GT-92RL-H: Acceed's uncompromising hardware for edge AI

As AI models become increasingly efficient in industry, the demands on hardware in the field are also growing: GPU performance must be mobile, robust and maintenance-free in vehicles, production lines, energy plants and security systems. With the new GT-92RL-H controller, German distributor Acceed now has another edge platform in its portfolio that is designed precisely for this purpose: powerful, fanless, with high I/O density and, thanks to its robust system design (in accordance with EN 50155 / EN 45545), largely independent of infrastructure conditions.

In the age of distributed intelligence, data-intensive processes are increasingly shifting from the data centre to the edge of the network, where machines need to see, analyse and make decisions. But what if these edge areas are dusty, hot, subject to intense vibration or simply lack infrastructure? This is exactly where the GT-92RL-H from Acceed's portfolio comes in. The new controller is an industrial-grade edge computer in a 19" rack enclosure with dedicated GPU performance and a completely fanless architecture, designed for AI applications in environments where traditional IT often fails.

Depending on the configuration, the controller runs on 12th to 14th generation Intel processors (up to 65W TDP). The integrated Nvidia RTX 2000 ADA GPU is known as an energy-efficient AI accelerator for demanding inference tasks. Both the CPU and GPU operate in a completely passively cooled system with no moving or contamination-prone parts. This ensures reliable and low-maintenance operation at ambient temperatures between -40 and +55 °C.

The GT-92RL-H has nine Gigabit Ethernet ports via robust and secure M12X sockets, suitable for direct connection of IP or GigE Vision cameras. There are also four USB 3.2 ports, for example for additional sensors. The internal M.2 slot with PCIe Gen4x4 enables NVMe storage with data rates of up to 7000 MB/s. Two additional SATA interfaces expand the storage options. The setup is complemented by two Mini PCIe slots, which can be used to connect mobile radio modules, GNSS or MVB (Multifunction Vehicle Bus, a serial fieldbus standard for vehicles).

The GT-92RL-H impresses with its ability to reliably perform GPU-accelerated data processing under difficult environmental conditions without active cooling and without sacrificing standardised interfaces. Combined with its compact 2U design and flexible mounting options, the system is recommended wherever real-time AI needs to operate robustly and maintenance-free at the point of action. The manufacturer Neousys is thus addressing not only applications in the railway and commercial vehicle sectors, but also industrial vision and AI systems where the installation of classic server hardware is not possible or not economical, such as in compact control rooms, mobile facilities or remote locations without air conditioning.