oval-border
configure this system contact us
Hyperion X2160 — AxiVector HPC Compute Cluster

design features

Two-node HPC with a real RDMA interconnect

AxiVector is built for CFD, FEA, MPI, mathematical computing and scientific workloads that need CPU density, memory bandwidth and low-latency communication.

  • Two Supermicro AS-4125GS-TNRT / CSE-418G2TS 4U rackmount nodes
  • 256 CPU cores and 2.3TB aggregate ECC DDR5 memory
  • Two professional NVIDIA GPUs with 192GB aggregate VRAM
  • Direct 100GbE RoCEv2 RDMA connection between compute nodes

Hyperion X2160 — AxiVector HPC Compute Cluster description

An engineered departmental compute platform

Hyperion X2160 balances tightly coupled CPU compute with professional GPU offload and fast enterprise NVMe. The complete quotation can include the matched Orion rack, manufacturer-approved support rails, UPS, console and Axiotech commissioning so the two nodes arrive as one validated platform.

The online configurator provides live indicative configuration pricing and validates compute, memory, storage, networking, rack, power and compliance selections together. Final price is subject to component availability, export compliance and formal quotation.

configure this Hyperion system contact us

Contact us at sales@axiotech.co.uk and we’ll be happy to advise.