Course Outline
Fundamentals of the Databricks Platform and Lakehouse
- Databricks Lakehouse architecture and key components.
- Organizing workspaces and catalogs.
Databricks Workspace and Notebooks
- Navigating the workspace and developing via notebooks.
- Structuring code for reuse across notebooks.
Apache Spark Architecture and Execution
- Spark runtime architecture and execution model.
- Understanding lazy evaluation and the job DAG.
PySpark DataFrames and the DataFrame API
- DataFrame abstractions and schemas.
- Core DataFrame operations and column expressions.
Translating SQL to PySpark DataFrames
- Mapping core SQL clauses to DataFrame operations.
- Implementing window functions and aggregations in PySpark.
Reading and Writing Data in Databricks
- Accessing data from various file formats and databases.
- Writing and partitioning data within the Lakehouse.
Delta Lake and Table Management
- Understanding Delta tables and ACID transactions.
- Leveraging time travel and schema evolution.
Data Cleaning and Transformation Patterns
- Data cleaning processes and type conversions.
- Developing reusable transformation logic.
User-Defined Functions and Modular Code
- Using Python UDFs and pandas UDFs.
- Modularizing procedural logic into functions.
Performance Tuning and Optimization
- Strategies for partitioning and caching.
- Identifying bottlenecks using the Spark UI.
Fundamentals of Structured Streaming
- Comparing batch versus streaming processing models.
- Working with streaming DataFrames and basic aggregations.
Databricks Jobs and Workflow Orchestration
- Scheduling notebooks as jobs and tasks.
- Constructing multi-step workflows with dependencies.
Unity Catalog and Data Governance
- Unity Catalog architecture and namespaces.
- Managing access control and data lineage.
Testing, Debugging, and Production Practices
- Unit testing PySpark logic.
- Debugging techniques and code quality standards.
End-to-End Financial Services Use Cases
- Building a comprehensive banking ETL pipeline.
- Migrating legacy SQL processes to PySpark.
Migrating SQL Workloads to PySpark
- Planning patterns and migration strategies.
- Incremental conversion of SQL workflows to PySpark.
Requirements
- Proficiency in Python programming, including functions and data types.
- Understanding of SQL concepts such as joins, aggregations, and subqueries.
- No previous experience with Databricks or PySpark is necessary.
Audience
- Data engineers, analysts, and general data professionals.
- Teams transitioning existing SQL-based workflows to Databricks and PySpark.
Testimonials (1)
I liked that it was practical. Loved to apply the theoretical knowledge with practical examples.
