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
CANN Optimization Capabilities Overview
- Mechanisms for handling inference performance within CANN
- Strategic optimization targets for edge and embedded AI environments
- Insights into AI Core utilization and memory allocation strategies
Leveraging the Graph Engine for Analysis
- Fundamentals of the Graph Engine and its execution pipeline
- Visualizing operator graphs and tracking runtime metrics
- Adjusting computational graphs to drive optimization
Profiling Tools and Key Performance Metrics
- Utilizing the CANN Profiling Tool (profiler) for comprehensive workload analysis
- Evaluating kernel execution times to identify bottlenecks
- Profiling memory access patterns and implementing tiling strategies
Building Custom Operators with TIK
- Exploring TIK and its operator programming model
- Developing custom operators using the TIK DSL
- Validating and benchmarking operator performance
Advanced Operator Optimization using TVM
- Introduction to TVM integration with the CANN ecosystem
- Implementing auto-tuning strategies for computational graphs
- Determining optimal scenarios for switching between TVM and TIK
Memory Optimization Strategies
- Controlling memory layout and buffer placement
- Methods to minimize on-chip memory consumption
- Best practices for asynchronous execution and resource reuse
Real-World Deployments and Case Studies
- Case study: Performance tuning for smart city camera pipelines
- Case study: Optimizing inference stacks for autonomous vehicles
- Frameworks for iterative profiling and sustained performance improvement
Conclusion and Path Forward
Requirements
- Solid grasp of deep learning model architectures and training processes
- Practical experience with model deployment via CANN, TensorFlow, or PyTorch
- Proficiency in Linux CLI, shell scripting, and Python programming
Target Audience
- AI performance engineers
- Specialists in inference optimization
- Developers focused on edge AI or real-time system applications
14 Hours
