DG Matrix, the global leader in solid-state transformer solutions, today unveiled at the OCP Global Summit the technologies that make Interport™ 360 Flex. It’s a complete, skid-mounted power solut...

A new solid-state transformer building block and real-time power-fabric software connect a single SST cell to a complete, skid-mounted power solution for modular datacenters. The Interport SST Cell is featured on the NVIDIA MGX AI Factory wall at OCP Global Summit
SAN JOSE, Calif.: DG Matrix, the global leader in solid-state transformer solutions, today unveiled at the OCP Global Summit the technologies that make Interport™ 360 Flex. It’s a complete, skid-mounted power solution for modular datacenters and next-generation AI factories based on the NVIDIA MGX architecture, deployable from a single building block. This includes the Interport SST Cell, the modular solid-state transformer (SST) at the core of the system; and Interport OS, the real-time software layer that operates and optimizes the power fabric.
Together, they give AI factory operators a path to transition from conventional AC distribution to NVIDIA 800 VDC architecture while protecting the land, power and shell investments they have already made. The Interport SST Cell is featured on the NVIDIA MGX AI Factory wall at the Summit, which runs Oct. 12–15 in San Jose, as a new power-infrastructure component for the NVIDIA MGX modular reference architecture and NVIDIA 800 VDC architecture.
AI factories are now built the way servers are: from standardized, modular building blocks that can be specified, simulated and deployed at scale. Power infrastructure must evolve to support the compute density and performance requirements of future AI factories. Conventional electrical rooms are still assembled stage by stage from discrete transformers, switchgear, UPS systems and rectifiers. Each are engineered on site and none designed for the rapid, cluster-wide changes in power demand as GPUs move between computing and communication or MGX architecture-compatible 800 VDC power infrastructure.
Operators that have secured land, power and shell for AC-distributed facilities face a choice between re-engineering that infrastructure for 800 VDC or leaving it stranded. Today’s announcement gives the power side of the AI factory the same modular path as the compute side, and a way to bring existing sites along: a skid-mounted system that accepts the AC power a site already has and delivers native 800 VDC to the white space, and software that turns it into an intelligent power fabric.
Interport 360 Flex: the complete, skid-mounted power solution. Interport 360 Flex is DG Matrix’s end-to-end power solution for modular datacenters, delivered as outdoor grey-space skids that can be paralleled as capacity grows. It integrates power conversion, energy storage for smoothing rapid changes in AI workload power demand, grid resources, on-site generation and controls into a single, software-defined stage that converts, routes and controls electrons between AC and DC sources, storage and compute loads and delivers native 800 VDC to higher-density AI compute racks.
Because the 34.5 kVAC-capable Interport 360 Flex takes in the medium- or low-voltage AC power an existing site already has and outputs 800 VDC, the system converts existing AC power to 800 VDC, enabling operators to introduce higher-voltage DC distribution while continuing to use existing land, power and facility investments. The key component of the system is the Interport SST Cell, the modular, galvanically isolated, multi-port power-conversion building block from which every Interport 360 Flex skid is assembled. Built on silicon-carbide power semiconductors with liquid cooling designed for the higher density of accelerated computing racks, the same cell featured on the NVIDIA MGX Ecosystem wall at OCP Global Summit scales, cell by cell, into the complete Interport 360 Flex system.
Interport OS: the real-time operating layer. Interport OS is the real-time operating and optimization layer for the physical power fabric. It coordinates power-routing scenarios across cells, ports, storage and sources within sub-millisecond timescales, and delivers as software the functions that legacy datacenters buy as separate equipment: UPS functionality, voltage and frequency ride-through, smoothing rapid changes in AI workload power demand, STATCOM grid-support functions, programmable protection, microgrid control and AC/DC load optimization. Interport OS is what makes a skid of SST cells behave as one programmable system rather than a collection of converters.
“At GTC we showed the industry that a solid-state transformer could deliver native 800 VDC power for the NVIDIA MGX architecture. At OCP we are showing how it becomes an entire AI factory’s power infrastructure,” said Haroon Inam, CEO and co-founder of DG Matrix. “Interport 360 Flex is the product our customers deploy, and the Interport SST Cell is the building block that makes it modular. Interport OS is the intelligence that coordinates power routing, ride-through, grid support, microgrid operation and smoothing rapid changes in AI workload power demand across the fabric in real time. Operators have made enormous investments in land, power and shell. We help them carry those investments into the 800 VDC era instead of stranding them. Speed to power is speed to inference, and this is how modular datacenters get both.”
NVIDIA MGX is an open modular reference architecture that enables ecosystem partners to build accelerated systems from single-node servers to rack-scale AI factories, and the NVIDIA 800 VDC power architecture reduces conversion stages, moves direct-current power closer to the rack and supports higher-density accelerated computing. NVIDIA curates the MGX Ecosystem wall at OCP Global Summit to illustrate hardware and technology contributions across the MGX ecosystem, including compute, networking, power, cooling, storage and controls. The Interport SST Cell is featured on the wall as the enabler of DG Matrix’s SST-based power fabric for MGX and 800 VDC AI factories, demonstrating how a single cell scales into the broader Interport 360 Flex system.
Today’s announcement builds on DG Matrix’s reveal of the Interport SST designed for AI factory infrastructure based on the NVIDIA MGX architecture at NVIDIA GTC 2026. Since the company’s founding in 2023, DG Matrix has worked closely with NVIDIA on technical alignment around 800 VDC architectures and MGX architecture compatibility, from prototype through the commercial launch of its Interport products, and has completed a yearlong program testing energy storage solutions for smoothing rapid changes in AI workload power demand in collaboration with NVIDIA.
DG Matrix operates the largest solid-state transformer engineering team of any company in the world, giving the Interport platform a multi-year lead on SST hardware, firmware and AI-driven control software. McKinsey estimates that global demand for datacenter capacity could almost triple by 2030, with AI-related capacity demand reaching 156 gigawatts. Research from SemiAnalysis projects that approximately 39 gigawatts of incremental AI datacenter capacity will be served by 800 VDC architecture by 2030, with a solid-state transformer total addressable market of approximately $13 billion.
DG Matrix will demonstrate Interport 360 Flex, the Interport SST Cell and Interport OS at OCP Global Summit, Oct. 12–15 at the San Jose McEnery Convention Center at booth S1: Grand Ballroom 220A. The Interport SST Cell can be seen on the NVIDIA MGX Ecosystem wall throughout the summit.
About DG Matrix
DG Matrix, the global leader in solid-state transformer solutions, is unlocking power as the binding constraint on AI. It is removing this critical bottleneck of stranded power by building the intelligent power fabric of the AI datacenter. Built on the world’s first commercially available, programmable multi-port solid-state transformer, the company’s AI-enabled Interport™ platform converts, routes and controls every electron, between the grid, on-site generation, storage and compute loads through a single, software-defined stage. DG Matrix delivers an integrated and simpler solution for faster deployment, lower energy costs, native 800 VDC readiness and protection from GPU pulse loads. Because speed to power is speed to inference, DG Matrix helps hyperscale, neocloud and colocation operators bring AI capacity online in months instead of years - at scale, anywhere in the world. Learn more at https://www.dgmatrix.com/.
Fonte: Business Wire
Alaa Abdul Nabi, Vice President, Sales International at RSA presents the innovations the vendor brings to Cybertech as part of a passwordless vision for…
G11 Media's SecurityOpenLab magazine rewards excellence in cybersecurity: the best vendors based on user votes
Always keeping an European perspective, Austria has developed a thriving AI ecosystem that now can attract talents and companies from other countries
Successfully completing a Proof of Concept implementation in Athens, the two Italian companies prove that QKD can be easily implemented also in pre-existing…
#Restaurants--SpotOn, a leading restaurant technology provider, is proud to partner with Chef Francis Ang on Masters of Filipino Flavors, a four-night…
#WeConnect--Credo Technology Group Holding Ltd (Credo) (NASDAQ: CRDO), an innovator in providing connectivity at scale through fast, reliable, and energy-efficient…
Lenovo, the global technology powerhouse, today announced that its annual global innovation event will return to CES® 2027 in Las Vegas. Hosted at Sphere…
#3DBoneMorphology--MRI2CT Inc. today announced FDA 510(k) clearance for NextMR, its AI powered imaging platform, for generating accurate 3D bone visualization…