ARCHER2 will be an HPE Cray EX supercomputing system with an estimated peak performance of 28 PFLOP/s. The machine will have 5,848 compute nodes, each with dual AMD EPYC Zen2 (Rome) 64 core CPUs at 2.2GHz, giving 748,544 cores in total.

ARCHER2 should be capable, on average, of over eleven times the science throughput of its predecessor, ARCHER. This estimate is based on benchmarks of five of the most heavily used research software packages on ARCHER. As with all new systems, the relative speedups over ARCHER vary by software and problem size. The research software applications used for the benchmarking evaluation are estimated to improve science throughput by the following factors: 8.7x for CP2K, 9.5x for OpenSBLI, 11.3x for CASTEP, 12.9x for GROMACS, and 18.0x for the HadGEM3 climate model.

Hardware details

Compute nodes

Nodes 5,848 nodes: 5556 standard memory, 292 high memory
Processor 2× AMD Zen2 7742, 2.25 GHz, 64-core
Cores per node 128 (2× 64-core processors)
NUMA structure 8 NUMA regions per node (16 cores per NUMA region)
Memory per node 256 GiB (standard memory), 512 GiB (high memory)
Memory per core 2 GiB (standard memory), 4 GiB (high memory)
Interconnect HPE Cray Slingshot, 2× 100 Gbps bi-directional per node

Login and data analysis nodes

Processor 2× AMD Zen2 7742, 2.25 GHz, 64-core
Cores per node 128 (2× 64-core processors)
NUMA structure 8 NUMA regions per node (16 cores per NUMA region)
Memory per node 512 GiB
Memory per core 4 GiB

Storage

home File Systems 1 PB NetApp NFS Available on login and data analysis nodes
work File Systems 14.5 PB HPE Cray ClusterStor Available on login, data analysis and compute nodes

Software details

Information on software packages and libraries installed by the ARCHER2 CSE team at EPCC can be found in the ARCHER2 Documentation.