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
Performance Concepts and Metrics
- Latency, throughput, power consumption, and resource utilization
- Distinguishing between system-level and model-level bottlenecks
- Differences in profiling for inference versus training
Profiling on Huawei Ascend
- Utilizing CANN Profiler and MindInsight
- Diagnostics for kernels and operators
- Analyzing offload patterns and memory mapping
Profiling on Biren GPU
- Leveraging Biren SDK for performance monitoring
- Kernel fusion, memory alignment, and execution queues
- Profiling with awareness of power and temperature constraints
Profiling on Cambricon MLU
- Using BANGPy and Neuware performance tools
- Gaining kernel-level visibility and interpreting logs
- Integrating the MLU profiler with deployment frameworks
Graph and Model-Level Optimization
- Strategies for graph pruning and quantization
- Operator fusion and restructuring of computational graphs
- Standardizing input sizes and tuning batch parameters
Memory and Kernel Optimization
- Optimizing memory layout and reuse strategies
- Managing buffers efficiently across different chipsets
- Platform-specific kernel tuning techniques
Cross-Platform Best Practices
- Achieving performance portability through abstraction strategies
- Establishing shared tuning pipelines for multi-chip environments
- Case study: Tuning an object detection model across Ascend, Biren, and MLU
Summary and Next Steps
Requirements
- Experience with AI model training or deployment pipelines
- Understanding of GPU/MLU compute principles and model optimization techniques
- Familiarity with performance profiling tools and metrics
Target Audience
- Performance engineers
- Machine learning infrastructure teams
- AI system architects
21 Hours
