public inbox for [email protected]
 help / color / mirror / Atom feed
From: Christoph Hellwig <[email protected]>
To: Keith Busch <[email protected]>
Cc: Christoph Hellwig <[email protected]>, Keith Busch <[email protected]>,
	[email protected], [email protected],
	[email protected], [email protected],
	[email protected], [email protected],
	[email protected], [email protected],
	Hannes Reinecke <[email protected]>
Subject: Re: [PATCHv10 9/9] scsi: set permanent stream count in block limits
Date: Wed, 30 Oct 2024 17:57:08 +0100	[thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: <[email protected]>

On Wed, Oct 30, 2024 at 10:42:59AM -0600, Keith Busch wrote:
> On Wed, Oct 30, 2024 at 04:50:52PM +0100, Christoph Hellwig wrote:
> > On Wed, Oct 30, 2024 at 09:48:39AM -0600, Keith Busch wrote:
> > > What??? You said to map the temperature hints to a write stream. The
> > > driver offers that here. But you specifically don't want that? I'm so
> > > confused.
> > 
> > In bdev/fops.c (or file systems if they want to do that) not down in the
> > driver forced down everyones throat.  Which was the whole point of the
> > discussion that we're running in circles here.
> 
> That makes no sense. A change completely isolated to a driver isn't
> forcing anything on anyone. It's the upper layers that's forcing this
> down, whether the driver uses it or not: the hints are already getting
> to the driver, but the driver currently doesn't use it.

And once it uses by default, taking it away will have someone scream
regresion, because we're not taking it away form that super special
use case.

> Here's something recent from rocksdb developers running ycsb worklada
> benchmark. The filesystem used is XFS.

Thanks for finally putting something up.

> It sets temperature hints for different SST levels, which already
> happens today. The last data point made some minor changes with
> level-to-hint mapping.

Do you have a pointer to the changes?

> Without FDP:
> 
> WAF:        2.72
> IOPS:       1465
> READ LAT:   2681us
> UPDATE LAT: 3115us
> 
> With FDP (rocksdb unmodified):
> 
> WAF:        2.26
> IOPS:       1473
> READ LAT:   2415us
> UPDATE LAT: 2807us
> 
> With FDP (with some minor rocksdb changes):
> 
> WAF:        1.67
> IOPS:       1547
> READ LAT:   1978us
> UPDATE LAT: 2267us

Compared to the Numbers Hans presented at Plumbers for the Zoned XFS code,
which should work just fine with FDP IFF we exposed real write streams,
which roughly double read nad wirte IOPS and reduce the WAF to almost
1 this doesn't look too spectacular to be honest, but it sure it something.

I just wish we could get the real infraѕtructure instead of some band
aid, which makes it really hard to expose the real thing because now
it's been taken up and directly wired to a UAPI.
one

  reply	other threads:[~2024-10-30 16:57 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-10-29 15:19 [PATCHv10 0/9] write hints with nvme fdp, scsi streams Keith Busch
2024-10-29 15:19 ` [PATCHv10 1/9] block: use generic u16 for write hints Keith Busch
2024-10-29 17:21   ` Bart Van Assche
2024-10-29 15:19 ` [PATCHv10 2/9] block: introduce max_write_hints queue limit Keith Busch
2024-10-29 15:19 ` [PATCHv10 3/9] statx: add write hint information Keith Busch
2024-10-29 15:19 ` [PATCHv10 4/9] block: allow ability to limit partition write hints Keith Busch
2024-10-29 15:23   ` Christoph Hellwig
2024-10-29 17:25   ` Bart Van Assche
2024-10-30  4:46     ` Christoph Hellwig
2024-10-30 20:11       ` Keith Busch
2024-10-30 20:26         ` Bart Van Assche
2024-10-30 20:37           ` Keith Busch
2024-10-30 21:15             ` Bart Van Assche
2024-10-29 15:19 ` [PATCHv10 5/9] block, fs: add write hint to kiocb Keith Busch
2024-10-29 15:19 ` [PATCHv10 6/9] io_uring: enable per-io hinting capability Keith Busch
2024-10-29 15:19 ` [PATCHv10 7/9] block: export placement hint feature Keith Busch
2024-10-29 15:19 ` [PATCHv10 8/9] nvme: enable FDP support Keith Busch
2024-10-30  0:24   ` Chaitanya Kulkarni
2024-10-29 15:19 ` [PATCHv10 9/9] scsi: set permanent stream count in block limits Keith Busch
2024-10-29 15:26   ` Christoph Hellwig
2024-10-29 15:34     ` Keith Busch
2024-10-29 15:37       ` Christoph Hellwig
2024-10-29 15:38         ` Keith Busch
2024-10-29 15:53           ` Christoph Hellwig
2024-10-29 16:22             ` Keith Busch
2024-10-30  4:55               ` Christoph Hellwig
2024-10-30 15:41                 ` Keith Busch
2024-10-30 15:45                   ` Christoph Hellwig
2024-10-30 15:48                     ` Keith Busch
2024-10-30 15:50                       ` Christoph Hellwig
2024-10-30 16:42                         ` Keith Busch
2024-10-30 16:57                           ` Christoph Hellwig [this message]
2024-10-30 17:05                             ` Keith Busch
2024-10-30 17:15                               ` Christoph Hellwig
2024-10-30 17:23                             ` Keith Busch
2024-10-30 22:32                             ` Keith Busch
2024-10-30 16:59                 ` Bart Van Assche
2024-10-30 17:14                   ` Christoph Hellwig
2024-10-30 17:44                     ` Bart Van Assche
2024-10-29 17:18     ` Bart Van Assche
2024-10-30  5:42       ` Christoph Hellwig
2024-10-29 15:24 ` [PATCHv10 0/9] write hints with nvme fdp, scsi streams Christoph Hellwig

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 \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    /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