Cisco
Product
UCS
Designing Cisco Systems for AI (DCSAI)

This course provides an in-depth understanding of designing Cisco networks tailored for artificial intelligence (AI) applications and solutions. Participants will learn how to leverage Cisco technologies to design, build and operate next-generation network infrastructure to support AI workloads effectively. The course covers key concepts, best practices, and practical implementation strategies for AI-ready networks using Cisco solutions.

About the course

Prerequisites:

The knowledge and skills that the learner should have before attending this course are as follows:

  • CCNA or equivalent experience
  • Basic Knowledge of AI Systems
  • Storage Admin Familiarity

Course Objectives:

Upon completing this course, the learner will be able to meet these overall objectives:

  • Understand Cisco Networking Technologies for AI
  • Describe AI and Network Design
  • Understand AI Workloads and Network Requirements
  • Learn Security and Compliance for AI Networks
  • Understanding the components if AI Infrastructure
Course content

Module 1: Introduction to AI and Network Design

  • Understanding the emergence of AI and its practical applications
  • Overview of Cisco technologies for AI integration
  • Challenges and considerations in designing AI-ready networks

Module 2: Understanding the components if AI Infrastructure

  • High Performance GPU Cloud
  • High Performance Storage and Storage Networks
  • High Performance Ethernet Networks
  • High Performance Compute
  • Data Processing and Machine Learning Frameworks
  • MLOps Platforms

Module 3: Understanding AI Workloads and Network Requirements

  • Kubernetes vs. Docker/Container
  • Workload characterization and Generative AI Workload
  • Distributed Workloads and Computing
  • Network architecture requirements for AI applications
  • Scalability, performance, and latency considerations
  • Integration of high-speed data pipelines for AI model training and inference

Module 4: Cisco Networking Technologies for AI

  • Deep dive into Cisco switches, routers, and data center fabrics
  • Introduction to SONiC and SONiC on Cisco 8000
  • Deep Dive into SONiC for AI Infrastructure Networking in the Hybrid Cloud Environments
  • Neural Networks and Parallelism
  • Cisco NDFC for network automation and policy-based management on Nexus Spine and Leaf Architecture
  • Integration of Cisco UCS for AI workload optimization

Module 5: Security and Compliance for AI Networks

  • Threat landscape for AI environments
  • Implementing security controls for data privacy and confidentiality
  • Compliance considerations for AI data handling and processing

Module 6: Network Optimization Techniques for AI

  • Quality of Service (QoS) for prioritizing AI traffic
  • Network slicing for efficient resource allocation
  • Load balancing and traffic engineering for AI workloads

Module 7: Monitoring and Analytics for AI Networks

  • Utilizing Cisco Observability Solutions for AI Networks
  • AI-driven analytics for proactive network management
  • MLOps Tools
  • Troubleshooting and optimization techniques for AI workloads connectivity challenges

Module 8: Case Studies and Best Practices

  • Real-world examples of AI network designs using Cisco technologies
  • Best practices for deploying and managing AI-ready networks
  • Lessons learned and future trends in AI network design

Module 9: Hands-on Labs and Practical Exercises

  • Lab exercises covering network design, configuration, and optimization
  • Simulation of AI workloads to test network performance
  • Troubleshooting scenarios and best practices demonstration

Key Takeaways:

  • Resources for further learning and certification pathways
  • Recommendations for implementing AI-ready networks in real-world environments
Who Should Attend

The primary audience for this course is as follows:

  • Network Solutions Architects
  • Presales Architects
  • Systems Engineers
  • Channel Partners
  • AI Infrastructure Administrators
  • AI Engineers
  • ML Engineers
  • Technical Managers and Decision Makers