From: | Thomas Munro <thomas(dot)munro(at)gmail(dot)com> |
---|---|
To: | 陈宗志 <baotiao(at)gmail(dot)com> |
Cc: | pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: Some questions about PostgreSQL’s design. |
Date: | 2024-08-20 10:09:27 |
Message-ID: | CA+hUKG+2j=kP-uh8rYo5qVmaXUThPgSpzSEHGK9QbuvOX-Lx1Q@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Tue, Aug 20, 2024 at 8:55 PM 陈宗志 <baotiao(at)gmail(dot)com> wrote:
> It seems like these issues have clear solutions, such as using
> DirectIO like InnoDB instead of buffered IO,
For this part: we recently added an experimental option to use direct
I/O (debug_io_direct). We are working on the infrastructure needed to
make it work efficiently before removing the "debug_" prefix:
prediction of future I/O through a "stream" abstraction which we have
some early pieces of already, I/O combining (see new io_combine_limit
setting), and asynchronous I/O (work in progress, basically I/O worker
processes or io_uring or other OS-specific APIs).
From | Date | Subject | |
---|---|---|---|
Next Message | Amul Sul | 2024-08-20 10:26:23 | Re: pg_verifybackup: TAR format backup verification |
Previous Message | Alexander Pyhalov | 2024-08-20 09:14:44 | Re: Asynchronous MergeAppend |