger.
- Sky Bridge · USA, decommissioned · 0.53 PFlop/s
- SANAM · Saudi Arabia, operational · 0.53 PFlop/s
- NEMO bwForCluster · Germany, operational · 0.53 PFlop/s
- KEK SAKURA · Japan, operational · 0.54 PFlop/s
- BrENIAC · Belgium, decommissioned · 0.55 PFlop/s
- SuperMIC · USA, decommissioned · 0.56 PFlop/s
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 |
|---|---|---|
| developer.nvidia.com | names system and part | Intel Xeon E5-2680 v4 |
| top500.org | names system and part | Intel Xeon E5-2680 v4 |
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.