Traffic model
Separate internet, file transfer, storage access and GPU communication. Identify east-west traffic, peaks and uplink bottlenecks before comparing port speeds.
Design switching, optics and operations from traffic patterns and topology.
Devices / compute nodes → Access switching → Aggregation / fabric → Storage & services
These planning and selection methods support discussion. Adapt the final configuration to workloads, site conditions and validation results.
Separate internet, file transfer, storage access and GPU communication. Identify east-west traffic, peaks and uplink bottlenecks before comparing port speeds.
Plan access/aggregation or leaf-spine topology and separate business, management and storage networks. VLANs require appropriate routing and access-control policies.
Check switches, NICs, optics, fibre and reach. Validate RoCE with congestion control, PFC/ECN and switch buffering; high-speed ports alone are insufficient.
Measure throughput, latency, loss and failover; retain configuration backups and port maps. AI clusters also require collective-communication and application tests.
Provide current equipment and versions, key applications and workloads, users or concurrency, data volume and growth, site power and networking, budget, timing and critical service windows.
Requirements brief, configuration and BOM, compatibility records, test conditions and acceptance checklist. Agree the actual scope of installation, migration, maintenance and on-site services during project confirmation.