Re: Show WAL write and fsync stats in pg_stat_io

On Mon, Nov 06, 2023 at 03:35:01PM +0300, Nazir Bilal Yavuz wrote:
> Results are:
>
> ╔═════════╦═══════════════════════════════╦════════╗
> ║         ║      track_wal_io_timing      ║        ║
> ╠═════════╬═══════════════╦═══════════════╬════════╣
> ║  clock  ║       on      ║      off      ║ change ║
> ║ sources ║               ║               ║        ║
> ╠═════════╬═══════════════╬═══════════════╬════════╣
> ║   tsc   ║               ║               ║        ║
> ║         ║ 514814.459170 ║ 519826.284139 ║   %1   ║
> ╠═════════╬═══════════════╬═══════════════╬════════╣
> ║   hpet  ║               ║               ║        ║
> ║         ║ 132116.272121 ║ 141820.548447 ║   %7   ║
> ╠═════════╬═══════════════╬═══════════════╬════════╣
> ║ acpi_pm ║               ║               ║        ║
> ║         ║ 394793.092255 ║ 403723.874719 ║   %2   ║
> ╚═════════╩═══════════════╩═══════════════╩════════╝

Thanks for the tests.  That's indeed noticeable under this load.
Better to keep track_io_timing and track_wal_io_timing as two
separated beasts, at least that's clear.
--
Michael