CLUSTERATLAS
A field guide to supercomputing

FPT AI Factory Japan

FPT AI Factory Japan

CPU Accelerator Interconnect
i
Reading this map

FPT AI Factory Japan node architecture

FPT AI Factory Japan is operated by FPT AI Factory Japan in Japan. This profile covers Ranked system / partition. Specifications are listed below; anything the operator has not published is itemised in the documented-fields list.

Published benchmark (TOP500 · June 2026)

HPL rank
#50
HPL Rmax
49.85 PFlop/s
Theoretical peak Rpeak
67.44 PFlop/s
Ranked core count
146,304

Cluster · HGX H200 · public profile

Operator / site
FPT AI Factory Japan
Country
Japan
Profile scope
Ranked system / partition
Data checked
14 September 2026
Manufacturer
HPE
Ranked core count
146304
Theoretical peak (PFlop/s)
67.44
Operating system
Ubuntu 22.04.5 LTS

HGX H200 · public profile

System / node model
HGX H200
CPU model
Intel Xeon Platinum 8558 · 48 cores
CPU sockets per node
2
Cores per socket
48
Physical CPU cores
96
Accelerator model
NVIDIA H200 SXM5 141 GB
Physical accelerators per node
8
Host / unified memory
2 TB DDR5-4800
Accelerator memory
141 GB HBM3e per GPU
Network interfaces per node
8
Interface speed
400 Gb/s per port (InfiniBand)
Management / storage networking
Two 200 Gb/s BlueField-3 DPU ports per node
Local storage
30 TB (8 × 3.84 TB NVMe SSD)

Scope and uncertainty

The per-node figures are FPT's published Japan H200 node configuration; FPT does not state how many such nodes form the ranked cluster, so no node count is recorded. FPT names NVLink but no generation or aggregate bandwidth.

Documented fields: 17 / 29

Every field this catalog checks. Fields listed above have a value; the remainder are shown here. A field marked absent is not publicly available in the sources we could verify, which does not mean the hardware is missing. Fields that cannot apply, such as benchmark results for an unranked system, are excluded from the total.

Cluster — 4 / 7

  • Manufacturer
  • Ranked core count
  • Theoretical peak (PFlop/s)
  • Measured power (kW)
  • Energy efficiency (GFlop/s per watt)
  • Operating system
  • Network topology

HGX H200 · public profile — 13 / 22

  • System / node model
  • Compute nodes of this type
  • CPU model
  • CPU sockets per node
  • Cores per socket
  • Physical CPU cores
  • CPU note
  • Accelerator model
  • Physical accelerators per node
  • Host / unified memory
  • Accelerator memory
  • Accelerator ↔ accelerator link
  • On-node fabric
  • CPU ↔ accelerator link
  • Network interfaces per node
  • Interface model
  • Interface speed
  • Interface-to-accelerator topology
  • Aggregate scale-out per node
  • GPUDirect RDMA
  • Management / storage networking
  • Local storage
  • NVMe data disks
  • Max remote data disks
  • Uncached disk IOPS
  • Uncached disk throughput (MBps)
  • Form factor

Sources

Browse all cluster architecture pages →

Data & sources