Skip to content
Pacific
Pacific

AI infrastructure, made physical.

Thermal systems / Reference design. Illustrative Pacific compute infrastructure design study.

02 / High-density colocation

Your hardware. A considered home.

Bring your GPU hardware. We assess the power, cooling, networking, and operating envelope around it, starting with engineering fit.

Thermal systems / Reference designIllustrative design study

The application

Fit comes first.

Owned hardware deserves more than a generic rack allocation. Every interface is part of the intake.

  1. 01

    Rack and node

    Map node dimensions, rack geometry, and maintenance access against the deployment envelope.

  2. 02

    Power and cooling

    Review the operating envelope and cooling interfaces, including how your nodes map to direct-to-chip liquid cooling.

  3. 03

    Network and storage

    Confirm the topology, fabric configuration, and storage requirements before hardware moves.

Close view of Pacific's illustrative internal infrastructure layout.
Pacific / Design studyWritten specifications govern the configuration

Reference specification

A reference. Not an assumption.

Pod 32 describes the reference infrastructure envelope. Your engineering review determines the configuration and whether your hardware fits.

Explore the technology
Pod 32 reference specification
Compute32 Supermicro HGX B300 nodes
GPUs256 NVIDIA B300
Racks4
Operating load645 kW
System memory128 TB
Storage983 TB raw NVMe per module
FabricRail-optimized 400G Ethernet
Fabric throughputConfigurable 3.2 or 6.4 Tbps per node
CoolingDirect-to-chip liquid, redundant CDUs
Site waterNo site water requirement
TenancySingle-tenant bare metal

Engineering and operations

Clear ownership. Documented handoff.

Managed infrastructure

Pacific operates remotely with contracted regional field service. Hardware, software, maintenance access, and exit responsibilities are defined during discovery.

Evidence at acceptance

Factory, site, and integration checks follow a 24-stage test program. Measured NCCL criteria apply where the network fabric is in scope.

Controlled changes

Configuration changes after handoff follow acceptance control, with evidence. Service responsibilities and terms are defined in the contract.

Start a conversation

A little context. A useful call.

Three short questions, then choose a time. No project names or sensitive details needed.

Question 1 of 3

Your role

Only broad answer categories are recorded. Keep sensitive project details for the conversation.

Book without the questions Open Cal.com in a new tab

Further detail

Before we talk.

What hardware fits?

An engineering intake maps your rack and node dimensions, power envelope, cooling interfaces, network topology, and service access against the infrastructure. Fit is confirmed by that review, not assumed.

Who operates what?

Pacific runs the infrastructure through remote operations and contracted regional field service. Responsibility for your hardware, software stack, and maintenance access is defined explicitly during discovery.

Is liquid cooling required?

The reference environment is direct-to-chip liquid cooling with redundant CDUs and no site water requirement. The fit review determines how your hardware's cooling profile maps onto it.

How is acceptance handled?

Handoff follows factory, site, and integration evidence gates, including measured NCCL fabric criteria where the network fabric is in scope for your deployment.

What are the commercial terms?

Terms are discussed privately on a qualified colocation call. Pacific does not publish rates.