• Topics of Interest • Tech. Prog. Schedule • Topic : Multi-Core • Topic : ARM Proc • Topic : Coprocessor • Topic : GPGPUs • Topic : HPC Cluster • Topic : App. Kernels • Lab. Overview • Key-Note/Invited Talks • Home





hyPACK-2013 Topics of Interest : Mode-04- Device GPUs

In Mode-04 programme, topics dealing with practical approaches to write Heterogeneous programs on GPGPUs are considered. CUDA enabled NVIDIA GPU Programming, and heterogeneous programming (OpenCL) on NVIDIA /AMD GPUs GPGPUs are used for development of programs on matrix computations. Tuning & performance aspects of codes on GPGPUs will be carried out. Programming approaches on Heterogeneous HPC GPU Clusters platforms will be discussed. Topics of interest are listed below.



  • Introduction to NVIDIA-PGI Complier Directives - OpenACC on GPUs; CUDA enabled NVIDIA GPUs

  • Performance of Matrix Computations - NVIDIA-PGI Complier Directives OpenACC on GPUs; CUDA enabled NVIDIA GPUs

  • Performance of Application Kernels - NVIDIA-PGI Complier Directives OpenACC on GPUs; CUDA enabled NVIDIA GPUs

  • Simple example programs on Multi-Core Processors with NVIDIA - GPU Computing CUDA 4.1 SDK.

  • Write programs based on the AMD APP Software Development Kit (SDK) based on OpenCL (Open Computing Language)

  • Performance of selective programs on Multi-Core Processors with NVIDIA - GPU Computing CUDA SDK and AMD-APP Tech. SDK based on OpenCL.

  • Special example programs using CUDA Tool Chain on Multi-Core Processors with NVIDIA - GPU Computing CUDA SDK (CULA Tools, CUBLAS, CUFFT, CUSPARSE)

  • Special example programs on matrix computations using Concurrent Asynchronous Execution APIs of CUDA 4.1 enabled NVIDIA GPUs (single/Multiple devices).

  • Special example programs based on Streams (Concurrent Asynchronous Execution) of CUDA 4.1 of NVIDIA GPU

  • LLVM-based CUDA complier and toolkit technologies for matrix computation and application kernels; GPU Accelerator Programming Model - Compiler Optimizations

  • Expousre to NVIDIA Parallel Nsight tool kit.

  • Codes to understand different memory types of CUDA enabled NVIDIA GPUs for matrix computations.

  • Example programs based on Numerical Linear Algebra using CUDA enabled NVIDIA GPUS and AMD-APP OpenCL.

  • Example programs (BLAS, FFTs) based on AMD Accelerated Parallel Processing Math Libraries (APPML) using OpenCL.

  • Example programs based on special class of problems- Dense &. Sparse Matrix Computations, Fast Search Algorithms, & Partial Differential Eqs.(PDEs) will be discussed using CUDA enabled NVIDIA GPUs & AMD-APP OpenCL of HPC GPU Cluster.

  • Example programs on Heterogeneous Programming - OpenCL based on CUDA enabled NVIDIA GPUs and AMD-APP GPUs.

  • Code Walk through and execution of parallel programs based on mixed programming environment using using TBB, Pthreads, OpenMP on host Multi-Core systems with GPU Accelerator devices.

  • Selective example programs on numerical and non-numerical computations using NVIDIA - GPU Computing CUDA SDK and AMD - APP SDK OpenCL.

  • Application & System Benchmarks related to HPC GPU Cluster based on CUDA/OpenCL NVIDIA & OpenCL AMD-APP programming paradigms.

  • Example programs based on The OpenACC Application Program Interface (a collection of compiler directives and the details are implicit in the programming model and are managed by the OpenACC API-enabled compilers and runtimes) for matrix computations on NVIDIA GPUs.

  • Example programs based on AMD APP - Aparapi Data Parallel workloads in Java

  • Example programs based on CUDA APIs to completely overlap CPU and GPU execution and I/O in HPC GPU Cluster environment.

  • Performance of memory (pinned/locked) & CUDA shared memory usage on CUDA enabled GPUs for application kernels.

  • Develop test suites to launch multiple kernels on CUDA enabled NVIDIA single & multiple GPU devices.

  • Programming exercises for Numerical Computations based on CUDA/OpenCL enabled NVIDIA, & AMD-APP OpenCL Programming for Matrix Computations (Dense & Sparse Matrices Computations)

  • Implementation of Image Processing applications (Edge Detection, Face Detection & Image inpainting algorithms) on GPGPUs using CUDA/OpenCL enabled NVIDIA GPUs and OpenCL AMD-APP GPUs of HPC GPU Cluster

  • Implementation of String Search Algorithms - CUDA/OpenCL enabled NVIDIA GPUs and OpenCL AMD-APP GPUs of HPC GPU Cluster

  • Solution of Partial Differential Equations (Poisson Equation in two dimensional & three dimensional regions) by finite element Method (FEM) using CUDA/OpenCL enabled NVIDIA GPUs & OpenCL on HPC GPU Cluster.

  • Tuning & Performance using CUDA enabled NVIDIA GPU Libraries; Memory Optimisation, Data-access optimization for matrix computations

  • Demonstration of Integrated Numerical Linear Algebra Kernels for Matrix Computations (using Open Source Software) on CUDA enabled NVIDIA GPUs & OpenCL enabled AMD-APP Tech.

  • Tuning & Performance - Selective Application Benchmarks CUDA enabled NVIDIA GPUs & AMD-APP SDK - OpenCL Programming

  • Demonstration of Application kernels - CUDA enabled NVIDIA GPUs; OpenCL & AMD-APP GPUs - SDKs





Centre for Development of Advanced Computing