hyPACK-2013 HPC GPU Cluster - Heterogeneous Programming

Two types of Hybrid Heterogeneous HPC GPU Cluster is used in laboratory sessions of the workshop. The two clusters i.e., Intel Xeon Processor nodes as host-cpus with CUDA enabled NVIDIA GPUs as device GPU accelerators and another cluster consists of AMD-Opteron processor nodes as host-cpu with AMD-ATI GPUs (AMDFire Stream & AMD-ATI FirePro) as device GPU accelerators These clusters can address some of the heterogeneous computing workloads. The implementation and programming issues of integrated cluster of Multi-Core processors with GPU accelerators, will be discussed. The HPC GPU Cluster supports Parallel Programming models, which include Shared memory programming (POSIX Threads, OpenMP, Intel TBB), and MPI 2.0 standard on Multi Core Processors. The Linux programming environment can be configured to match different workloads of cluster as per application demands and execute highly scalable customized applications.

• AMD's consumer targeted APUs have both an GPU and CPU harbored inside. The AMD A6-3500 APU is rated at 65W power consumption and is running at 2.1 GHz, and AMD's Turbo Core is supported, clocks of up to 2.4GHz depending on the workload. For more details, visit AMD A6 3500 APU review

• AMD A6-3650 APU Processor Review AMD A8-3850 'Llano' Accelerated Processing Unit (APU) AMD A6 3500 APU Llano

• HP Pavailion AMD A8-4500K (Trinity) APU with AMD Radeon 7640G chip the Pavilion dv6-7010 features an AMD A8-4500M APU with four cores, a 1.9 GHz clock frequency and a 2.8 GHz Turbo boost. Graphics are provided by a Radeon 7640G chip. Further specifications include 6 GB of memory, a 750 GB hard disk, Gigabit LAN, 802.11/b/g/n WiFi and Bluetooth. The 15.6-inch screen has a resolution of 1366x768 pixels.

The sustainable bandwidth from the host to device (and vice versa) plays a key role in the acceleration of single DGEMM (Matrix - Matrix computations) library calls. DGEMM is a Matrix-matrix multiply basic building block and there are a set of functions in BLAS (Basic Linear Algebra Subroutines) to perform this operation:

Peak performance (in double precision) of AMD GPU node in having OpencL enabled AMD-ATI GPU is 495.5 Gflop/s

Host-CPU : AMD Opteron X86 12 Core;
Device GPU : AMD Fire Stream 9350 & 9250; AMD FirePro V5900 & V7900

Host-CPU (AMD)

• One AMD Opteron X86 24 Core Multi-Core Processor systems with two PCI-e 2.0 x16 Slots; RAM-48 GB; Clock Speed : 3.0 GHz; Cent OS 5.2; GCC Version 4.1.2; Dual Socket 12 Core (24 cores)

• ACML version , OpenCL and BLAS Libraries; Peak Performance : CPU : 144 Gflops (1 Node - 12 Cores) and AMD-APP with OpenCL Prog. Env.

GPUs (AMD-ATI)

• AMD Fire Stream 9250 GPU Accelerator :
  Double Precision Floating Point : The FireStream 9250 supports double precision floating point operations in hardware;
  High Performance per Watt : Up to 8 GFLOPS per watt of single precision performance potential
  
  Optimized for computation The AMD FireStream product line provides the industry's first double-precision floating point capability on a GPU. The AMD FireStream 9250 is our second generation DP-FP product. With 1GB GDDR3 memory on board and single-precision performance of 1 TFLOPS.

• AMD Fire Stream 9350 GPU Accelerator :
  Technology Need : AMD FireStream Computing Solution
  High DPFP performance : 528 GFLOPS double precision
  High performance per Watt : 2.4 GFLOPS / Watt
  Open standards : OpenCL, Direct Compute
  Performance optimization tools : OpenCL SDK
  PCIe 2.1 Host Interface : 8 GB/S Host-GPU bandwidth
  
  The FireStream 9350 offers maximum GPU performance with 4GB of DDR5 memory in a 2-slot configuration. The FireStream 9350 offers maximum performance / slot with 2GB DDR5 memory in a 1-slot configuration.

• AMD FirePro V5900 :
  The AMD FirePro V5900 features 2GB of blazing-fast GDDR5 memory, 512 stream processors, and support for three simultaneous monitor outputs from a single AMD FirePro V5900 graphics card with AMD technology. The AMD FirePro V5900 supports OpenCL and it has parallel processing capabilities of 512 stream processors and PCI Express 2.1 compliant.

• AMD FirePro V7900 :
  The AMD FirePro V7900 features : 2GB of ultra-fast GDDR5 memory and 1280 stream processors. The AMD FirePro V7900 supports OpenCL and it has parallel processing capabilities of 1280 stream processors and PCI Express 2.1 compliant.

• Demonstrate codes using different memory types of OpenCL Architectures on AMD APP GPU Cluster and AMD APUs

• Incorporation of Error Checks on HPC GPU Cluster based on OpenCL for matric computation test suites

• Example programs on Heterogeneous Programming - OpenCL based on CUDA enabled NVIDIA GPUs

• Tuning & Performance using OpenCL enabled AMD-APP Libraries; Memory Optimization, Data-access optimization for matrix computations

• Matrix Computations : Matrix - Vector Multiplication, Matrix-Matrix Multiplication based on MPI and OpenCL Implementation on HPC GPU Cluster with AMD-ATI GPUs

• Application Kernels demonstration on HPC GPU Clusters (Heterogeneous Programming & MPI, Pthreads & Intel TBB)

• Performance of Matrix Computations using vendor supplied tuned mathematical libraries (OpenCL based BLAS on AMD-ATI GPUs) on HPC GPU Cluster with GPU Accelerators)

• Selective Numerical Computational kernels on Parallel Processing Systems with GPU Accelerator devices using MPI & OpenCL enabled AMD-ATI GPUs on HPC GPU Cluster

• Numerical Linear algebra on Multi-Core Processors using Mixed Mode of Programming ( MPI-OpenCL, Pthreads-OpenCL) on HPC GPU Cluster.

• Special Class of Application Kernels, and Numerical Linear algebra on Multi-Core Processors using Heterogeneous Programming ( OpenMP-OpenCL, MPI-OpenCL, Pthreads-OpenCL) on HPC GPU Cluster.

• HPC-GPU Cluster (MPI on host-CPU & GPU - OpenCL - Solution of Partial differential Equations

• HPC GPU Cluster (MPI on host-CPU & GPU - OpenCL - Image Processing -Edge Detection algorithms

• Heterogeneous Programming (MPI on host-CPU & GPU - OpenCL - String Search algorithms & Sequence Analysis Applications

• Develop test suites on HPC GPU Cluster based on MPI programming in Host-CPU to launch multiple kernels on GPU devices on each node of HPC GPU Cluster in an MPI- OpenCL programming environment

• HPC GPU Cluster (MPI on host-CPU & GPU-OpenCL - Open source software Benchmarks - Solution of Matrix system Ax=b of Linear Equations (OpenCL based LINPACK solvers)

• HPC GPU Cluster (MPI on host-CPU & GPU-OpenCL - Open source software Benchmarks - LINPACK (Solution of Matrix system Ax=b of Linear Equations)

• Performance of MAGMA (Numerical Linear Algebra Kernels) on CUDA enabled GPUs & L HPC GPU Cluster (MPI on host-CPU & GPU - OpenCL - Image Processing -Edge Detection algorithms using OpenACC

• Bio-Informatics: Sequence analysis (Smith Waterman Algorithms) on HPC GPU Cluster - OpenCL enabled NVIDIA GPUs

• Solution of Partial Differential Equations (Poisson Equation in two dimensional & three dimensional regions) by finite element Method (FEM) using OpenCL AMD-APP on HPC GPU Cluster.

• Image Processing -Face Detection and Image Inpainting algorithms on HPC GPU Cluster - AMD APP