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.
Duration 21 hours
Course Outline
Exploring the Chinese AI GPU Landscape
- Comparative analysis of Huawei Ascend, Biren, and Cambricon MLU
- Contrasting CUDA with CANN, Biren SDK, and BANGPy paradigms
- Market trends and vendor ecosystem dynamics
Pre-Migration Preparation
- Evaluating your existing CUDA codebase
- Selecting target platforms and appropriate SDK versions
- Setting up the toolchain and development environment
Strategies for Code Translation
- Adapting CUDA memory access patterns and kernel logic
- Aligning compute grid and thread models
- Exploring automated versus manual translation workflows
Platform-Specific Execution
- Utilizing Huawei CANN operators and custom kernels
- Navigating the Biren SDK conversion pipeline
- Restructuring models using BANGPy (Cambricon)
Inter-Platform Validation and Tuning
- Profiling execution performance on each target platform
- Optimizing memory usage and comparing parallel execution
- Monitoring performance metrics and iterating on improvements
Overseeing Hybrid GPU Environments
- Implementing hybrid deployments across multiple architectures
- Developing fallback mechanisms and device detection protocols
- Introducing abstraction layers to enhance code maintainability
Real-World Examples and Proven Methods
- Porting vision and NLP models to Ascend or Cambricon
- Modifying inference pipelines for Biren clusters
- Resolving version conflicts and API discrepancies
Conclusions and Future Directions
Requirements
- Hands-on experience programming with CUDA or GPU-based applications
- Solid grasp of GPU memory models and compute kernels
- Proficiency in AI model deployment or acceleration workflows
Target Audience
- GPU Developers
- System Architects
- Porting Specialists
