Developing FPGA-accelerated cloud applications with SDAccel: Practice

Get introduced to reconfigurable cloud infrastructure and its role in accelerating computational workloads. Understand the basics of cloud computing and explore FPGA-based solutions, particularly Intel and Xilinx FPGA-based solutions. Learn about the AWS EC2 F1 instances and their relevance in cloud acceleration.

Explore the applicative domains and real-world stories of FPGA acceleration in the cloud, focusing on the AWS EC2 F1 instances. Understand the platform model and learn how to create kernels from RTL IP, C/C++, and OpenCL. Gain insights into compiling the platform, creating an Amazon FPGA Image, and developing and executing host applications on F1 instances.

Dive into the Smith-Waterman algorithm, a fundamental algorithm in computational biology. Understand the problem description, analyze the algorithm and its code, and explore the Roofline model and code profiling. Gain insights into static code analysis, performance prediction via the Roofline model, and SDAccel environment profiling and optimization guide.

Explore the details of the Smith-Waterman algorithm, including its implementation, parallelism, systolic array architecture, input compression, and shift register. Understand the acceleration of Smith-Waterman on Amazon EC2 F1 instances and gain access to the source codes for further exploration and implementation.

Conclude the course with insights into architectural optimizations for high performance and energy-efficient Smith-Waterman implementation on FPGAs using OpenCL. Gain a comprehensive understanding of the course content and explore future directions in FPGA-accelerated cloud applications with SDAccel.