Supercomputers/Systems/AFI-NITY II
Japan · operational · hpc
AFI-NITY II
Also known as Integrated Supercomputation System (2024 update)
Operated by Institute of Fluid Science, Tohoku University at Tohoku IFS Sendai .
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
- 2.82 PFlop/s
- Rpeak
- 3.18 PFlop/s
- Rmax ÷ Rpeak
- 88.7%
- Cores
- 49,728
- 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.
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 AFI-NITY II, and those in Japan. 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 | 2.82 PFlop/s | #215 of 234 · bottom 9% | #49 of 131 · top 37% |
| Rmax ÷ Rpeak | 88.7 % | #30 of 231 · top 13% | #27 of 130 · top 20% |
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 | - HPE Cray XD2000 nodes (centre) / HPE Cray XD220v (TOP500) | Hewlett Packard Enterprise | - | Verified | ifs.tohoku.ac.jptop500.org |
| CPU | - Intel Xeon Platinum 8480+ 56C 2.0GHz | Intel | - | Reported | top500.org |
| Interconnect | NVIDIA InfiniBand HDR Interconnect Fat tree or dragonfly+ · 200 Gb/s per port · About 0.6 microseconds InfiniBand HDR200 | NVIDIA | - | Reported | top500.org |
| Operating system | Rocky Linux Operating systems RHEL-compatible Linux, community-maintained · Downstream rebuild of Red Hat Enterprise Linux source Rocky Linux | Rocky Enterprise Software Foundation | - | Reported | top500.org |
The read
The 2024 generation of the Integrated Supercomputation System at Tohoku University's Institute of Fluid Science, branded AFI-NITY II by its Advanced Fluid Information Research Center. The centre's hardware page splits a 4.4 PFlop/s total peak into 3.2 PFlop/s for the distributed memory part (135.5 TB of memory), 0.2 PFlop/s for the shared memory part (HPE Superdome Flex 280) and 0.8 PFlop/s for a fluid-data analysis server. TOP500's entry (HPE Cray XD220v, Xeon Platinum 8480+ 56C 2GHz, InfiniBand HDR200, Rocky Linux; 49,728 cores, 2.82 PFlop/s Rmax, 3.18 PFlop/s Rpeak) lines up with the distributed memory part only, and the shared memory and analysis parts are not in those figures. The centre cites 447 HPE Cray XD2000 nodes while 49,728 cores is exactly 444 nodes of 2x56 cores, so the benchmarked node count appears slightly smaller than the installed count. The centre dates the update to August 2024; TOP500 gives installation year 2024 and ranks the entry 424th in June 2025 and 453rd in November 2025). It replaces the 2018 Fujitsu PRIMERGY AFI-NITY (see afi-nity-tohoku-ifs), which the centre's older page rated at 3.7 PFlop/s peak with 19 PB of external storage; the new page keeps the 19 PB figure. TOP500 does not list a power figure for this entry.
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
- 2026-10-29 Added AFI-NITY II
Nearby in scale
The machines closest to AFI-NITY II by Rmax, smaller to larger.
- SuperMUC Phase 2 · Germany, decommissioned · 2.81 PFlop/s
- JFRS-1 · Japan, operational · 2.79 PFlop/s
- Volkswagen LX HPC 224Rk-1 · Norway, operational · 2.78 PFlop/s
- Freeman (ERDC) · USA, operational · 2.83 PFlop/s
- Henri · USA, operational · 2.88 PFlop/s
- SuperMUC · Germany, decommissioned · 2.9 PFlop/s
Related systems
Same site
- AFI-NITY · Japan, operational, 2018
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.