Lab Notes: How We Made Joins 23 Thousand Times Faster, Part Three
The final part of a three-part miniseries that looks at how we improved join performance in the CrateDB 3.0 release.
Lab Notes: How We Made Joins 23 Thousand Times Faster, Part One
We made faster joins by implementing a block hash algorithm and distributing its execution across the cluster. Find out more in this post.
Lab Notes: Adding Support for Joins on Virtual Tables and Multi-Row Subselects
We recently added support for multi-row subselects. Memory can cause bottlenecks, but fortunately, we found the solution.
Lab Notes: Implementing Non-Collapsable Subselects
CrateDB now supports non-collapsable subselects. Let us introduce you to the problem, and show you the steps Crate.io took to solve it.