May 2017
Beginner
416 pages
10h 37m
English
Traditionally, a query had to run on a single CPU. While this was just fine in the OLTP world, it started to be a problem for analytical applications, which were bound to the speed provided by a single core. With PostgreSQL 9.6, parallel queries were introduced. Of course, implementing parallel queries was hard and so a lot of infrastructure has already been implemented over the years. All this infrastructure is now available to provide the end user with parallel sequential scans. The idea is to make many CPUs work on complicated WHERE conditions during a sequential scan. Version 9.6 also allowed for parallel aggregates and parallel joins. Of course, there is a lot of work left, but we are already looking at a ...
Read now
Unlock full access