#005 - Scaling Databases Every Which Way with Glauber Costa
Backend Banter
00:00
How to Optimize for Reading Larger Buffers
The patterns of utilization with spinning discs and NVMe were so different that you had to throw your intuition into the trash. You can optimize for reading larger buffers, but it also maxes out in a couple of megabytes. It doesn't make sense to put a terabyte file sequential anymore, because you don't need to do sequential scan. But obviously, this is a stupid example, because it just makes the semaphore bigger.
Transcript
Play full episode
