Thank you for sending your enquiry! One of our team members will contact you shortly.
Thank you for sending your booking! One of our team members will contact you shortly.
Course Outline
Introduction to Biren GPU Architecture
- Overview of Biren technologies and primary use cases.
- Detailed hardware layout, including cores, memory structures, and compute clusters.
- Comparative analysis with NVIDIA and AMD GPU architectures.
Establishing the Biren Programming Environment
- Installation procedures for the Biren SDK and runtime components.
- Explaining the toolchain and compiler models involved.
- Reviewing basic project structures and build processes.
GPU Programming via the Biren Stack
- Exploring thread and block models within the Biren context.
- Managing memory and handling data transfers efficiently.
- Developing kernels and understanding launch patterns.
Migrating from CUDA to Biren
- Techniques for translating existing CUDA codebases.
- Identifying common API mappings and necessary adaptations.
- Practical labs focused on code conversion and application.
Debugging and Profiling Strategies
- Utilizing Biren’s integrated debugger and profiler tools.
- Systematic identification of performance bottlenecks.
- Analyzing memory access patterns for optimization opportunities.
Advanced Optimization Techniques
- Optimizing thread scheduling and instruction pipelining.
- Applying loop unrolling and leveraging shared memory effectively.
- Advanced kernel tuning to maximize throughput.
Case Studies and Application Examples
- Executing model training tasks using Biren accelerators.
- Practical migration and profiling of computer vision or NLP models.
- Performance benchmarking against CUDA/NVIDIA standards.
Course Summary and Recommended Next Steps
Requirements
- A solid grasp of GPU architecture and parallel processing concepts.
- Hands-on experience with GPU programming environments such as CUDA, OpenCL, or equivalent technologies.
- Proficiency with deep learning frameworks, including PyTorch or TensorFlow.
Intended Audience
- HPC developers.
- AI infrastructure engineers.
- Specialists in performance optimization.
21 Hours
Testimonials (2)
The extensive selection of tools presented
Miruna Buzduga - Aeronamic Eastern Europe
Course - AI Enablement Training for Engineers
Step by step training with a lot of exercises. It was like a workshop and I am very glad about that.