[MDEV-21955] Packed sort keys: Evaluate performance of dynamic-length mem-comparable data encoding Created: 2020-03-16 Updated: 2020-08-17 Resolved: 2020-08-17 |
|
| Status: | Closed |
| Project: | MariaDB Server |
| Component/s: | Optimizer |
| Fix Version/s: | N/A |
| Type: | Task | Priority: | Major |
| Reporter: | Sergei Petrunia | Assignee: | Sergei Petrunia |
| Resolution: | Fixed | Votes: | 0 |
| Labels: | None | ||
| Issue Links: |
|
||||||||
| Description |
|
Based on the experiments in |
| Comments |
| Comment by Sergei Petrunia [ 2020-03-16 ] |
|
A branch with a crude patch implementing this: Benchmark results: |
| Comment by Sergei Petrunia [ 2020-03-17 ] |
|
Summary of benchmark results so far:
Looks like this is worth considering implementing for 10.6... (Note that this encoding could also be used with priority queue. If one is sorting VARCHAR(255) there should be a gain from sort keys being smaller). |
| Comment by Sergei Petrunia [ 2020-03-18 ] |
|
Relative popularities of charsets according to the feedback plugin: https://gist.github.com/cvicentiu/143ad0750c59b523a47a1aebe0706068 |