New Energy-Efficient CUDA ARM Development Kit From SECO

Integrating a PCI Express x4 interface on PCI-e x16, the SECO mITX GPU DevKit allows the use of a range of NVIDIA GPUs, including those based on the NVIDIA Kepler compute architecture.

The SECO mITX GPU DevKit will support the recently announced NVIDIA Kayla platform for computing on the ARM architecture, which is expected to be available this spring. The Kayla platform is powered by an NVIDIA Tegra quad-core ARM processor and a Kepler GPU to deliver the highest performance and highest efficiency for next-generation of CUDA and OpenGL applications.

In a world where power-hungry hardware is becoming more and more a liability, the new SECO mITX GPU DevKit enables software developers to explore the world of low power and high performance ARM/GPU based applications, said Alessandro Santini, SECO’s HPC sales manager.

“ARM CPUs when combined with the compute power of GPUs can deliver maximum performance per watt,” said Ian Buck, general manager of GPU computing software at NVIDIA. “The SECO mITX GPU DevKit makes it easy for programmers to develop new CUDA-based applications for mobile devices, HPC servers and a broad range of other systems.”

SECO mITX GPU DevKit is available for pre-order at a suggested retail price of 349 EUR excl. taxes, and is comprised of:
- QuadMo747-X/T30 - QSeven v1.2 module, NVIDIA Tegra 3 with 2 GB RAM and 4 GB eMMC
- SECO Mini-ITX Carrier Board
- Cable Kit

SECO mITX GPU DevKit Specifications:

-Energy-efficient Tegra 3 quad-core ARM CORTEX A9 processor with 2 GB processor memory
-PCI Express x16 mechanical connector, x4 electrical
-1 Gigabit Ethernet network support
-Direct attached storage support with SATA port
-HDMI
-3 USB port + 1 OTG port

Shipments will start on April 8th.