Rigel is a compact and air-cooled hardware solution with a depth of 25.6" that fits in half of a 4U rack, making it about half the size of most other air-cooled quad GPU solutions. The liquid-cooled version of the unit reduces that size to half of a 3U rack space. The small form factor is ideal for use in confined environments, such as an equipment bay of autonomous vehicles, mobile command centers, under the seats of helicopters, or in the equipment bay of an aircraft, where two Rigel systems can fit side-by-side in a 4U rack.

So-called AI Transportables differ from traditional edge AI infrastructures by implementing the latest technologies in areas such as high-speed data centers, input/output, networking and storage to enable operation in adverse environments. AI Transportables meet stringent MIL-SPEC requirements for shock and vibration, redundancy, operating temperature ranges, altitude ranges and uninterruptible power supplies.

Additional features or options, include up to 800 Gbps bandwidth using up to four NVIDIA® ConnectX-6™ 200Gb/s InfiniBand adapters, 64TB of NVMe flash memory, AMD EPYC™ 7002 processors (with 24-cores), and advanced OSS PCIe Gen 4 switching direct connectivity between these critical components. Rigel also includes a secure, unified management and control module that supports Redfish industry standard specifications and provides a user-friendly interface to all system resources.

"Virtually all AI edge computing platforms on the market today do not integrate and optimize data center-class computing with compact, rugged form factors," said Benjamin Nigl, senior account manager for high-performance solutions at BRESSNER Technology. " As our most compact and powerful transportable AI compute server, we believe the Rigel Edge supercomputer will transform the use of real-time AI for the most demanding applications in the automotive, maritime and aerospace industries."

Rigel addresses the growing demand for AI Transportables, one of the fastest growing segments of edge computing. According to MarketsandMarkets, the edge computing market is expected to grow at a compound annual growth rate of 34 percent to $15.7 billion by 2025.