Creative Compute
0

HPE AI-native routers (Juniper PTX Series)

ProductConfigure to orderManufacturer documented

Routers rather than switches: used where AI traffic has to travel between datacentres or between sites, instead of staying inside one room. Built on Juniper's custom silicon following HPE's acquisition of Juniper.

Ask Carl about this

No payment is taken online. Every request is an enquiry answered by an engineer.

Representative system image

Where does this sit?

  1. 01Local
  2. 02Workstation
  3. 03Departmental
  4. 04Enterprise
  5. 05Cluster
  6. 06Rack-scale

Rack-Scale & Supercomputing. Integrated racks where compute, interconnect, storage, power and liquid cooling are engineered as a single system. Typically bought by a datacentre operator, a national facility or a large ai platform. Datacentre with liquid cooling and high-density power.

Fit

Is this the right thing for you?

Who it's for

  • Operators running AI infrastructure across multiple datacentres
  • Service providers building AI-capable backbones
  • Enterprises with distributed AI deployments

What it runs

  • Data-centre interconnect
  • Large-scale AI fabric routing
  • Distributed cloud

When it is the wrong answer

  • Single-site deployments — a switch fabric is what you need
Where it goes
Datacentre or service-provider point of presence
Complexity
specialist
Cooling
Air cooled
Specification

What the numbers mean

Every figure below is explained in plain English. Switch to the technical view for the bare specification.

Platform

Series

Juniper PTX Series, including PTX12000 and PTX10002-36QDD

Silicon

Juniper Express 5 custom silicon

Capability

1.6T-ready platform, high density and throughput

Positioning

Large-scale AI fabrics, distributed cloud and data-centre interconnect

Announcement

Expanded lineup announced at Mobile World Congress, February 2026

Verification

Manufacturer documented — HPE routers page and Juniper PTX router announcement, February 2026. Reviewed 2026-08-31.

Manufacturer source
Before you commit

Practical considerations

The things that catch people out after the hardware has already been ordered.

  • If all your AI infrastructure is in one room, you do not need this. Say so early and save the budget.
Completeness

Nothing arrives working on its own

What else will I need?

  • A network architecture covering both sites
  • Circuit provision between locations
  • Routing expertise, in-house or contracted
AI networkingAI storageInstallationAsk Carl what I’m missing
Goes with

What normally sits alongside it

Relationships documented by the manufacturer, or by us during a deployment.