From: | Xuneng Zhou <xunengzhou(at)gmail(dot)com> |
---|---|
To: | Jeff Davis <pgsql(at)j-davis(dot)com>, Andres Freund <andres(at)anarazel(dot)de> |
Cc: | pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: aio/README.md comments |
Date: | 2025-08-30 04:02:49 |
Message-ID: | CABPTF7WmFQ6S1F6CmzhZOm3EXYVv9408s=ZfQhUeqVcFnZyfSA@mail.gmail.com |
Views: | Whole Thread | Raw Message | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Hi,
On Sat, Aug 30, 2025 at 6:24 AM Jeff Davis <pgsql(at)j-davis(dot)com> wrote:
>
> On Fri, 2025-08-29 at 12:32 -0400, Andres Freund wrote:
> > I don't really see an advantage of sync in those cases either.
>
> It seems a bit early to say that it's just there for debugging. But
> it's just in a README, so I won't argue the point.
>
> I attached some proposed changes based on my understanding.
>
> Regards,
> Jeff Davis
>
+ These memory copies can become the bottleneck when the
+ underlying storage has high enough throughput, which is common for
+ solid-state drives or fast network block devices.
Would it be helpful to be more specific on the types of solid-state
drives like PCIe/NVMe SSD?
SATA SSDs' ~600 MB/s theoretical ceiling bandwidth might not be high
enough. The rate of out-of-date
could be less concerning since the advancement of hardwares has slowdowned.
Best,
Xuneng
From | Date | Subject | |
---|---|---|---|
Next Message | Elliot Haisley | 2025-08-30 04:18:05 | [PATCH] meson: Update meson to enable building postgres as a subproject |
Previous Message | Nathan Bossart | 2025-08-30 02:51:38 | Re: Improve LWLock tranche name visibility across backends |