Get Started
Menu

Scientific computing, and for that matter all of high-performance computing, is quickly migrating to GPU acceleration. Because clock speed is no longer scaling, but computing bandwidth continues to scale, computing algorithms written for single-instruction, multiple data (SIMD) architectures have been and will continue to scale along with Moore’s Law. Both GPU-accelerated algorithms and CPU-only algorithms utilize coarse-level parallelization by dividing the chip or mask data into partitions, and computing the partitions in CPU cores or CPU cores accelerated with GPU(s). But each computing unit computes much more data much faster with GPU acceleration. The performance difference increases every node along Moore’s Law.

Computing capacity scales only by processors now

Peak Double Precision FLOPS
Created with Sketch. 8000 6000 7000 5000 4000 3000 2000 1000 2011 2012 2013 2014 2015 2016 2017 2010 2008 2007 Year GFLOPS Volta 7,000 Pascal 4,000 K80 1,864 K40 1,430 K20 1,175 M2090 665.6 M2050 515.2 M1060 x86 CPU NVIDIA GPU
Peak Memory Bandwidth
Created with Sketch. 1400 1000 1200 800 600 400 200 2011 2012 2013 2014 2015 2016 2017 2010 2008 2007 Year GFLOPS Volta 1,200 Pascal 1,000 K80 480 K40 288 K20 206 M2090 178 M2050 148 M1060 x86 CPU NVIDIA GPU
Source: NVIDIA

D2S GPU+CPU Solutions Produce Optimal Overall Performance

Optimal GPU acceleration is not the result of a simple replacement of GPUs for CPUs. To realize the >10X acceleration potential of GPU-based computing, you must understand when to deploy GPUs and when to use CPUs. The D2S GPU-acceleration approach employs sophisticated software engineering to combine the strength of each to the benefit of the whole system, deciding what to put on CPUs, what to put on GPUs, then scheduling and load-balancing the two to achieve optimal overall performance.

Free White Paper Download
To download the white paper "GPU-Accelerated Computing: Maximizing Performance for Semiconductor Manufacturing" in PDF format, please register below.