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
- Ascend AI hardware: An overview of 310, 910, and 910B chips
- Key components: MindSpore, CANN, and AscendCL
- Strategic positioning and core architectural principles
The Function of CANN in Huawei’s AI Stack
- Defining CANN: SDK objectives and internal layering
- ATC, TBE, and AscendCL: Mechanisms for model compilation and execution
- CANN’s role in supporting inference optimization and deployment
MindSpore: Overview and Architecture
- Training and inference processes within MindSpore
- Graph mode, PyNative, and hardware abstraction layers
- Connecting with Ascend NPUs through the CANN backend
AI Lifecycle on Ascend: From Training to Deployment
- Building models in MindSpore or migrating from other frameworks
- Exporting and compiling models via ATC
- Deploying on Ascend hardware using OM models and AscendCL
Benchmarking Against Competing AI Stacks
- MindSpore vs. PyTorch and TensorFlow: Distinctions in focus and market position
- Deployment processes on Ascend compared to GPU-centric stacks
- Potential benefits and constraints for enterprise applications
Enterprise Integration Use Cases
- Applications in smart manufacturing, government AI initiatives, and telecommunications
- Considerations for scalability, regulatory compliance, and ecosystem health
- Hybrid cloud/on-premises deployment strategies utilizing the Huawei stack
Conclusion and Future Directions
Requirements
- General awareness of AI workflows or platform architectures
- Foundational knowledge of model training and deployment processes
- No prior hands-on experience with CANN or MindSpore is necessary
Target Audience
- AI platform assessors and infrastructure architects
- AI/ML DevOps specialists and pipeline integration experts
- Technology leaders and strategic decision-makers
