d-Matrix Acquires GigaIO to Tackle AI's System-Level Challenge

SANTA CLARA, CA – April 02, 2026 – In a strategic move signaling a major shift in the AI hardware landscape, d-Matrix, a pioneer in low-latency AI inference, today announced the acquisition of GigaIO's data center business. The deal brings GigaIO's deep expertise in rack-scale systems and its innovative interconnect technology in-house, positioning d-Matrix to address what its leadership calls the primary bottleneck in scaling artificial intelligence: the system itself.

This acquisition is more than a simple expansion; it's a calculated response to the evolving demands of modern AI. As models grow in complexity, the industry is realizing that the performance of a single accelerator chip is only one piece of a much larger puzzle. By integrating GigaIO's technology, d-Matrix is making a clear statement that the future of AI hinges on building cohesive, highly-efficient, rack-scale systems.

Beyond the Chip: The Rise of the 'Systems Problem'

The era of judging AI performance solely by the speed of an individual processor is rapidly coming to a close. The new frontier is system-level efficiency, a challenge that has become more acute as data centers struggle to keep pace with the voracious demands of generative AI and other frontier models.

"Inference is bigger than any one chip. It's now a systems problem," said Sid Sheth, founder and CEO of d-Matrix, in the company's official announcement. This single statement encapsulates the strategic rationale behind the acquisition. Sheth explained that leading AI developers are increasingly disaggregating workloads, breaking down massive computational tasks into smaller pieces that run simultaneously across a mix of CPUs, GPUs, and specialized AI accelerators like d-Matrix's own Corsair platform.

This disaggregated approach, while powerful, creates a massive data-flow challenge. "That means data must move efficiently across chips, nodes, racks, and entire data centers in real time," Sheth noted. Any latency or bottleneck in this data movement can negate the performance gains of the powerful processors themselves. The acquisition, he added, "accelerates our ability to deliver infrastructure built for this new reality, where low latency, efficiency,and scale all matter at once."

Integrating the Fabric: What d-Matrix Gains with GigaIO

At the heart of the acquisition are GigaIO's core data center technologies, particularly its innovative FabreX™ PCIe-based memory fabric. Traditionally, PCI Express (PCIe) has been a high-speed bus used to connect components inside a single server. GigaIO's breakthrough was to transform PCIe into a true, low-latency network fabric that can span an entire rack, or even multiple racks.

This technology enables what is known as composable disaggregated infrastructure (CDI). With a fabric like FabreX, data centers can create flexible pools of resources—compute, memory, and accelerators—that can be dynamically allocated to workloads on demand. This provides the performance of a tightly-coupled system with the flexibility and efficiency of a networked architecture.

The acquisition includes GigaIO's SuperNODE, an engineered system built upon the FabreX fabric that allows dozens of accelerators to be connected to a single node with the performance of being inside the same server. This technology, which d-Matrix began integrating in a 2025 partnership, proved to be a perfect match for its own product stack. By owning the interconnect fabric, d-Matrix can now tightly integrate its Corsair™ inference accelerators and JetStream™ networking products into a complete, optimized system, all managed by its Aviator™ software.

This approach offers a compelling alternative to the proprietary, closed-ecosystem interconnects offered by some market giants. By leveraging an extension of the open PCIe standard, d-Matrix can offer customers greater flexibility and avoid vendor lock-in, a significant advantage in the rapidly evolving AI hardware market.

Reshaping the AI Infrastructure Landscape

The deal is not just about technology; it's also about talent and market position. With the acquisition, d-Matrix absorbs a highly skilled team of systems engineers based in Carlsbad, California. This establishes a new Southern California engineering hub for d-Matrix, expanding its global footprint to six innovation centers across North America, Europe, and Asia. This infusion of specialized expertise in rack-scale design and high-performance interconnects is critical for executing on the company's system-level vision.

By moving up the stack from a component provider to a systems integrator, d-Matrix is strengthening its competitive position against industry heavyweights like Nvidia, Intel, and AMD. The company is betting that customers will increasingly seek complete, pre-validated solutions that solve the complex integration challenges of building AI infrastructure at scale. The acquisition enhances the value of d-Matrix's SquadRack™, a reference architecture developed with partners Broadcom and Arista, by providing a powerful, in-house fabric to tie it all together.

Industry analysts see the move as a savvy consolidation that addresses a critical need. As the market matures, the ability to deliver not just a fast chip, but a fast, efficient, and scalable system, will become the key differentiator for success.

A New Chapter for GigaIO at the Edge

While d-Matrix absorbs its data center business, GigaIO, Inc. is not disappearing. Instead, the company is embarking on a new, highly focused chapter. It will continue to operate as an independent entity with a singular mission: conquering the burgeoning market for edge computing.

The divestiture allows GigaIO to pour all its resources into products like Gryf, a suitcase-sized, field-deployable AI supercomputer. Powered by the same core FabreX technology, Gryf is designed to bring data center-class processing power to remote and rugged environments where traditional infrastructure is impossible, from defense applications to energy exploration.

This strategic split appears to be a win-win. d-Matrix gains the technology and talent needed to dominate the data center inference market, while GigaIO emerges as a leaner, more focused company poised to become a leader in the high-growth edge AI sector. For both companies, the deal marks a pivotal step in their respective journeys to shape the future of computing.