CLUSTERCluster · A100 compute
- Operator / site
- Argonne Leadership Computing Facility
- Country
- United States
- Profile scope
- Documented compute profile
- Data checked
- 14 September 2026
- Operating system
- SUSE Linux Enterprise Server 15 SP7
Polaris
Argonne Leadership Computing Facility
CPU Accelerator Interconnect
i
Reading this map
Polaris node architecture
Polaris is operated by Argonne Leadership Computing Facility in United States. This profile covers Documented compute profile. Specifications are listed below; anything the operator has not published is itemised in the documented-fields list.
A100 compute
- System / node model
- HPE Apollo 6500 Gen 10+
- Compute nodes of this type
- 560
- CPU model
- AMD EPYC 7543P · 32 cores
- CPU sockets per node
- 1
- Cores per socket
- 32
- Physical CPU cores
- 32
- Accelerator model
- NVIDIA A100
- Physical accelerators per node
- 4
- Host / unified memory
- 512 GiB DDR4
- Accelerator memory
- 40 GiB per GPU
- Accelerator ↔ accelerator link
- NVLink · all GPU pairs
- On-node fabric
- NVLink — four bonded links between every GPU pair (NV4 in the operator affinity matrix), 600 GB/s
- CPU ↔ accelerator link
- PCIe 4.0
- Network interfaces per node
- 2
- Local storage
- 2 × 1.6 TB SSD in RAID0
- Form factor
- 2 nodes per chassis · 7 chassis per rack · 40 racks
Scope and uncertainty
Compute nodes only; login, gateway and service nodes are excluded.
Documented fields: 17 / 25
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 — 1 / 3
- Manufacturer
- Ranked core count
- Theoretical peak (PFlop/s)
- Measured power (kW)
- Energy efficiency (GFlop/s per watt)
- Operating system
- Network topology
A100 compute — 16 / 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