11.1-gpus-1 - CS6230 HPC Tools and Applications...

Info iconThis preview shows pages 1–13. Sign up to view the full content.

View Full Document Right Arrow Icon
CS6230 HPC Tools and Applications Heterogeneous Computing with GPUs Jeffrey S. Vetter Computational Science and Engineering College of Computing Georgia Institute of Technology http://ft.ornl.gov/~vetter vetter@computer.org Many slides in this lecture are courtesy of NVIDIA.
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Thinking about Parallelism 2 Assembler SIMD, AVX Compiler Libraries, Frameworks Core Threads Pthreads, OpenMP Distributed memory model like MPI, or GAS Languages Libraries, Frameworks Socket: Multicore Threads Pthreads, OpenMP Distributed memory model like MPI, or GAS Languages Memory-Thread affinity becomes much more important Libraries, Frameworks Node Distributed memory model like MPI, GAS Languages Libraries, Frameworks System
Background image of page 2
Thinking about Parallelism (2) 3 Assembler SIMD, AVX Compiler Libraries, Frameworks Core Threads Pthreads, OpenMP Distributed memory model like MPI, or GAS Languages Libraries, Frameworks Socket: Multicore Threads Pthreads, OpenMP Distributed memory model like MPI, or GAS Languages Memory-Thread affinity becomes much more important Libraries, Frameworks Node Distributed memory model like MPI, GAS Languages Libraries, Frameworks System Heterogeneity can exist at all levels Existing GPU, FPGA Archs
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Heterogeneous Computing with GPUs CPU + GPU Co-Processing 4 cores CPU 48 GigaFlops (DP) GPU 515 GigaFlops (DP)
Background image of page 4
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Amazingly fluid, real-time video editing Quick preview of real time edits and effects Realistic preview of final content Faster encoding Mainstream Applications Going Parallel CUDA Accelerates Adobe CS5
Background image of page 6
146X Medical Imaging U of Utah 36X Molecular Dynamics U of Illinois, Urbana 18X Video Transcoding Elemental Tech 50X Matlab Computing AccelerEyes 100X Astrophysics RIKEN 149X Financial simulation Oxford 47X Linear Algebra Universidad Jaime 20X 3D Ultrasound Techniscan 130X Quantum Chemistry U of Illinois, Urbana 30X Gene Sequencing U of Maryland 50x 150x
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Workstations 2 to 4 Tesla GPUs Integrated CPU-GPU OEM CPU Server + Tesla S-series 1U Tesla S-series 1U Systems S2050 S1070 Tesla M-series GPUs M2070 M2050 M1060 Tesla C-series GPUs C2070 C2050 C1060
Background image of page 8
NVIDIA’s Fermi 3B transistors ECC 8x the peak double precision arithmetic performance over NVIDIA's last generation GPU. 512 CUDA Cores featuring the new IEEE 754-2008 floating-point standard NVIDIA Parallel DataCache NVIDIA GigaThread Engine Debuggers, language support 9
Background image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Programming GPUs
Background image of page 10
Doing GPU Computing Right Combination of Hardware and Software C Fortran Java Python Wrappers Direct Compute OpenCL tm OpenCL is trademark of Apple Inc. used under license to the Khronos Group In c C++ Libraries and Middleware cuFFT cuBLAS cuRAND cuSPARSE LAPACK CULA MAGMA NPP cuDPP Thrust VSIPL SVM OpenCurre nt PhysX Video OptiX Ray tracing iray Rendering RealityServ er MATLAB Mathemati ca NVIDIA GPU CUDA Parallel Computing Architecture GPU Computing Applications
Background image of page 11

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
CUDA Programming Effort / Performance Source : MIT CUDA Course x86 CPU NVIDA GPU IBM Cell MATLAB
Background image of page 12
Image of page 13
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 61

11.1-gpus-1 - CS6230 HPC Tools and Applications...

This preview shows document pages 1 - 13. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online