Skip to content

Latest commit

 

History

History
29 lines (15 loc) · 1.67 KB

KafkaIntro.md

File metadata and controls

29 lines (15 loc) · 1.67 KB

What is Kafka?

Kafka is a big data streaming service that allows multiple sources to upload and download data, such as log files or bug reports. Kafka organizes data and ensures that it is kept in order, and that all data goes where it needs to. Kafka was originally created by LinkedIn, but is now maintained by Apache.

Messages

Kafka transmits data in the form of messages, which are small files containing any type of data. Messages are created and transmitted in real-time, and some examples of messages include:

  • program crash logs
  • customer support chat records
  • forms received through a webpage

These messages are given to Kafka via API calls. Once a message is given to Kafka, the data is queued and stored in servers until users access the messages

Producers and Consumers

Programs or users that create and send data are called producers, and programs or users that read data are called consumers. There are separate APIs for producers and consumers, and you can find examples of how to use them here.

Kafka Topics

Producers and consumers subscribe to Kafka topics, which are the structure Kafka uses for data organization. A topic is a group of related data that is separated into its own sub-stream. Organizations can pick and choose who has access to which topics, which allows companies to keep data secure by limiting the number of people who have access to it.

Kafka Clusters

Data Queueing

Data Partitioning