Atlera OpenCL SDK overview and case studies - CPU + Hardware Accelerator, ARM®- based system-on-a-chip(SoC) devices by Intel, Apple,NVIDIA, Qualcomm, TI, and others. Massive Parallelism - Millions of logic elements, Thousands of 20Kb memory blocks, Thousands of DSP blocks, Dozens of High-speed transceivers. OpenCL Driving Force. Developed and published by consortium http://www.khronos.org Royalty - free standard. Application-driven Consumer Electronics, Image Processing & Video, Encoding, Augmented reality & Computational Photography. Military - Video and Radar Analytics. Scientific / High Performance Computing - Financial, Molecular Dynamics, Bioinformatics, etc. Medical rendering algorithms. Build extremely deep pipelines with wide parallelism - Deep Pipelined Parallelism, Map coarse grained thread parallelism to fine-grained FPGA parallelism, Create deeply pipelined representation of a kernel replicated across FPGA, On each clock cycle, send in input data for a new thread. Modular Architecture - The OpenCL Compiler converts the kernel code and builds the rest of the device within its resource and timing constrains.
|