Aircloud Compute is now Lightyard. Aircloud continues as a private inference platformAircloud continues for inference

Hardware · Previous generation

NVIDIA HGX B200

We build and run clusters of HGX B200, the eight-GPU Blackwell system before B300, from new systems or ones you already own.

Sheet 01 · HGX B2008 GPUs per node
GPUs
8 × B200
GPU memory
180 GB HBM3e each
Network
400 Gb/s per GPU
Power
Up to 14.3 kW per system
Cooling
Air, front to back
Drawing is illustrative · Figures are NVIDIA's
Specifications

NVIDIA HGX B200 specifications.

One system on NVIDIA's figures, with notes where it differs from B300.

Generation
BlackwellThe generation before Blackwell Ultra, which HGX B300 uses.
GPUs
8 GPUs per nodeAll eight share one NVLink domain.
GPU memory
180 GB HBM3e per GPU, 1.44 TB per system108 GB less per GPU than HGX B300. It sets how large a model one system can hold.
NVLink
14.4 TB/s across 8 GPUs1.8 TB/s per GPU, unchanged in HGX B300.
Scale-out network
400 Gb/s per GPU (ConnectX-7)Against 800 Gb/s per GPU on HGX B300. Less bandwidth for jobs that span systems, and a lighter fabric to build.
Power
Up to 14.3 kW per systemNVIDIA gives it as a ceiling, not a typical draw.
Cooling
AirNo facility water at the rack. Rooms built for air can take it.
Air-cooled rack
Set in the designThe count per rack follows what each position can feed and cool.
Reference design
32 nodesThe scalable unit in NVIDIA's reference architecture for B200.
Lead time
Set in the proposalFor systems you already hold, the schedule depends on the site and the network, not on supply.

Figures are NVIDIA's: its specifications for the eight-GPU B200 system and its B200 reference architecture.

Site requirements

Fitting B200 into an air-cooled hall.

An example rack of four systems, keyed to the notes. The design sets the real count per rack.

  1. 1

    Feeds sized to NVIDIA's maximum

    We size each position to NVIDIA's figure, up to 14.3 kW per system, rather than to typical draw. Every system takes power from an A and a B strip on separate circuits.

  2. 2

    Room air, no facility water

    The fan wall draws cold-aisle air in at the front and exhausts it at the back. That's why B200 fits rooms built for air, provided the hot aisle is contained.

  3. 3

    ConnectX-7 links to the leaf switches

    One per GPU at 400 Gb/s, so eight fabric links leave each system. On systems you already own, we design the fabric around the NICs they came with.

  4. 4

    Four systems, up to 57.2 kW

    The drawing's count times NVIDIA's per-system maximum, as an example. If your positions can't feed that, we put fewer systems in each rack.

Choosing a generation

HGX B200 or HGX B300?

Two eight-GPU systems on air, at about the same power. B300 adds memory and scale-out bandwidth per GPU; B200 makes sense where the models fit and the systems are already bought.

HGX B200 or HGX B300?
HGX B300HGX B200
GenerationBlackwell UltraBlackwell
GPU memory288 GB HBM3e per GPU180 GB HBM3e per GPU
NVLink1.8 TB/s per GPU1.8 TB/s per GPU
Scale-out network800 Gb/s per GPU (ConnectX-8)400 Gb/s per GPU (ConnectX-7)
PowerAbout 14 kW per systemUp to 14.3 kW per system
Lead timeAbout 10 to 14 weeks from signingSet in the proposal
Fits bestNew builds, larger models and long contextModels that fit in 1.44 TB per system, and systems you already own

NVIDIA's figures for both. If you already hold B200 systems, this table is about what to add next.

Other platforms

Other NVIDIA platforms we build on.

The same four stages on each: design, source, build and operate. All hardware.

Questions

HGX B200 questions.

We build HGX B200 clusters as a deployment your team runs, a managed cluster you own and we operate, or a reserved cluster we own. Other platforms.

Get started

Tell us where you're starting.

Capital, hardware, a site or a team: tell us what you have and when you need it running. We usually reply within a few hours and come back with a written plan.

sales@lightyard.ai