Do Flink on Web with FLOW
summary
A web service named FLOW lets users build Flink jobs on a drag-and-drop GUI and run them or download an auto-generated Maven project. Demo shows preview fetching 1,000 records per Kafka partition (3,000 total), filtering out invalid NYC coordinates down to ~2,918, then sliding-window aggregation (10-minute windows every 5 minutes) and a threshold filter leaving 20 cells marked popular. Users upload uploadable UDF jars, see preview computed with per-session Flink Table API, and push results to Elasticsearch/Kibana for charts and maps.
AI-written summary · about
chapters (8)
- 0:00Introduction to Flow
- 3:11Creating a New Flow
- 5:50Data Preview and Setup
- 7:01Building the Pipeline
- 10:22Implementing Logic with UDFs
- 12:50Advanced Transformations
- 14:40Code Generation and Deployment
- 15:55Technical Architecture
related talks
46mFlink SQL in 2020: Time to show off!
40mFrom concept to production; the fastest way to develop & deploy Apache Flink jobs
43mLeverage the power and simplicity of SQL on Apache Flink
27mIntroduction to Flink in 30 minutes
39mBetter, Faster, Stronger Streaming: Your First Dive into Flink SQL
38mApache Flink: A Very Quick Guide
45mApache Flink — Real-Time Data Processing
42mApache Flink deep-dive #bbuzz