OSS Wins Multi-Million-Dollar AI Transportable Program with Leading 5G Cellular Provider

One Stop Systems, Inc. a leader in AI Transportable solutions on the edge, has won a multi-year, multi-million-dollar program with a Fortune 50 cellular provider to supply OSS transportable compute and storage hardware for a mobile 5G AI application.

The deployment will involve installing OSS SDS rugged servers in service vehicles operating in more than 100 cities nationwide. The vehicles will use onboard 5G antennas to gather and characterize data in real-time in the field, and then transport the data back to a service hub where it is uploaded to the cloud and combined with other data sets.

“This leading 5G cellular provider chose OSS because of our technology leadership and long history of delivering rugged compute and storage solutions for transportable applications, particularly those involving high-speed data recording,” stated OSS CEO, David Raun. “We anticipate a long-term relationship developing under this commercial program win and it leading to other opportunities with this marquee customer.”

The planned deployment will leverage OSS’s unique AI transportable edge technology consisting of an OSS Gen 4 PCIe® SDS-3U GPU-accelerated server mounted in a vehicle equipment rack. The SDS will be powered by the latest AMD EPYC™ 7003 Series Processors and 16 NVMe drives—an ideal configuration for high-speed NVMe data acquisition and recording.

The system features the first server class, dual socket AMD EPYC platform ruggedized for harsh edge environments and will be air-cooled inside the vehicle.

Each SDS-3U system will be capable of ingesting, recording, and storing up to 256 terabytes of raw 5G RF data signals onto high-speed NVMe SSDs encased in two easily removable, encrypted drive packs.

OSS’s proprietary high-speed PCIe Gen 4 switched interconnect will transport the data to the compute processors for real-time, in-vehicle signal analysis using the latest AI deep learning frameworks. For uploading the massive amount of collected data to the cloud, the drive packs can be easily removed and transported to OSS SDS servers installed in-ground stations with high-bandwidth Internet connections.

“This 5G cellular signal capture and analysis application is a perfect example of our strength in the AI Transportable segment of edge computing which requires the highest performance in challenging mobile environments,” added Raun. “Our rugged SDS server demonstrates how we continue to push the boundaries of high-performance edge computing, where speed, security, and reliability are the most critical factors for AI transportable applications”.

OSS SDS-3U gives users the flexibility to choose to encrypt data on the move using optional OSS Ion Accelerator 6.x SAN and NAS software or encrypting data at rest using NVMe self-encrypting drives. The servers are ideally suited for field applications, where they can be housed in highly ruggedized mobile containers, shelters, or vehicles. These mobile or field locations can then support end-to-end AI workflows and on-the-spot decision-making without the need to access a remote centralized data center for processing.

SDS options include up to sixteen PCIe 4.0, FIPS 140-2 encryption compliant U.2 NVMe™ SSD drives, up to four of the latest Gen 4 GPUs, and high-speed Network Adapters or SmartNICs. These components enable the system to deliver unprecedented acceleration, security, and flexibility for AI, data analytics, and high-performance edge computing. Altogether, the system represents the highest performance, most secure FIPS 140-2 solution required by today’s government and enterprise AI edge applications.

OSS expects initial deliveries of the mobile 5G AI system to begin before the end of this year, with deployments continuing through 2026.

OSS is focused on the fastest-growing segment of the edge computing market, AI Transportables.  According to MarketsandMarkets, the global edge computing market is projected to grow at a compound annual growth rate of 17.8% to reach $101.3 billion by 2027.

Leave a Reply

Your email address will not be published. Required fields are marked *