Agenda

Ververica Bootcamp Program

The Ververica Bootcamp Program is an intensive training initiative that transforms Apache Flink users into proficient data processing professionals. By translating complex Flink concepts into practical exercises rooted in real-world scenarios, we empower participants to tackle their toughest data challenges. Leveraging Ververica Cloud services, participants gain a deep understanding of Flink and learn to optimize the scalability and efficiency of their cloud-based solutions. This program is not just about learning; it’s about mastering Apache Flink and leading the future of data processing.

Level Up Your Stream Processing Skills

This intensive, 2-day face-to-face program is designed for Apache Flink users with 2-4 years of experience who want to take their skills to the intermediate level. We'll delve into advanced Flink concepts and techniques, empowering you to build and deploy highly scalable and efficient real-time data processing pipelines. Leveraging Ververica Cloud services, you'll gain a deeper understanding of Flink and explore best practices for production deployments.

Target Audience:

Apache Flink users with 2-4 years of experience who are comfortable with core concepts and want to become proficient in advanced functionalities.

Key Topics

• Advanced Windowing Operations
• Time Management Strategies
• State Management Techniques
• Serialization Optimization
• Exactly Once Processing
• Fault Tolerance

• Enrichment Techniques
• Scalability Optimization
• Flink SQL Functions
• Table API Features
• Workflow Design
• Using Paimon Effectively

Learning Outcomes

Master Advanced Windowing Operations in Apache Flink:

• Understand and implement session windows, tumbling/sliding windows with triggers, and time management strategies (Event Time, Processing Time, Ingestion Time).

Optimize State Management for High Performance in Flink Applications:

• Apply advanced state management techniques including state partitioning and RocksDB integration.
• Optimize state size and access patterns for enhanced performance.

Improve Workflow Performance via Advanced Serialization Techniques:

• Learn how to reduce time spent serializing and deserializing data, for data sources and sinks (connectors), and over the network.

Deep Dive into Exactly-Once Processing and Failure Recovery:

• Understand the differences between at-least-once, exactly-once, and exactly-once end-to-end. Learn how to effectively use exactly-once processing when faced with bad data, infrastructure failures, and workflow bugs.

Develop Complex Real-Time Pipelines:

• Build a workflow that processes a continuous stream of events to generate both dashboard and analytics results.
• Learn how best to enrich data from a variety of data sources.
• Optimize complex workflows using pre-filtering, pruning, async I/O, broadcast streams, parallel partial enrichments, and other techniques.

Use Flink SQL & Table APIs to Implement Workflows:

• Utilize the advanced functionalities of Flink SQL, including UDFs and Table Functions, and master the Flink Table API for unified data transformations and real-time analytics.
• Compare and contrast the resulting workflow with the Java API.

Designing Optimized Workflows:

• Learn about situations where splitting a workflow into multiple components improves efficiency and reduces operational complexity.
• Learn how to use Paimon as an efficient and low-overhead data bridge between workflows.