AWS ML Engineer Associate 3.1 Select a Deployment Infrastructure

This course provides a comprehensive understanding of model deployment in the machine learning (ML) pipeline. In the introduction, you learn about fundamental deployment concepts. Then, in the first section, you learn about essential components of a production infrastructure. This section guides you in selecting the best orchestration services for ML workflows. Then, you learn about tools and services from Amazon Web Services (AWS) that you can use during the deployment phase of the ML lifecycle. The next section covers inference infrastructure. You learn how to select the best deployment target based on key benefits. You also learn how to select the appropriate environment for training and inference based on specific requirements. Next, you explore various AWS compute instance types and learn how to differentiate between on-demand and provisioned resources. Finally, you review how to provision compute resources in production and test environments.

• Course level: Advanced
• Duration: 1 hour

Activities

• Online materials
• Knowledge check questions
• A course assessment

Course objectives

• Define model deployment in the ML pipeline.
• Describe a production infrastructure and its components.
• Compare and contrast orchestration services for ML workflows.
• Describe deployment infrastructure design considerations.
• Select the best deployment target based on key benefits.
• Describe ML model deployment strategies and their endpoint requirements.
• Select the best model deployment hosting strategies based on key benefits.
• Select multi-model or multi-container deployments based on key benefits.
• Select the best container option based on key benefits.
• List and describe AWS compute instance types for ML solutions.
• Select the best compute environment for training and inference based on specific requirements.
• Differentiate between on-demand and provisioned resources for performance and scaling issues.
• Describe how to provision compute resources in production and test environments.
• Describe methods for optimizing models on edge devices.

Intended audience

• Cloud architects
• Machine learning engineers

Recommended Skills

• Completed at least 1 year of experience using SageMaker and other AWS services for ML engineering
• Completed at least 1 year of experience in a related role, such as backend software developer, DevOps developer, data engineer, or data scientist
• A fundamental understanding of programming languages, such as Python
• Completed preceding courses in the AWS ML Engineer Associate Learning Plan

Course outline

• Section 1: Introduction
  • Lesson 1: How to Use This Course
  • Lesson 2: Domain 3 Introduction
  • Lesson 3: Course Overview
  • Lesson 4: Fundamentals of Model Deployment
• Section 2: Model Building and Deployment Infrastructure
  • Lesson 5: Building a Repeatable Framework
  • Lesson 6: Workflow Orchestration Options
• Section 3: Inference Infrastructure
  • Lesson 7: Deployment Considerations and Target Options
  • Lesson 8: Choosing a Model Inference Strategy
  • Lesson 9: Container and Instance Types for Inference
  • Lesson 10: Optimizing Deployment with Edge Computing
• Section 4: Conclusion
  • Lesson 11: Course Summary
  • Lesson 12: Assessment
  • Lesson 13: Contact Us