Equipped with up to four, long-life GPUs, the 2U GPU server brings scalable computing power for high-density graphics, virtual displays, situational awareness and artificial intelligence in the field. It is designed specifically to handle challenging, yet critical inference obstacles, with low latency even in volatile, mission-critical conditions.
The server’s rugged design and durability delivers tensor core performance, allowing the capture and analysis of data at the tactical edge.
The integrated system combines two 3rd generation Intel Xeon Scalable processors and GPUs that tackle workloads at 260 teraflops. A dedicated air plenum prevents thermal throttling from the high GPU workload. It weighs just under 12.7 kg.
The platform can be configured for temporary or permanent data storage applications, with up to six NVMe drives. InfiniBand I/O connectivity provides rapid data transfer for backhaul applications.
Suitable for the ever-changing landscape of battlespace management, command and control, intelligence gathering and dissemination and sensor fusion across complex networks for pinpoint situational awareness, the server takes compute-intensive workloads out of the data centre and into the field to provide seamless operation in time-sensitive, unpredictable situations.
Originally published here.
Phone: 07 3868 4255
STMicroelectronics STM32WBA6 wireless microcontrollers
Packing extra memory and digital system interfaces compared to the previous series, while...
Quectel LG680P quad-band GNSS module
Quectel Wireless Solutions has launched the LG680P — a multi-constellation, quad-band GNSS...
Epson OG7050CAN oven-controlled crystal oscillator (OCXO)
Epson's OG7050CAN OCXO consumes 56% less power (0.2 W at 25°C) than its earlier products...