SSP, Strategic Supply Partners Live inventory
The desk
GPU procurementNodes and clusters, landed to your rack ● Live inventoryVerified lines, indicative pricing
By platform
B300 and GB300Current-generation Blackwell H200 and H100New stock and fleet rotation All platformsThe full catalogue
Rest of the stack
Memory, servers and storageThe kit around the accelerators Networking and opticsTransceivers, switching, cabling
Capacity without capex
GPU-as-a-serviceReserved, committed spend, on demand Bare-metal leasingDedicated clusters on a term
Decide first
Cloud vs bare metalWhere each one actually wins Reserved vs on demandThe break-even, calculated Cluster cost calculatorThe whole bill, not the GPU line
Space and power
Colocation and powerSingle racks to 100 MW Above 40 kW per rackWhere the shortlist collapses
Where
United States Europe Asia Pacific Middle East and Africa All regions →
Supply side
Have capacity to fill?Tell us what you hold
Analysis
InsightsWhat things cost, with the numbers Sizing guidesMemory, nodes, power, cost ComparisonsBuy, lease or rent
Start here
How an engagement worksThe first week, held to dates Find your fitThree questions, the right desk Glossary
Talk to the desk
Use case · HPC and research

HPC and research

For research institutions and engineering teams: we structure deals to pass formal procurement, precision requirements first. Scientific computing has been running accelerated workloads for far longer than the AI market has existed, and its requirements differ in ways that catch AI-focused suppliers out.

Send your HPC requirement See live inventoryAcknowledged within one hour, first sourcing pass within one business day.
At a glance

What decides this requirement.

Deciding factor
Precision requirements and interconnect topology
Differs from AI
Double precision often matters; AI workloads rarely need it
Procurement
Frequently public sector, with formal processes and audit
Facility
Often existing institutional sites with real power constraints
Precision is the first divergence

AI silicon is not automatically HPC silicon.

A great deal of scientific computing depends on double-precision floating point, where much of the current AI-focused product line is optimised for lower precision formats. A card that dominates AI benchmarks can be a poor value proposition for a simulation code that needs FP64.

That is not a reason to avoid current-generation hardware, but it is a reason to check the requirement before buying it. We ask what the codes actually need rather than assuming the AI answer transfers.

MPI-based codes also stress the interconnect differently from AI training, with communication patterns that reward topology choices AI clusters do not always make.

Research supercomputing racks with visible liquid cooling manifolds
Procurement and the facility

Institutional buying has its own shape.

Research computing is frequently bought through public or institutional procurement, with formal processes, audit expectations and disclosure requirements. We structure deals to pass that review: named counterparties, auditable pricing on direct supplier terms, and documentation built for an auditor rather than a sales meeting. The same discipline as the sovereign desk.

Facilities are often existing institutional buildings rather than purpose-built halls, which makes power and cooling the binding constraint more often than in commercial deployments. We qualify what the room can genuinely take before proposing what goes in it.

Where an upgrade to power or cooling is required, that is part of the same conversation rather than a surprise afterwards. Details on the space and power desk.

Where to next

Start from what is verified.

Current verified lines with quantities, lead times and indicative pricing are public on the live inventory. Anything not listed becomes a sourcing requirement with a first pass inside one business day.

Space and powerSovereign and public sectorFabric and opticsLive inventoryProcurement desk
Tell us what you need See live inventoryAcknowledged within one hour, first sourcing pass within one business day.
Straight answers

Asked first, answered straight.

Is AI hardware suitable for HPC?

Sometimes, and not automatically. Where your codes need double precision, current AI-optimised parts may be poor value. We check what the codes actually require rather than assuming the AI answer transfers.

Can you work with public procurement processes?

Yes. Deals are structured to survive formal review: disclosed relationships, auditable pricing on direct supplier terms and documentation built for audit.

What if our building cannot take the density?

That is common with existing institutional facilities and it is better discovered now. Power and cooling upgrades are part of the same conversation rather than a surprise after delivery.

Do you supply the storage and interconnect too?

Yes. Fabric, storage and the supporting infrastructure come through the same desk, sized to the codes rather than to a generic AI template.

Talk to the desk

Working through this on a real requirement?

Twenty minutes with the desk, no pitch and no quote at the end of it. Tell us roughly what you need and we will come back within one business day.

Acknowledged within one hour, first sourcing pass within one business day.
✓

Sent. We are on it.

Your enquiry has landed with the desk. Acknowledged within one hour.