[TBD]Understanding Apache flink

Flink ecosystem

• It has runtime core processing which does stream processing

• It has APIs related to data

  • Batch processing API - Can be used with JAVA, Scala and Python

  • Stream processing API - Can be used with JAVA, Scala (Validate it) .

• Allows to process following operations on the data

  • Relational ( For Structured) - TABLE API

  • Graph data processing - GELLY API

  • Machine learning operations - FLINK ML

• For visulalization of data, provides Zeeplin libray

  • Provides interactive data analysis

• Deployment

  • With Java VM

  • In cluster, as standalone or as distributed cluster node mode

  • Any Cloud

• Storage

  • Local, HDFS or S3

• Database integration

  • MongoDB

  • Kafka

  • Flume

  • RabbitMQ