Newsletter

Newsletter
 

OpenCL Training Courses for AMD APUs and FirePro GPUs

 

AccelerEyes has been working with AMD develop premium training courses for AMD APUs and AMD FirePro GPUs. These training courses are the fastest way for developers to become proficient at programming OpenCL for AMD devices.

AccelerEyes is uniquely equipped to provide training for AMD OpenCL devices due to our extensive experience programming ArrayFire support for AMD APUs and GPUs. In our training courses, we are able to share important first-hand experience that will greatly benefit your development efforts.

AMD devices are uniquely capable of high-performance computations due to their advantage of coupling top-notch AMD CPUs with FirePro GPUs, resulting in zero-copy capabilities between devices. Contact us today to schedule a training course for your organization.

Attendees will receive the latest industry knowledge and techniques for computing with OpenCL and accelerator devices. We have helped thousands of organizations speedup their code and our primary objective is to help you increase productivity while maximizing the return on your hardware. AccelerEyes training will empower you with the knowledge your organization needs in order to achieve success in accelerated computing.

     Training picture

Upcoming Training Courses and Locations

OpenCL

 

 



OpenCL Training Course Syllabus

  • Day 1: Introduction to OpenCL
    • Lectures:
    • OpenCL Computing Overview
    • The OpenCL Programming Model
    • Architectures Supported
    • Basic Dataset Mapping Techniques
    • OpenCL Libraries, ArrayFire
    • Asychronos Operation
    • Portability

    • Practice:
    • A Simple OpenCL Kernel
    • Equivalent ArrayFire Example
    • Monte Carlo Pi Estimation
    • Using OpenCL Libraries
    • Timing OpenCL and ArrayFire
    • Porting Code for Multiple Devices

  • Day 3: Scaling Across Multiple Devices
    • Lectures:
    • Multi-Device Use Cases
    • Heterogeneous Computing
    • Scaling Across Multiple Devices

    • Practice:
    • Out of Core Problems: Matrix Multiply
    • Task Level Parallelism: Optimization
    • ArrayFire Multi-Device

 

  • Day 2: OpenCL Optimization
    • Lectures:
    • OpenCL Architecture: Work Groups, Work-Items, WaveFronts
    • OpenCL Memory Model: Global, Local and Constant Memory
    • OpenCL Command Queues: Asynchronos Launches and Concurrent Execution
    • Advanced Mapping Techniques
    • Architecture Specific Limitations and Optimizations
    • ArrayFire: Lazy Evaluation and Code Vectorization

    • Practice:
    • Matrix Transpose
    • Optimization Using Local Memory
    • Median Filter
    • Optimization Using Constant Memory
    • Command Queues Example
    • ArrayFire Example: Nearest Neighbor Algorithm

  • Day 4: OpenCL Algorithm Problems
    • Lectures and Practice:
    • Reductions
    • Scan Algorithms
    • Sort
    • Convolution
    • Customer-Specific Problem

Interested in OpenCL Training for AMD Devices

Fill out this form and we'll send you an email to get started.

* Required Fields

First Name *
Last Name *
Organization *
Email *
Phone
How can we help?