CUDA and OpenCL Training

We provide high-quality 2- or 4-day CUDA™ and OpenCL training courses.

Since we specialize solely in CUDA and OpenCL work, we can uniquely immerse students in GPU and heterogeneous computing. Students of our courses walk away proficient at programming CUDA or OpenCL, receive the latest industry knowledge and techniques for GPU computing, and learn the tricks to maximize performance from heterogeneous computing devices.

For groups, we either travel to your location, host in our Atlanta office, or train remotely via video conference, tailoring our instruction to meet your application-specific needs. For individuals, we offer 2-day training quarterly.

We recommend that attendees have a working knowledge of C/C++ for a fruitful learning experience.

Talk to Us to Register a Group Training Course

"Can't ask for better individualized instruction than the environment I was fortunate enough to encounter. The instructor was able to completely focus on my particular needs and concerns."

-Brian Rapp, US Army Research Lab

Included in All Courses

You provide the minds, and we'll take care of the rest. Each training comes with the following:

  • Instruction by an excellent and exciting expert
  • Hands-on exercises
  • Use of a laptop with CUDA and OpenCL capable GPUs and CPUs
  • Choice of Linux or Windows operating system
  • Printed manual of lecture material
  • Electronic copy of programming exercises

CUDA and OpenCL Training Syllabus

* Courses are taught in either CUDA or OpenCL. Similar principles apply in each framework.

Day 1, Introduction


  • GPU Computing Overview
  • The Programming Model
  • Basic Dataset Mapping
  • Techniques
  • Libraries, ArrayFire
  • Profiling Tools


  • A Simple Kernel
  • Equivalent ArrayFire Example
  • Using Libraries
  • Monte Carlo Pi Estimation
  • Timing and ArrayFire
  • Debugging Code
Day 2, Optimization


  • Architecture: Grids, Blocks, and Threads
  • Memory Model: Global, Shared, and Constant Memory
  • Advanced Mapping Techniques
  • Streams: Asynchronos Launches and Concurrent Execution
  • ArrayFire: Lazy Evaluation and Code Vectorization


  • Matrix Transpose
  • Optimization Using Shared Memory
  • Median Filter
  • Optimization Using Constant Memory
  • Stream Example
  • ArrayFire Example: Nearest Neighbor Algorithm
Day 3, Multi-GPU

Lectures (customizable):

  • Multi-GPU Use Cases
  • Multi-GPUs: Contexts
  • Existing Libraries
  • Scaling Across Multiple GPUs


  • Out of Core Problems: Matrix Multiply
  • Task Level Parallelism: Optimization
  • ArrayFire Multi-GPU
Day 4, Algorithm Problems

Lectures and Practice (customizable):

  • Reductions
  • Scan Algorithms
  • Sort
  • Convolution
  • Customer-Specific Problem

Xilinx SDAccel Training

In addition to CUDA & OpenCL training, we offer training for Xilinx SDAccel. ArrayFire is North America's exclusive Xilinx SDAccel Authorized Training Partner (ATP). Our SDAccel training courses help design teams leverage Xilinx FPGAs for OpenCL application acceleration.

Course Name:  "Developing and Optimizing Applications Using the OpenCL Framework for FPGAs"

Talk to Us to Register an SDAccel Training
(or email us at with any questions)

Individual 2-Day Course

For single individuals, we provide an online 2-day CUDA-only training course once a quarter, following the Days 1 and 2 syllabi, as shown above.

Upcoming Dates

  • Q2:  June 11-12, 2024
  • Q3:  September 10-11, 2024
  • Q4:  December 10-11, 2024

Each course has limited spots, so reserve your place as soon as possible by emailing us at