All of lore.kernel.org
 help / color / mirror / Atom feed
From: Patrick Goetz <pgoetz@math.utexas.edu>
To: Chuck Lever III <chuck.lever@oracle.com>,
	Bruce Fields <bfields@fieldses.org>
Cc: Daire Byrne <daire@dneg.com>,
	Linux NFS Mailing List <linux-nfs@vger.kernel.org>
Subject: Re: parallel file create rates (+high latency)
Date: Tue, 25 Jan 2022 15:50:05 -0600	[thread overview]
Message-ID: <42867c2c-1ab3-9bb6-0e5a-57d13d667bc6@math.utexas.edu> (raw)
In-Reply-To: <F7C721F7-D906-426F-814F-4D3F34AD6FB1@oracle.com>



On 1/25/22 09:30, Chuck Lever III wrote:
> 
> 
>> On Jan 25, 2022, at 8:59 AM, J. Bruce Fields <bfields@fieldses.org> wrote:
>>
>> On Tue, Jan 25, 2022 at 12:52:46PM +0000, Daire Byrne wrote:
>>> Yea, it does seem like the server is the ultimate arbitrar and the
>>> fact that multiple clients can achieve much higher rates of
>>> parallelism does suggest that the VFS locking per client is somewhat
>>> redundant and limiting (in this super niche case).
>>
>> It doesn't seem *so* weird to have a server with fast storage a long
>> round-trip time away, in which case the client-side operation could take
>> several orders of magnitude longer than the server.
>>
>> Though even if the client locking wasn't a factor, you might still have
>> to do some work to take advantage of that.  (E.g. if your workload is
>> just a single "untar"--it still waits for one create before doing the
>> next one).
> 
> Note that this is also an issue for data center area filesystems, where
> back-end replication of metadata updates makes creates and deletes as
> slow as if they were being done on storage hundreds of miles away.
> 
> The solution of choice appears to be to replace tar/rsync and such
> tools with versions that are smarter about parallelizing file creation
> and deletion.
> 

Are these tools available to mere mortals?  If so, what are they called. 
  This is a problem I'm currently dealing with; trying to back up 
hundreds of terabytes of image data.


> 
> --
> Chuck Lever
> 
> 
> 

  reply	other threads:[~2022-01-25 21:50 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-01-23 23:53 parallel file create rates (+high latency) Daire Byrne
2022-01-24 13:52 ` Daire Byrne
2022-01-24 19:37 ` J. Bruce Fields
2022-01-24 20:10   ` Daire Byrne
2022-01-24 20:50     ` J. Bruce Fields
2022-01-25 12:52       ` Daire Byrne
2022-01-25 13:59         ` J. Bruce Fields
2022-01-25 15:24           ` Daire Byrne
2022-01-25 15:30           ` Chuck Lever III
2022-01-25 21:50             ` Patrick Goetz [this message]
2022-01-25 21:58               ` Chuck Lever III
2022-01-25 21:59               ` Bruce Fields
2022-01-25 22:11                 ` Patrick Goetz
2022-01-25 22:41                   ` Daire Byrne
2022-01-25 23:01                     ` Patrick Goetz
2022-01-25 23:25                       ` Daire Byrne
2022-01-25 21:15   ` Patrick Goetz
2022-01-25 21:20     ` J. Bruce Fields
2022-01-26  0:02       ` NeilBrown
2022-01-26  0:28         ` Daire Byrne
2022-01-26  2:57         ` J. Bruce Fields
2022-02-08 18:48           ` Daire Byrne
2022-02-10 18:19             ` Daire Byrne
2022-02-11 15:59               ` J. Bruce Fields
2022-02-17 19:50                 ` Daire Byrne
2022-02-18  7:46                   ` NeilBrown
2022-02-21 13:59                     ` Daire Byrne
2022-04-25 13:00                       ` Daire Byrne
2022-04-25 13:22                         ` J. Bruce Fields
2022-04-25 15:24                           ` Daire Byrne
2022-04-25 16:02                             ` J. Bruce Fields
2022-04-25 16:47                               ` Daire Byrne
2022-04-26  1:36                                 ` NeilBrown
2022-04-26 12:29                                   ` Daire Byrne
2022-04-28  5:46                                     ` NeilBrown
2022-04-29  7:55                                       ` Daire Byrne

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=42867c2c-1ab3-9bb6-0e5a-57d13d667bc6@math.utexas.edu \
    --to=pgoetz@math.utexas.edu \
    --cc=bfields@fieldses.org \
    --cc=chuck.lever@oracle.com \
    --cc=daire@dneg.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 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.