Penguin Solutions and Lektra Collaborate on Distributed EdgeScale AI Micro Data Centers

Penguin Solutions and Lektra are collaborating on the deployment and optimisation of distributed EdgeScale AI infrastructure, rather than concentrating compute capacity in a single large campus.

The EdgeScale model is designed around smaller AI data-center sites in multiple locations, supplied by locally available distributed carbon-free energy resources. Capacity at each individual site is capped at less than 20 MW. Lektra says it has already deployed eight sites and plans further expansion, but has not disclosed their locations, combined capacity, or the operating status of each deployment.

Penguin Solutions is providing NVIDIA-based hardware through its Full-Stack AI Factory Platform, alongside accelerated-computing server design, systems integration, deployment and ongoing operations. Lektra describes its approach of aligning distributed energy generation with computing located where power is available as “Grid 2.0.”

The source characterises EdgeScale as an architecture that can reduce data-center deployment timelines from several years to several months. This should be treated as an architectural claim rather than a disclosed record of delivery performance: no reference project, baseline conditions or site-level results were provided.

For the data-center and power-infrastructure market, the collaboration highlights the potential role of repeatable, modular AI deployments where a single site cannot secure the land, grid capacity or interconnection timetable required for a large campus. Distributed deployments could also lessen the concentrated grid burden associated with one large facility. The model may increase demand for modular infrastructure, local-energy integration and phased GPU expansion, but the available information does not establish the economics, reliability arrangements, generation mix, grid-connection parameters or aggregate scale of Lektra’s network.