• Mode-6 App. Kernels • PDE Solvers : FDM /FEM • Image Processing - FFT • Monte Carlo Methods • String Search. • Sequence Analysis • Video Processing • Intr. Detcn. Sys • App. Power & Perf. • Home




hyPACK-2013 HPC GPU Cluster - Application Kernels



Video Processing

Video Processing : The graphics performance of specialized software, e.g. scientific software, image manipulation, video decoders/encoders, games that make GPU performance is pretty important. In video processing, the GPU processor takes a video stream and perform image processing operations (effects) on its content using graphic APIs to generate a video stream that can be presented to the display in real time. The scalability of video processing can be achieved when multiple video streams are enhanced to combine multiple video streams together to produce a singel output stream. On GPUs, hardware decoding with OpenVideo Decode API have been used. Many decoding libraries exist for the various platforms supporting video devcode using the power of CPU Multi-core processors. Video processing effects can be carried out on GPU. The CUDA enabled NVIDIA GPU or OpenCL can be used to decode video on the GPU. Working in OpenCL and output video display using OpenGL is one programming paradigm. To decide a frame of video, we must fist initialize the decoding framework and request tha tit open the file. The decode framework will then decode frame of video in the designated format. The process of data is done using OpenCL. Important steps of Video Processing will be discussed in laboratory sessions.

HPC GPU Cluster : In hyPACK-2013 workshop, a prototype HPC GPU cluster (CUDA /OpenCL enabled NVIDIA GPUs & AMD-ATI OpenCL Prog. env) is used to solve application kernels, that are based on Heterogenous programming model In this workshop, programming and performance issues for applications on HPC GPU Clusters will be discussed. In laboratory session, a prototype Hybrid Heterogneous HPC GPU Cluster is made available, which can address some of the heterogeneous computing workloads. The HPC GPU Cluster can be made "adaptive" to the application it is running, assigning the most effective resources in real-time as per application demands, without requiring modifications to the application. One of the objectives of HPC GPU Cluster (hybrid computing system) is to allocate resources of CPUs & GPUs in an optimal way to solve applications of different characterstics.

Centre for Development of Advanced Computing