designing data intensive applications book

The chapter on replication explains of single-, multi-, and no- leader replication, touching of potential problems you might face with them, such as “ split brain“, “ seeing into the future” because of the replication lag, etc, and solutions to those problems.

This chapter is looking at different types of encoding, including popular JSON, XML but also those that are language specific or those that are used to write data to disk and also all kinds of problems that come with backwards compatibility and rolling upgrades.

In this chapter the author also talks about two types of categories of databases – those optimized for transaction processing and those optimized for analysis.

And then expanding this to log-structured databases (Bitcask, SSTables, LSM-trees, Cassandra, Lucene, etc) and to update-in-place ones, with B-trees being the main example of these and core for effectively all relational databases.

Awesome chapter where the author builds database starting with simplest possible file commands like:.

Data models as in three main types of databases: document, relational, and graph, and sometimes full-text search can be considered as another data model type.

Maintainable – better life for engineering and operations.

Scalable – strategies for keeping performance good under load.

Reliable – system works correctly, even when faults happen.

Martin Kleppmann brings in main terminology and fundamentals in this chapter:.

Reliable, Scalable, and Maintainable Applications.

Here is the list of all of the chapters, with some of the notes, and random comments from me (not well structured, but gives a good idea on what the book is about):

The book is also based on lots of scientific research which is properly referenced so you can always go deeper into any of given topics if you would like to.

The book is very crisp and well structured dive into architecture of under-the-hood workings of distributed data systems, such as databases, caches, queues focusing on fundamentals rather than specific implementations. I personally made three small tech talks within my tech team based on this book and plan to maybe do few more. Also it is probably the best one you can use to prepare for system design interviews in big tech companies, so take a note.

The book “Designing Data-Intensive Applications” is definitely one of the best technical software engineering books I’ve ever read.

Designing data intensive applications book