Supercomputers/Systems/SAKURA Apollo 6500 V100 Cluster
Japan · operational · mixed
SAKURA Apollo 6500 V100 Cluster
Also known as SAKURA internet Tesla V100 cloud supercomputer (2019)
Operated by SAKURA internet at SAKURA internet (Ishikari) .
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
- 3.71 PFlop/s
- Rpeak
- 5.37 PFlop/s
- Rmax ÷ Rpeak
- 69.2%
- Cores
- 65,208
- 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 SAKURA Apollo 6500 V100 Cluster, 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 | 3.71 PFlop/s | #100 of 230 · top 43% | #42 of 131 · top 32% |
| Rmax ÷ Rpeak | 69.2 % | #92 of 224 · top 41% | #83 of 130 · bottom 37% |
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 | - 96 HPE Apollo 6500 Gen10 System servers | Hewlett Packard Enterprise | 96 reported | Verified | it.impress.co.jptop500.org |
| Accelerator | NVIDIA Tesla V100 Accelerators CUDA (Volta GV100) · 5,120C · 1.53 GHz · 12 nm · 300 W 768 NVIDIA Tesla V100 GPUs (SXM2) | NVIDIA | 768 reported | Reported | it.impress.co.jptop500.org |
| CPU | - Intel Xeon Gold 6150 18C 2.7GHz | Intel | - | Verified | it.impress.co.jptop500.org |
| Interconnect | Mellanox InfiniBand EDR Interconnect Fat tree · 100 Gb/s per 4x port · About 0.9 microseconds Mellanox InfiniBand EDR | NVIDIA | - | Verified | it.impress.co.jptop500.org |
| Operating system | CentOS Operating systems Linux distribution, RHEL-compatible rebuild · Open source, community-maintained CentOS | The CentOS Project | - | Reported | top500.org |
The read
SAKURA internet's 2019 cloud-type GPU supercomputer at its Ishikari Data Center in Hokkaido, built with a partner that the Japanese press article writes as Planus Solutions and TOP500 spells Prunus Solutions, so the English spelling of the partner is unsettled. The Impress IT report of 18 June 2019 (the day the June 2019 TOP500 appeared) describes 96 HPE Apollo 6500 Gen10 servers with 768 NVIDIA Tesla V100 GPUs, Xeon Gold 6150 CPUs and Mellanox InfiniBand EDR, 3,712 TFlop/s Linpack against 5,365 TFlop/s peak, 54th in the world and 8th in Japan, offered for research use. TOP500 (system 179696) agrees on the figures (65,208 cores, 3.712 PFlop/s Rmax, 5.365 PFlop/s Rpeak; the processor is written 6150/6152 in the system name) and shows it still listed in June 2025 at 309th and November 2025 at 339th, which is why status_as_of is 2025-11. No power figure is published for the entry. It is the V100 generation that preceded the H100-class SAKURAONE family (see sakuraone and sakura-koukaryoku-h200-container) at the same operator and is unrelated to the PEZY-based sakura-pezy machine.
Timeline
- 2019 First operational source
- 2019-06-18 Announced source
- 2025-06 List debut at #309 (June 2025) 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
- 2026-10-29 Added SAKURA Apollo 6500 V100 Cluster
Nearby in scale
The machines closest to SAKURA Apollo 6500 V100 Cluster by Rmax, smaller to larger.
- Ruby (LLNL) · USA, operational · 3.7 PFlop/s
- Orion (MSU HPC2) · USA, operational · 3.67 PFlop/s
- IARA · Brazil, operational · 3.66 PFlop/s
- Pratyush · India, operational · 3.76 PFlop/s
- Delta · USA, operational · 3.81 PFlop/s
- hotlum · France, operational · 3.81 PFlop/s
Related systems
Same site
- Koukaryoku PHY H200 Container DC · Japan, operational, 2025
- SAKURAONE CYB · Japan, operational, 2025
- SAKURAONE CYC · Japan, operational, 2025
- SAKURAONE CYD · Japan, operational, 2026
- SAKURAONE · Japan, operational
Same accelerator
- Artemis (G42) · United Arab Emirates, operational, 2019
- Barbora · Czechia, operational, 2019
- Béluga · Canada, operational, 2019
- Cannon (FASRC) · USA, upgraded, 2019
- cHARISMa · Russia, operational, 2019
- Christofari · Russia, operational, 2019
Further reading
Operating a supercomputer
Acceptance testing, why "installed" and "in production" are six months apart, and the machine lifecycle.
Inside a node
Why accelerators exist, what the memory hierarchy costs, and how the CPU and accelerator merged onto one package.
The interconnect
Topologies, why latency and tail behaviour matter more than bandwidth, and how the fabric market consolidated.