Frontier

Oak Ridge National Laboratory

Node type

CPU Accelerator Interconnect

Reading this map

Frontier node architecture

Frontier is operated by Oak Ridge 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: #3
HPL Rmax: 1353 PFlop/s
Theoretical peak Rpeak: 2055.72 PFlop/s
Ranked core count: 9,066,176
Measured power: 24607 kW
Energy efficiency: 54.98 GFlop/s per watt

Cluster · MI250X compute

Operator / site: Oak Ridge National Laboratory
Country: United States
Profile scope: Documented compute inventory
Data checked: 13 September 2026
Manufacturer: HPE
Ranked core count: 9066176
Theoretical peak (PFlop/s): 2055.72
Measured power (kW): 24607
Energy efficiency (GFlop/s per watt): 54.98435404559678
Operating system: HPE Cray OS
Network topology: Dragonfly

MI250X compute

Compute nodes of this type: 9856
CPU model: AMD optimized EPYC · 64 cores
CPU sockets per node: 1
Cores per socket: 64
Physical CPU cores: 64
Accelerator model: AMD Instinct MI250X
Physical accelerators per node: 4
Host / unified memory: 512 GB DDR4
Accelerator memory: 128 GB HBM2e per accelerator
Accelerator ↔ accelerator link: Infinity Fabric
On-node fabric: Infinity Fabric — 200 GB/s between the two GCDs of one MI250X; 50–100 GB/s between GCDs on different MI250X depending on link count
CPU ↔ accelerator link: Infinity Fabric · 36+36 GB/s CPU ↔ GCD
Network interfaces per node: 4
Interface model: HPE Slingshot 11
Interface speed: 200 Gb/s (25 GB/s) per NIC · 800 Gb/s node injection
Local storage: 2 × 1.92 TB NVMe SSD
Form factor: 77 Olympus HPE Cray EX cabinets, 128 nodes per cabinet

Scope and uncertainty

Each MI250X has two GPU compute dies: four physical accelerators appear as eight logical GPU devices. Diagram counts physical accelerators.

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

MI250X compute — 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 →