This means it will be used with Amazon’s Elastic Compute Cloud (Amazon EC2) F1 instances to provide reconfigurable, custom-hardware acceleration with the firm’s 16nm Virtex UltraScale+ FPGAs.
SDAccel automates the acceleration of software application written in C, C++ or OpenCL by building application-specific FPGA kernels for Amazon EC2 F1.
The hope is this will open up the application of SDAccel technology to software developers less familiar with FPGAs.
These are typically used for compute-intensive workloads like data analytics, video processing and machine learning.
SDAccel, which is accessible through the AWS FPGA developer AMI, also supports traditional hardware designers writing accelerators in lower-level hardware description languages such as VHDL Verilog.