Cloud Data Architecture

Build Scalable, AI-Ready Data Platforms

Cloud Data Architecture is the strategic design of data systems in the cloud that transform raw data into meaningful, actionable insights. This course provides a comprehensive foundation in building modern, cloud-native data platforms using scalable, secure, and high-performance architectures. You’ll learn how to leverage managed cloud services, decouple compute and storage, and implement elastic scaling to support today’s data-driven and AI-powered business environments.

As organizations accelerate digital transformation, cloud data architecture has become essential for enabling real-time analytics, machine learning, and AI innovation. Traditional on-premises systems can’t keep pace with the volume, variety, and velocity of modern data. This course explores how cloud-based solutions—such as data lakes, data warehouses, and lakehouse architectures—unlock agility, cost-efficiency, and advanced analytics capabilities.
You’ll also dive into emerging trends like semi-structured and unstructured data processing, vector embeddings, retrieval-augmented generation (RAG), and infrastructure optimized for large language models (LLMs) and intelligent agents—key components of next-generation data ecosystems.

Course Highlights:

• Foundations of Cloud Data Architecture: Cloud-Native Platforms
• Architectural Evolution of Cloud Platforms: Virtualization, Containers, Microservices, and Managed Services
• Data Platform Architectures
• Architectural Patterns for Modern Data Systems: Layered Architectures, Medallion Design, and Domain-Oriented Data Organization
• Compute and Storage Architecture: Elasticity, Workload Isolation, Performance, and Cost Optimization
• Data Modeling for Analytics: Relational, Dimensional, and Hybrid Approaches in Cloud Platforms
• Semantic Views and Semantic Layers: Reusable Metrics, Dimensions, and Governed Access for Reporting and Self-Service Analytics
• Architectures for Semi-Structured Data
• Architectures for Unstructured and Multi-Modal Data: Documents, Images, Staged Files, and External Object Storage

Course Learning Outcomes: 

• Explain the core principles of cloud-native data architecture, including virtualization, managed services, separation of compute and storage, elasticity, workload isolation, and consumption-based pricing
• Evaluate the roles of data lakes, data warehouses, lakehouses, and the Data Cloud in modern analytics ecosystems
• Design a layered cloud data architecture using Snowflake and AWS S3 to support ingestion, storage, duration, governance, and analytics consumption
• Design semantic views and governed reporting layers that support reusable metrics, dimensions, and self-service analytics
• Assess architectural options for work flow automation, security, access control, and secure data sharing in enterprise cloud data platforms
• Explain how modern cloud data architectures support AI use cases through vector embeddings, RAG patterns, Cortex AI services, and LLM-oriented data design

Course Typically Offered: Online in Fall and Spring quarters.

Prerequisites: Basic proficiency in SQL, including SELECT, JOIN, and GROUP BY, and familiarity with relational database concepts. Prior exposure to cloud computing concepts is helpful but not required.

Next Step: After completing this course, consider taking other courses in our Database Management Certificate to continue learning.

More Information: For more information about this course, please email unex-techdata@ucsd.edu.