Re: Changing WAL Header to reduce contention during ReserveXLogInsertLocation()
От | Pavan Deolasee |
---|---|
Тема | Re: Changing WAL Header to reduce contention during ReserveXLogInsertLocation() |
Дата | |
Msg-id | CABOikdOFM9i906PMpuE7-a3N+xqjYFS=B-9C5Ozh9+wKA-onsQ@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: Changing WAL Header to reduce contention duringReserveXLogInsertLocation() (Peter Eisentraut <peter.eisentraut@2ndquadrant.com>) |
Ответы |
Re: Changing WAL Header to reduce contention during ReserveXLogInsertLocation()
|
Список | pgsql-hackers |
On Fri, Mar 9, 2018 at 8:49 PM, Peter Eisentraut <peter.eisentraut@2ndquadrant.com> wrote:
On 2/1/18 19:21, Simon Riggs wrote:
> If we really can't persuade you of that, it doesn't sink the patch. We
> can have the WAL pointer itself - it wouldn't save space but it would
> at least alleviate the spinlock.
Do you want to send in an alternative patch that preserves the WAL
pointer and only changes the locking?
Sorry for the delay. Here is an updated patch which now replaces xl_prev with xl_curr, thus providing similar safeguards against corrupted or torn WAL pages, but still providing benefits of atomic operations.
Patched:
======
#clients #tps
1 34.195311
2 29.001584
4 31.712009
8 35.489272
16 41.950044
Master:
======
#clients #tps
1 24.128004
2 12.326135
4 8.334143
8 16.035699
16 8.502794
Thanks,
Pavan
Pavan Deolasee http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services
PostgreSQL Development, 24x7 Support, Training & Services
Вложения
В списке pgsql-hackers по дате отправления: