Data Management - (Transaction|Request|Commit|Redo) Log
Table of Contents
1 - About
They permits the implementation of :
- fault tolerance (Undo/Redo)
A transaction log is made up of entries (also records), which record all changes made to data.
This log are also known as:
If we keep events in a transaction log, it starts to behave like a version control system for data (versus code).
Most databases write every single insert, update and delete operation to a transaction log. This serves as a source of truth to the database and in case of errors, the database state is recovered from them.
2 - Articles Related
3 - Database Usage
- Recover consistent state of a database
- Replicate the database (Dataguard, Streams, GoldenGate…)
- Update materialized view
If have all your logs file, you can reconstruct the entire database
4 - Distributed
5 - Management
5.1 - Transfer them
How to transfer log files into a log manager such as Kafka
- Kafka Connect
- FileBeats from Elastic
- Log Tailers
- Develop an Application's logger that send straight to (No files involved at all).
- Telegraf (for metrics oriented messages)
5.2 - Read them
6 - Implementation
7 - Documentation / Reference
- The Log: What every software engineer should know about real-time data's unifying abstraction (What is a log, Log in Database, Log in Distributed database)