Mitsubishi Heavy Industries Rolls Out Secure Edge Data Center Brand DIAVAULT

Why It Matters

Edge AI reduces latency and data‑privacy risks, giving industrial firms real‑time insight and compliance, while opening a new revenue stream for MHI in a fast‑growing market.

Key Takeaways

• DIAVAULT offers secure, on‑prem AI inference.
• Supports deployments from containers to multi‑megawatt sites.
• Uses two‑phase direct chip cooling, low PUE.
• Integrated power, cooling, AI compute in one platform.
• Targets manufacturing, defense, 5G low‑latency applications.

Pulse Analysis

The rapid expansion of industrial IoT and AI workloads is reshaping how manufacturers handle data. While public cloud services provide scale, they often introduce latency and expose sensitive information. Edge data centers bridge that gap by processing data close to the source, delivering real‑time insights and tighter security controls. Mitsubishi Heavy Industries’ introduction of DIAVAULT reflects this shift, positioning the company to capture a growing segment of the edge‑infrastructure market that values performance, compliance, and on‑site reliability. Enterprises seeking to meet stringent regulatory standards are especially likely to adopt such localized solutions.

DIAVAULT differentiates itself through a tightly integrated hardware stack. The Yokohama Hardtech Hub demo houses next‑generation GPUs paired with two‑phase direct chip cooling, a method that removes heat at the silicon level and drives Power Usage Effectiveness below industry averages. Built‑in power distribution, uninterruptible supplies, and digital air‑conditioning control automatically balance server load with climate management, reducing operational costs. Security is baked in via two‑factor access and remote monitoring, enabling truly unmanned operation while safeguarding critical AI inference workloads. The modular design also allows scaling from container‑size units to multi‑megawatt facilities.

The launch positions MHI as a serious contender against established edge providers such as HPE, Dell and Schneider Electric. By leveraging its legacy in power systems and heavy‑industry engineering, MHI can offer end‑to‑end solutions that combine reliability with cutting‑edge AI performance. Customers in manufacturing, defense, and 5G networks stand to gain faster decision loops and reduced data‑exfiltration risk. As global AI adoption accelerates, the demand for secure, low‑latency compute will likely drive partnerships and co‑development projects, expanding DIAVAULT’s ecosystem beyond Japan. Early adopters could also influence future feature roadmaps, ensuring alignment with sector‑specific requirements.