CLUSTERATLAS
A field guide to supercomputing

Aurora

Argonne National Laboratory

CPU Accelerator Interconnect
i
Reading this map

Aurora node architecture

Aurora is operated by Argonne National Laboratory in United States. 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
#4
HPL Rmax
1012 PFlop/s
Theoretical peak Rpeak
1980.01 PFlop/s
Ranked core count
9,264,128
Measured power
38698 kW
Energy efficiency
26.15 GFlop/s per watt

Cluster · GPU Max compute

Operator / site
Argonne National Laboratory
Country
United States
Profile scope
Documented compute inventory
Data checked
13 September 2026
Manufacturer
Intel
Ranked core count
9264128
Theoretical peak (PFlop/s)
1980.01
Measured power (kW)
38698
Energy efficiency (GFlop/s per watt)
26.151222285389427
Operating system
SUSE Linux Enterprise Server 15 SP4
Network topology
Dragonfly

GPU Max compute

Compute nodes of this type
10624
CPU model
Intel Xeon CPU Max 9470 · 52 cores
CPU sockets per node
2
Cores per socket
52
Physical CPU cores
104
Accelerator model
Intel Data Center GPU Max
Physical accelerators per node
6
Host / unified memory
1,024 GB DDR5 + 128 GB CPU HBM
Accelerator memory
128 GB HBM per GPU
Accelerator ↔ accelerator link
Xe Link
CPU ↔ accelerator link
PCIe
Network interfaces per node
8
Form factor
166 racks · 8 chassis per rack · 8 nodes per chassis (64 nodes per rack) · each rack is one Dragonfly group

Scope and uncertainty

Six physical GPUs per node, each with two tiles. On-node links are summarized; not an exact PCIe switch map.

Documented fields: 20 / 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

GPU Max compute — 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