|
@AndresFreundTec | |||||
|
Probably should look at one of the profile guided binary layout optimizers (e.g.facebook's bolt) and see how much that helps.
But I think we also need to look at why our code density/locality is so shitty that we get such a huge number of faults. twitter.com/AndresFreundTe…
|
||||||
|
||||||
|
Torsten Grust
@Teggy
|
7. sij |
|
Re low code locality (in query processing, in particular):
Possibly a consequence of PostgreSQL's adoption of the Volcano iterator model?
|
||
|
|
||
|
Andres Freund (Tech)
@AndresFreundTec
|
7. sij |
|
That's a problem, but not in the workload I was testing. The query executor itself only takes up a small percentage of the time in a readonly, fully cached, pgbench -S workload.
|
||
|
|
||