As systems grow, data becomes expensive. You pay to store it on disk, move it over the network, keep it in memory, back it up, and read it later.

Data compression reduces that cost by storing or sending the same information using fewer bytes.

Fewer bytes can mean cheaper storage, less network traffic, faster reads from disk, more items in cache, and smaller backups.

Compression is not free. It uses CPU to make data smaller, and reads often need an extra decompression step before the data can be used.

A good design uses compression where bytes are expensive and avoids it where CPU time, request speed, or fast lookups matter more.

In this chapter, we will look at how compression works, where systems use it, and how to decide whether it is worth the tradeoff.