Get in Touch
 Duration 21 hours

Course Outline

Introduction to Biren GPU Architecture

  • Overview of Biren technology and key use cases.
  • Hardware composition: cores, memory structure, and compute clusters.
  • Comparative analysis with NVIDIA and AMD GPUs.

Establishing the Biren Programming Environment

  • Installation of the Biren SDK and runtime components.
  • Understanding the toolchain and compiler methodology.
  • Basic project organization and build processes.

GPU Programming within the Biren Stack

  • Thread and block execution models.
  • Memory management and data transfer mechanisms.
  • Kernel development and launch strategies.

Migrating from CUDA to Biren

  • Techniques for translating CUDA code.
  • Common API mappings and necessary adaptations.
  • Practical labs and exercises for code conversion.

Debugging and Profiling

  • Utilizing Biren’s debugger and profiling tools.
  • Identifying system bottlenecks.
  • Analyzing memory access patterns for optimization.

Advanced Optimization Techniques

  • Thread scheduling and instruction pipelining.
  • Loop unrolling and efficient shared memory utilization.
  • Advanced kernel tuning for maximizing throughput.

Case Studies and Application Examples

  • Training models using Biren accelerators.
  • Porting and profiling vision or NLP models.
  • Performance comparison against CUDA/NVIDIA solutions.

Summary and Future Directions

Requirements

  • A solid grasp of GPU architecture and parallel processing concepts.
  • Hands-on experience with CUDA, OpenCL, or comparable GPU programming environments.
  • Working knowledge of deep learning frameworks such as PyTorch or TensorFlow.

Audience

  • HPC developers.
  • AI infrastructure engineers.
  • Performance optimization specialists.

Number of participants


Price per participant

Testimonials (2)

Upcoming Courses

Related Categories