hard20 pts · first answer countsdatabasesindexingperformance

Inserts slow down with UUID keys

A table uses random UUIDv4 primary keys in a B-tree. As it grows past memory size, insert throughput drops and the index bloats. What is a good fix?

A.Switch the primary key to a hash indexB.Increase the database page size so each index page holds more keys and splits happen less oftenC.Use time-ordered ids such as UUIDv7, so new keys land near the end of the indexD.Drop the primary key and rely on application checks

← Index for the few pending rows Burst at a fixed window boundary →