Kafka and infrastructure satellites

Nowadays many people want Kafka (even when they don't need it). If the case is not trivial, the story does not end after deployment. Questions arise and there are many of them:

• How to organize monitoring;
• What Burrow is and why is it needed;
• Do we need backups;
• How to organize integration with the database;
• Why do we need Schema Registry and what is REST Proxy;
• What topology is needed in several DCs;
• Why there are a dozen replicators, how to choose. How MM differs from MM2;
• How to do nothing and have the cluster administer itself;
• What to do with data security and access.

The report will provide answers to the questions above and the motivation for choosing each specific ecosystem tool.

All tools appeared in Avito in production as a solution to specific operational problems or elimination of developer pains. All tools, open source, or internal development.