Acceleration

Advancements in FPGA silicon, IPs and tool flows have proven FPGAs to be the ideal solution for accelerating complex computational applications for solving science, engineering and business problems. The inherent re-programmability of FPGAs, deep pipelining structures and massive parallel compute resources to favour the ever-evolving technology of High-Performance Computing (HPC) algorithms and applications in areas including Neural Networks, Cryptography, Computer Vision, Genomics, Big Data, High-Frequency Trading. Moreover, with the same or even better results, FPGA can dramatically reduce power consumption and allow future upgrades and modifications.

Large Scale HPC

Conventional servers do not offer enough computing power despite their high energy consumption. That is why FPGA acceleration is the solution to current computationally intensive challenges.

Aldec’s scalable FPGA accelerators have been developed and optimized for large-scale HPC applications. Today’s generation of FPGA accelerator cards have low-power Xilinx® Ultrascale ™ FPGAs that offer outstanding computing performance with an energy efficiency that cannot be achieved with GPU-based accelerators.

Large Scale HPC
Embedded HPC

Embedded HPC

Recently, there has been an increasing need for embedded systems, many of which incorporate compelling microprocessor designs that exceed the capabilities of desktop computers available ten years ago. Smartphones are a common example, but there are hundreds of others. Computer vision, automotive systems, UAV (drones), security cameras and network security are just a few categories in which embedded electronics are present.

Embedded HPC is the right solution for the growing demand for computing capacity in embedded applications. These applications require the CPU and FPGA to be on the same board and ideally on the same chip. For this reason, Aldec’s TySOM boards are based on the Xilinx Zynq family of MPSoC and chipsets that integrate the ARM Programmable System and FPGA.