HPE HDR InfiniBand switches and adapters
200Gb/s InfiniBand switching and adapters for HPC and AI clusters.
- Who it's for
- Organisations building a multi-server GPU cluster
- Scale
- Level 5
Ethernet engineered specifically for AI traffic. It reaches speeds close to InfiniBand while remaining Ethernet, so the networking team you already have can run it without learning a second technology.
No payment is taken online. Every request is an enquiry answered by an engineer.
Where does this sit?
AI Factory & Cluster. Several GPU servers joined by a dedicated fabric with shared storage, operating as one pool of capacity. Typically bought by an enterprise ai platform team or a research institution. Datacentre.
Who it's for
What it runs
When it is the wrong answer
Every figure below is explained in plain English. Switch to the technical view for the bare specification.
Fifth-generation Spectrum AI Ethernet switch
64-port 800GbE OSFP switch
MeansSixty-four ports each capable of 800 gigabits per second.
Management switching within the same portfolio
NVIDIA InfiniBand switch, also offered within the HPE networking portfolio
WhenWhen you have chosen InfiniBand rather than Ethernet.
10GbE up to 800GbE across the portfolio
Verification
Manufacturer documented — HPE NVIDIA Networking collateral a00038966enw and HPE Store SN5610 listing. Reviewed 2026-08-31.
Manufacturer sourceThe things that catch people out after the hardware has already been ordered.
What else will I need?
Relationships documented by the manufacturer, or by us during a deployment.
200Gb/s InfiniBand switching and adapters for HPC and AI clusters.
High-density routing platforms for AI fabrics spanning more than one site.