All of lore.kernel.org
 help / color / mirror / Atom feed
From: Remi Gauvin <remi@georgianit.com>
To: Zygo Blaxell <ce3g8jdj@umail.furryterror.org>,
	linux-btrfs <linux-btrfs@vger.kernel.org>
Subject: Re: Btrfs autodefrag wrote 5TB in one day to a 0.5TB SSD without a measurable benefit
Date: Tue, 15 Mar 2022 15:22:43 -0400	[thread overview]
Message-ID: <97800cf4-b96d-27f9-1ed9-b508501e5532@georgianit.com> (raw)
In-Reply-To: <YjDgKzAx/tawKHCz@hungrycats.org>

On 2022-03-15 2:51 p.m., Zygo Blaxell wrote:

> The main advantage of larger extents is smaller metadata, and it doesn't
> matter very much whether it's SSD or HDD.  Adjacent extents will be in
> the same metadata page, so not much is lost with 256K extents even on HDD,
> as long as they are physically allocated adjacent to each other.
> 

When I tried enabling compress-force on my HDD storage, it *killed*
sequential read performance.  I could write a file out at over
100MB/s... but trying to read that same file sequentially would trash
the drives with less than 5MB/s actually being able to be read.


No such problems were observed on ssd storage.

I was under the impression this problem was caused trying to read files
with the 127k extents,, which, for whatever reason, could not be done
without excessive seeking.



  reply	other threads:[~2022-03-15 19:22 UTC|newest]

Thread overview: 71+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-03-06 15:59 Btrfs autodefrag wrote 5TB in one day to a 0.5TB SSD without a measurable benefit Jan Ziak
2022-03-07  0:48 ` Qu Wenruo
2022-03-07  2:23   ` Jan Ziak
2022-03-07  2:39     ` Qu Wenruo
2022-03-07  7:31       ` Qu Wenruo
2022-03-10  1:10         ` Jan Ziak
2022-03-10  1:26           ` Qu Wenruo
2022-03-10  4:33             ` Jan Ziak
2022-03-10  6:42               ` Qu Wenruo
2022-03-10 21:31                 ` Jan Ziak
2022-03-10 23:27                   ` Qu Wenruo
2022-03-11  2:42                     ` Jan Ziak
2022-03-11  2:59                       ` Qu Wenruo
2022-03-11  5:04                         ` Jan Ziak
2022-03-11 16:31                           ` Jan Ziak
2022-03-11 20:02                             ` Jan Ziak
2022-03-11 23:04                             ` Qu Wenruo
2022-03-11 23:28                               ` Jan Ziak
2022-03-11 23:39                                 ` Qu Wenruo
2022-03-12  0:01                                   ` Jan Ziak
2022-03-12  0:15                                     ` Qu Wenruo
2022-03-12  3:16                                     ` Zygo Blaxell
2022-03-12  2:43                                 ` Zygo Blaxell
2022-03-12  3:24                                   ` Qu Wenruo
2022-03-12  3:48                                     ` Zygo Blaxell
2022-03-14 20:09                         ` Phillip Susi
2022-03-14 22:59                           ` Zygo Blaxell
2022-03-15 18:28                             ` Phillip Susi
2022-03-15 19:28                               ` Jan Ziak
2022-03-15 21:06                               ` Zygo Blaxell
2022-03-15 22:20                                 ` Jan Ziak
2022-03-16 17:02                                   ` Zygo Blaxell
2022-03-16 17:48                                     ` Jan Ziak
2022-03-17  2:11                                       ` Zygo Blaxell
2022-03-16 18:46                                 ` Phillip Susi
2022-03-16 19:59                                   ` Zygo Blaxell
2022-03-20 17:50                             ` Forza
2022-03-20 21:15                               ` Zygo Blaxell
2022-03-08 21:57       ` Jan Ziak
2022-03-08 23:40         ` Qu Wenruo
2022-03-09 22:22           ` Jan Ziak
2022-03-09 22:44             ` Qu Wenruo
2022-03-09 22:55               ` Jan Ziak
2022-03-09 23:00                 ` Jan Ziak
2022-03-09  4:48         ` Zygo Blaxell
2022-03-07 14:30 ` Phillip Susi
2022-03-08 21:43   ` Jan Ziak
2022-03-09 18:46     ` Phillip Susi
2022-03-09 21:35       ` Jan Ziak
2022-03-14 20:02         ` Phillip Susi
2022-03-14 21:53           ` Jan Ziak
2022-03-14 22:24             ` Remi Gauvin
2022-03-14 22:51               ` Zygo Blaxell
2022-03-14 23:07                 ` Remi Gauvin
2022-03-14 23:39                   ` Zygo Blaxell
2022-03-15 14:14                     ` Remi Gauvin
2022-03-15 18:51                       ` Zygo Blaxell
2022-03-15 19:22                         ` Remi Gauvin [this message]
2022-03-15 21:08                           ` Zygo Blaxell
2022-03-15 18:15             ` Phillip Susi
2022-03-16 16:52           ` Andrei Borzenkov
2022-03-16 18:28             ` Jan Ziak
2022-03-16 18:31             ` Phillip Susi
2022-03-16 18:43               ` Andrei Borzenkov
2022-03-16 18:46               ` Jan Ziak
2022-03-16 19:04               ` Zygo Blaxell
2022-03-17 20:34                 ` Phillip Susi
2022-03-17 22:06                   ` Zygo Blaxell
2022-03-16 12:47 ` Kai Krakow
2022-03-16 18:18   ` Jan Ziak
2022-06-17  0:20 Jan Ziak

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=97800cf4-b96d-27f9-1ed9-b508501e5532@georgianit.com \
    --to=remi@georgianit.com \
    --cc=ce3g8jdj@umail.furryterror.org \
    --cc=linux-btrfs@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.