Scribd Logo
Upload

Welcome to Scribd!

  • Lecture 4 - GPU Architecture and Programming

    Uploaded by

    Sibgha Israr
    6 views30 pages
    CPUs are designed for sequential serial processing while GPUs have a highly parallel structure designed for manipulating computer graphics and image processing. GPUs have hundreds to thousands of smaller, more efficient cores compared to CPUs which have only a few cores but are faster at sequential tasks. While CPUs have been optimized for flexibility, GPUs are optimized for throughput on highly parallel workloads like graphics rendering where the same operations are performed on many data elements in parallel.

    Original Description:

    Copyright

    © © All Rights Reserved

    Available Formats

    PPTX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document
    CPUs are designed for sequential serial processing while GPUs have a highly parallel structure designed for manipulating computer graphics and image processing. GPUs have hundreds to thousands of smaller, more efficient cores compared to CPUs which have only a few cores but are faster at sequential tasks. While CPUs have been optimized for flexibility, GPUs are optimized for throughput on highly parallel workloads like graphics rendering where the same operations are performed on many data elements in parallel.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Flag for inappropriate content
    6 views30 pages

    Lecture 4 - GPU Architecture and Programming

    Uploaded by

    Sibgha Israr
    CPUs are designed for sequential serial processing while GPUs have a highly parallel structure designed for manipulating computer graphics and image processing. GPUs have hundreds to thousands of smaller, more efficient cores compared to CPUs which have only a few cores but are faster at sequential tasks. While CPUs have been optimized for flexibility, GPUs are optimized for throughput on highly parallel workloads like graphics rendering where the same operations are performed on many data elements in parallel.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Flag for inappropriate content
    of 30

    CS-423 Parallel and Distributed Computing

    Lecture 4
    GPU Architecture and Programming
    Agenda
    • GPU
    • CPU vs. GPU
    • Traditional Workflow
    • CUDA
    • Thread Hierarchy
    • CUDA Memory Hierarchy
    • Myths about CUDA
    GPU
    • Graphics Processing Unit
    • Originally designed as a graphics processor

    • Nvidia'sGeForce 256 (1999) –first GPU


    • single-chip processor for mathematically-intensive tasks
    • transforms of vertices and polygons
    • lighting
    • polygon clipping
    • texture mapping
    • polygon rendering  3D Model to 2D image (Model: Mathematical Representation of image)
    CPU vs GPU

    Core: A small CPU or processor built into a big CPU or CPU socket. It can independently perform or process all
    computational tasks. A core is considered to be a smaller CPU within a big processor.
    CPU vs GPU
    CPU vs GPU
    Recall:

    You might also like