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 14 hours
Course Outline
Introduction to the Huawei AI Ecosystem
- Overview of Ascend AI processors: 310, 910, and 910B
- Key components: MindSpore, CANN, and AscendCL
- Architectural principles and industry positioning
The Function of CANN in Huawei's AI Infrastructure
- Defining CANN: SDK objectives and internal structure
- Leveraging ATC, TBE, and AscendCL for model compilation and execution
- Role of CANN in inference optimization and deployment strategies
MindSpore Framework: Overview and Design
- Workflows for training and inference within MindSpore
- Graph mode, PyNative, and hardware abstraction layers
- Seamless integration with Ascend NPUs via the CANN backend
Managing the AI Lifecycle on Ascend: From Training to Production
- Developing models in MindSpore or migrating from third-party frameworks
- Compiling and exporting models using ATC
- Deploying optimized OM models on Ascend hardware via AscendCL
Comparative Analysis with Competing AI Stacks
- Positioning MindSpore against PyTorch and TensorFlow
- Contrast between Ascend-centric deployment workflows and GPU-based alternatives
- Assessing enterprise opportunities and potential constraints
Enterprise Integration and Use Cases
- Applications in smart manufacturing, public sector AI, and telecommunications
- Evaluating scalability, regulatory compliance, and ecosystem maturity
- Implementing cloud/on-prem hybrid architectures with the Huawei stack
Key Takeaways and Roadmap for Next Steps
Requirements
- Working knowledge of AI workflows or platform architecture
- Fundamental understanding of model training and deployment processes
- No prior practical experience with CANN or MindSpore is necessary
Target Audience
- AI platform evaluators and infrastructure architects
- AI/ML DevOps specialists and pipeline integrators
- Technology leaders and strategic decision-makers
