Linux-NFS Archive mirror
 help / color / mirror / Atom feed
From: Lionel Cons <lionelcons1972@gmail.com>
To: linux-nfs <linux-nfs@vger.kernel.org>
Subject: Re: Suggestions to drastically reduce Linux nfsd I/O latency on BTRFS?
Date: Tue, 12 May 2026 16:52:34 +0200	[thread overview]
Message-ID: <CAPJSo4UFrWbuC1hKNg8oxnJtSZAE=xEVv5T_UfuSOP1uZgeQEA@mail.gmail.com> (raw)
In-Reply-To: <e44aa868-5ec8-4c35-b5bc-5066487a28be@app.fastmail.com>

On Tue, 12 May 2026 at 15:23, Chuck Lever <cel@kernel.org> wrote:
>
>
> On Tue, May 12, 2026, at 5:47 AM, Lionel Cons wrote:
> > Are there suggestions on how to reduce Linux nfsd I/O latency on
> > BTRFS?
>
> For us mere humans, the first step is you need to root-cause
> your issue. But consider using an LLM to help you with this
> process.
>
> Example questions to guide your analysis:
>  - What particular NFS operations are taking too long?

Read, Write, issued to serve single-file disk images with bhyve and qemu

>  - How full is your exported file system?

9% full, 2 PiB total size

>  - What backing storage is in use, and what mount options?

Very, very big hardware RAID0 with battery-backed ram cache, shows up
as one single SCSI disk

>  - What transport (TCP? RDMA?)

TCP6

>  - How much server CPU is available when the it appears slow?

64 CPU cores, 512GB RAM

>  - If you are using NFSv4, are there other clients that emit
>    conflicting OPENs and LOCKs?

No, just read and write are slower, compared to doing the same setup
with something like ISCSI.

Lionel

      reply	other threads:[~2026-05-12 14:53 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-05-12  9:47 Suggestions to drastically reduce Linux nfsd I/O latency on BTRFS? Lionel Cons
2026-05-12 13:22 ` Chuck Lever
2026-05-12 14:52   ` Lionel Cons [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CAPJSo4UFrWbuC1hKNg8oxnJtSZAE=xEVv5T_UfuSOP1uZgeQEA@mail.gmail.com' \
    --to=lionelcons1972@gmail.com \
    --cc=linux-nfs@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).