CLUSTERATLAS
A field guide to supercomputing

JUPITER Booster

Jülich Supercomputing Centre

CPU Accelerator Interconnect
i
Reading this map

JUPITER Booster node architecture

JUPITER Booster is operated by Jülich Supercomputing Centre in Germany. This profile covers the documented compute inventory. Specifications are listed below; anything the operator has not published is itemised in the documented-fields list.

Published benchmark (TOP500 · June 2026)

HPL rank
#5
HPL Rmax
1000 PFlop/s
Theoretical peak Rpeak
1226.28 PFlop/s
Ranked core count
4,801,344
Measured power
15794 kW
Energy efficiency
63.32 GFlop/s per watt

Cluster · GH200 booster

Operator / site
Jülich Supercomputing Centre
Country
Germany
Profile scope
Documented compute inventory
Data checked
13 September 2026
Manufacturer
Bull
Ranked core count
4801344
Theoretical peak (PFlop/s)
1226.28
Measured power (kW)
15794
Energy efficiency (GFlop/s per watt)
63.31518298087882
Operating system
RedHat Enterprise Linux
Network topology
Dragonfly+

GH200 booster

Items marked ▧ come from the documented NVIDIA GH200 Grace Hopper Superchip platform specification.

System / node model
Eviden BullSequana XH3000
Compute nodes of this type
5884
CPU model
NVIDIA Grace · 72 cores
CPU sockets per node
4
Cores per socket
72
Physical CPU cores
288
Accelerator model
NVIDIA Hopper (GH200)
Physical accelerators per node
4
Host / unified memory
480 GB LPDDR5X
Accelerator memory
96 GB HBM3 per GPU
Accelerator ↔ accelerator link
NVLink 4 · all GPU pairs
On-node fabric
NVLink 4 — 300 GB/s between GPU pairs (150 GB/s per direction); cNVLink between CPUs at 100 GB/s bi-directional to each of the three neighbours
CPU ↔ accelerator link
NVLink-C2C · 900 GB/s CPU ↔ GPU
Network interfaces per node
4
Interface model
NVIDIA ConnectX-7 (InfiniBand NDR200)
Interface speed
200 Gbit/s per HCA · 800 Gbit/s per node
Interface-to-accelerator topology
A dedicated InfiniBand HCA per GH200 superchip, attached over PCIe Gen 5 to that superchip's Grace CPU

Scope and uncertainty

Booster module only. Each Grace CPU is paired with a Hopper GPU; the Cluster module is outside this profile. Node specifications follow the documented NVIDIA GH200 Grace Hopper Superchip platform (Superchip module — nodes per system and per-node superchip counts are operator-specific); operator-specific deviations are not published.

Documented fields: 24 / 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 — 7 / 7

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

GH200 booster — 17 / 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