Get in Touch
 Duration 21 hours

Course Outline

Overview of Biren GPU Architecture

  • Introduction to Biren and its primary use cases
  • Hardware configuration: cores, memory structures, and compute clusters
  • Architectural comparison with NVIDIA and AMD GPUs

Configuring the Biren Programming Environment

  • Installation of the Biren SDK and runtime components
  • Comprehending the toolchain and compiler architecture
  • Project organization fundamentals and build workflows

GPU Programming with the Biren Stack

  • Thread and block modeling concepts
  • Memory management strategies and data transfer mechanisms
  • Kernel development workflows and launch patterns

Migrating from CUDA to Biren

  • Techniques for converting CUDA codebases
  • Mapping common APIs and adapting interfaces
  • Practical labs focused on code conversion and refinement

Debugging and Profiling Methodologies

  • Leveraging Biren’s integrated debugger and profiler
  • Detecting performance bottlenecks
  • Analyzing memory access patterns and optimization opportunities

Advanced Optimization Techniques

  • Thread scheduling strategies and instruction pipelining
  • Utilizing loop unrolling and shared memory effectively
  • Advanced kernel tuning for maximum throughput

Case Studies and Real-World Applications

  • Training machine learning models using Biren accelerators
  • Porting and profiling computer vision or NLP models
  • Performance benchmarking against CUDA/NVIDIA standards

Conclusion and Future Directions

Requirements

  • Knowledge of GPU architecture and parallel processing concepts
  • Experience with CUDA, OpenCL, or comparable GPU programming environments
  • Familiarity with deep learning frameworks like PyTorch or TensorFlow

Target Audience

  • HPC developers
  • AI infrastructure engineers
  • Performance optimization specialists

Number of participants


Price per participant

Testimonials (2)

Upcoming Courses

Related Categories