Re: Handing off SLRU fsyncs to the checkpointer

On 2020-Aug-25, Jakub Wartak wrote:

> Turning on/off the defer SLRU patch and/or fsync doesn't seem to make
> any difference, so if anyone is curious the next sets of append-only
> bottlenecks is like below:
> 
>     14.69%  postgres  postgres            [.] hash_search_with_hash_value
>             ---hash_search_with_hash_value
>                |--9.80%--BufTableLookup
>                |          ReadBuffer_common
>                |          ReadBufferWithoutRelcache
>                |          XLogReadBufferExtended
>                |          XLogReadBufferForRedoExtended
>                |          |--7.76%--btree_xlog_insert
>                |          |          btree_redo
>                |          |          StartupXLOG
>                |           --1.63%--heap_xlog_insert
>                 --4.90%--smgropen
>                           |--2.86%--ReadBufferWithoutRelcache

Looking at an earlier report of this problem I was thinking whether it'd
make sense to replace SMgrRelationHash with a simplehash table; I have a
half-written patch for that, but I haven't completed that work.
However, in the older profile things were looking different, as
hash_search_with_hash_value was taking 35.25%, and smgropen was 33.74%
of it.  BufTableLookup was also there but only 1.51%.  So I'm not so
sure now that that'll pay off as clearly as I had hoped.

-- 
Álvaro Herrera                https://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services