Fusion Grid Sim 300
- Compute nodes
- 16
- Cores
- 1,536
- Accelerators
- 32 × 48GB
What is in it
- Login, head and controller nodes3x
- Fabric switches2xNDR 400Gb/s
- Accelerated compute nodes16x2 accelerators each
What arrives, across 2 cabinets, racked, cabled and burned in before it ships. The nearest single server we sell is the Fusion Rack Fury 300, shown in the photograph, with every part listed.
Why this cluster
Sixteen dual-socket Turin nodes, 1,536 cores on fully filled memory channels. Non-blocking NDR InfiniBand, with SHARP handling collective operations in the network instead of on the hosts. Half a petabyte of Lustre on its own storage network, so a checkpoint does not compete with the solver's own traffic.
Made for your workflow
Your cluster, connected
Explore the hardware, shared capacity and connections that bring your cluster together.
- Compute nodes
- 16 nodes
- 2 cabinets · Engineering simulation
- Multiple machines share the workload across your team or applications.
- CPU cores
- 1,536 cores
- Across the specified nodes
- Processing capacity for parallel jobs, shared services and data workloads.
- Connecting the nodes
- NDR 400Gb/s
- Specified network fabric
- The interconnect moves data between nodes so the cluster can work together.
- Shared storage
- 500TB Lustre
- Specified filesystem capacity
- A shared home for the files, datasets and results your nodes need to access.
Hardware platform
How the cluster comes together
Network fabricNDR 400Gb/s · 2 switches
- Control & access3 login, head and controller nodes
- GPU compute16 accelerated compute nodes
Each square represents one node. Final capacity and cabinet layout are sized to your site.
What it is for
MPI jobs such as ANSYS Fluent and MATLAB split one problem across many machines, and every step waits for the slowest link between them. So the network decides how far a case scales, long before core count does. Memory speed is the second limit, because most scientific codes run out of memory bandwidth before they run out of cores. Built on OpenHPC and Slurm, with the network sized so that scaling is limited by the software, not the hardware.
Both sizes of engineering simulationSpecified
- Compute nodes
- 16
- Cores
- 1,536
- Accelerators
- 32 × 48GB
- Interconnect
- NDR 400Gb/s
- Filesystem
- 500TB Lustre
- Each node
- 2x AMD EPYC 9455, 768GB DDR5 ECC, 2x RTX 6000 Ada 48GB
- Cabinets
- 2
- Deployment
- Rack, in your own room or a colocation
These are the figures the design was specified and tested to, not figures computed from a parts list. A cluster is a network, a filesystem and a node count, and no single build expresses that. The final numbers come from sizing it to your room, your power and your work, and every node in it is measured the usual way.
