Low performance on usb-media and DVD-RAM in sync mode

Bug #19982 reported by pato101
18
This bug affects 1 person
Affects Status Importance Assigned to Milestone
linux (Ubuntu)
Invalid
Medium
Unassigned
linux-source-2.6.20 (Ubuntu)
Won't Fix
Medium
Unassigned

Bug Description

USB sticks perform very slow if using filesystems other than reiserfs or vfat.

Same behaviour seen at DVD-RAM: with UDF filesystem is about 5 or 10 times
slower than with reiserfs.

However, if DVD-RAM with UDF is mounted async, performance is on par with
reiserfs. Reiserfs behaves almost the same using sync or async. Exactly the same
happens with usb-sticks.

How I've checked:

---- for async cases: -------------------
sudo mount -t udf -o noatime /dev/hda /media/dvdram/
time cp LargeFile /media/dvdram/tmp/.; sudo time umount /media/dvdram/
----------------------------------------

---- for sync cases: -------------------
sudo mount -t udf -o noatime,sync /dev/hda /media/dvdram/
time cp LargeFile /media/dvdram/tmp/.
----------------------------------------
(same for reiserfs formatted dvd-ram, using also the nolog option to avoid
journal updates)

Revision history for this message
Jonathan Carter (jonathan) wrote :

Do these performance issues persist with Ubuntu 6.06 LTS?

Revision history for this message
pato101 (pato101) wrote : Re: [Bug 19982] Re: Low performance on usb-media and DVD-RAM in sync mode

I'm afraid they problem persists.
However, I've been told that it does not happen with the LiveCD. I
cannot check the liveCD with the dvdram, as I have only a single
drive, but the liveCD behaves OK with the usb-stick while the
installed version does not.

On 6/15/06, Jonathan Carter <email address hidden> wrote:
> Do these performance issues persist with Ubuntu 6.06 LTS?
>
> --
> Low performance on usb-media and DVD-RAM in sync mode
> https://launchpad.net/bugs/19982
>

Revision history for this message
kko (kko) wrote :

Confirmed (on Breezy).

When "man mount" tells us that "sync" isn't even a valid option for udf filesystems, I wonder why this has an effect.

Revision history for this message
pato101 (pato101) wrote :

Please note that same happens with ext2 filesystem that supports
"sync" and "async", but not with reiserfs. The problem was first
detected on usb pendrives, where "sync" is an interesting option to be
used whereas it is not at DVD-RAM.

On 7/13/06, kko <email address hidden> wrote:
> Confirmed (on Breezy).
>
> When "man mount" tells us that "sync" isn't even a valid option for udf
> filesystems, I wonder why this has an effect.
>
> ** Changed in: Ubuntu
> Status: Unconfirmed => Confirmed
>
> --
> Low performance on usb-media and DVD-RAM in sync mode
> https://launchpad.net/bugs/19982
>

Revision history for this message
Gordon Mckeown (gordon-mckeown) wrote :

This is still a problem under Kubuntu Feisty.

An external USB 2.0 hard disk formatted with Ext3 writes at about 600KB/s with the sync option enabled.
Remounting the drive async gives around 20MB/s write speed.

Revision history for this message
Brian Murray (brian-murray) wrote :

I am assigning this bug to the 'ubuntu-kernel-team' per their bug policy. For future reference you can learn more about their bug policy at https://wiki.ubuntu.com/KernelTeamBugPolicies .

Changed in linux-source-2.6.20:
assignee: nobody → ubuntu-kernel-team
Revision history for this message
Launchpad Janitor (janitor) wrote : This bug is now reported against the 'linux' package

Beginning with the Hardy Heron 8.04 development cycle, all open Ubuntu kernel bugs need to be reported against the "linux" kernel package. We are automatically migrating this bug to the new "linux" package. However, development has already began for the upcoming Intrepid Ibex 8.10 release. It would be helpful if you could test the upcoming release and verify if this is still an issue - http://www.ubuntu.com/testing . If the issue still exists, please update this report by changing the Status of the "linux" task from "Incomplete" to "New". We appreciate your patience and understanding as we make this transition. Thanks!

Revision history for this message
Leann Ogasawara (leannogasawara) wrote :

The Ubuntu Kernel Team is planning to move to the 2.6.27 kernel for the upcoming Intrepid Ibex 8.10 release. As a result, the kernel team would appreciate it if you could please test this newer 2.6.27 Ubuntu kernel. There are one of two ways you should be able to test:

1) If you are comfortable installing packages on your own, the linux-image-2.6.27-* package is currently available for you to install and test.

--or--

2) The upcoming Alpha5 for Intrepid Ibex 8.10 will contain this newer 2.6.27 Ubuntu kernel. Alpha5 is set to be released Thursday Sept 4. Please watch http://www.ubuntu.com/testing for Alpha5 to be announced. You should then be able to test via a LiveCD.

Please let us know immediately if this newer 2.6.27 kernel resolves the bug reported here or if the issue remains. More importantly, please open a new bug report for each new bug/regression introduced by the 2.6.27 kernel and tag the bug report with 'linux-2.6.27'. Also, please specifically note if the issue does or does not appear in the 2.6.26 kernel. Thanks again, we really appreicate your help and feedback.

Revision history for this message
pato101 (pato101) wrote : Re: [Bug 19982] Re: Low performance on usb-media and DVD-RAM in sync mode

Where can I get these linux-image-2.6.27-* packages?

as for 2), I commented that, strangely, at the liveCD this issue did
not happen (couldn't test the DVD-RAM issue but the usb-stick one
yes).

2008/8/28 Leann Ogasawara <email address hidden>:
> The Ubuntu Kernel Team is planning to move to the 2.6.27 kernel for the
> upcoming Intrepid Ibex 8.10 release. As a result, the kernel team would
> appreciate it if you could please test this newer 2.6.27 Ubuntu kernel.
> There are one of two ways you should be able to test:
>
> 1) If you are comfortable installing packages on your own, the linux-
> image-2.6.27-* package is currently available for you to install and
> test.
>
> --or--
>
> 2) The upcoming Alpha5 for Intrepid Ibex 8.10 will contain this newer
> 2.6.27 Ubuntu kernel. Alpha5 is set to be released Thursday Sept 4.
> Please watch http://www.ubuntu.com/testing for Alpha5 to be announced.
> You should then be able to test via a LiveCD.
>
> Please let us know immediately if this newer 2.6.27 kernel resolves the
> bug reported here or if the issue remains. More importantly, please
> open a new bug report for each new bug/regression introduced by the
> 2.6.27 kernel and tag the bug report with 'linux-2.6.27'. Also, please
> specifically note if the issue does or does not appear in the 2.6.26
> kernel. Thanks again, we really appreicate your help and feedback.
>
> ** Tags added: cft-2.6.27
>
> --
> Low performance on usb-media and DVD-RAM in sync mode
> https://bugs.launchpad.net/bugs/19982
> You received this bug notification because you are a direct subscriber
> of the bug.
>

Revision history for this message
pato101 (pato101) wrote :

I confirm the bug is still present at Gutsy with

Linux 2.6.24-21-generic #1 SMP Mon Aug 25 16:57:51 UTC 2008 x86_64 GNU/Linux

from proposed?/backports? packages

2008/8/29 pato <email address hidden>:
> Where can I get these linux-image-2.6.27-* packages?
>
> as for 2), I commented that, strangely, at the liveCD this issue did
> not happen (couldn't test the DVD-RAM issue but the usb-stick one
> yes).
>
>
> 2008/8/28 Leann Ogasawara <email address hidden>:
>> The Ubuntu Kernel Team is planning to move to the 2.6.27 kernel for the
>> upcoming Intrepid Ibex 8.10 release. As a result, the kernel team would
>> appreciate it if you could please test this newer 2.6.27 Ubuntu kernel.
>> There are one of two ways you should be able to test:
>>
>> 1) If you are comfortable installing packages on your own, the linux-
>> image-2.6.27-* package is currently available for you to install and
>> test.
>>
>> --or--
>>
>> 2) The upcoming Alpha5 for Intrepid Ibex 8.10 will contain this newer
>> 2.6.27 Ubuntu kernel. Alpha5 is set to be released Thursday Sept 4.
>> Please watch http://www.ubuntu.com/testing for Alpha5 to be announced.
>> You should then be able to test via a LiveCD.
>>
>> Please let us know immediately if this newer 2.6.27 kernel resolves the
>> bug reported here or if the issue remains. More importantly, please
>> open a new bug report for each new bug/regression introduced by the
>> 2.6.27 kernel and tag the bug report with 'linux-2.6.27'. Also, please
>> specifically note if the issue does or does not appear in the 2.6.26
>> kernel. Thanks again, we really appreicate your help and feedback.
>>
>> ** Tags added: cft-2.6.27
>>
>> --
>> Low performance on usb-media and DVD-RAM in sync mode
>> https://bugs.launchpad.net/bugs/19982
>> You received this bug notification because you are a direct subscriber
>> of the bug.
>>
>

Revision history for this message
Gordon Mckeown (thefluffyone) wrote :

@pato101: I found some at http://kernel.ubuntu.com/pub/next/2.6.27-rc3/hardy/ -- can't guarantee these are exactly what's needed though.

Revision history for this message
Leann Ogasawara (leannogasawara) wrote :

Hi pato101,

If testing via a LiveCD won't be sufficient, you can enable the Intrepid repository to grab the newer kernel.

Revision history for this message
Gordon Mckeown (thefluffyone) wrote :

I tested again using Kubuntu Hardy updated to the 2.6.27-rc3 kernel mentioned above.

A 300GB Caviar SE PATA drive attached using a USB-IDE bridge cable gave the following transfer speeds when copying files using Dolphin:

Async: 20MB/s
Sync: 2.4MB/s

There was a lot more head activity in sync mode (understandably, I think). I can't directly compare this result to my last test back in June 2007 since I'm using a different drive, however there's still a massive performance between sync and async.

Changed in linux:
status: Incomplete → New
Changed in linux:
assignee: nobody → ubuntu-kernel-team
importance: Undecided → Medium
status: New → Triaged
Revision history for this message
Launchpad Janitor (janitor) wrote : Kernel team bugs

Per a decision made by the Ubuntu Kernel Team, bugs will longer be assigned to the ubuntu-kernel-team in Launchpad as part of the bug triage process. The ubuntu-kernel-team is being unassigned from this bug report. Refer to https://wiki.ubuntu.com/KernelTeamBugPolicies for more information. Thanks.

Revision history for this message
kernel-janitor (kernel-janitor) wrote :

This bug report was marked as Triaged a while ago but has not had any updated comments for quite some time. Please let us know if this issue remains in the current Ubuntu release, http://www.ubuntu.com/getubuntu/download . If the issue remains, click on the current status under the Status column and change the status back to "New". Thanks.

[This is an automated message. Apologies if it has reached you inappropriately; please just reply to this message indicating so.]

tags: added: kj-triage
Changed in linux (Ubuntu):
status: Triaged → Incomplete
Revision history for this message
Schugy (ich-schugy) wrote :

The slow usb-storage performance is still valid in Kubuntu 9.10 (linux-image-2.6.31-14-generic).
It was important for me to format my SDHC card (16 GB Class 10!) with a non journaling filesystem.
I've tried

I've tried several card readers which have slighty different performance but they have in general the same problem.

ID 05ca:1880 Ricoh Co., Ltd
ID 05e3:070e Genesys Logic, Inc. X-PRO CR20xA USB 2.0 Internal Card Reader
ID 05e3:0715 Genesys Logic, Inc. USB 2.0 microSD Reader (labeled as Hama 6 in 1)
ID 05e3:0715 Genesys Logic, Inc. USB 2.0 microSD Reader (labeled as Sandisk Micromate)

I have tried adding mount options like async, noatime, nodiratime or all at once.

I guess the module needs to "deparallize" or "sequentialize" i/o transactions.

Problems arise when processes like updatedb, firefox cache writes, ldconfig, update mandbare running. These processes run at 10 - 110 kb/sec instead of 10 mb. Booting seems to be almost read only and speed is fast.

But when I tried to write a big iso image onto the SDHC card it was running with 400 - 800 kb/sec. It became amazingly fast when I entered sync. Then I had between 6 and 14 mb/sec for the rest of the iso image.

I hope someone can write a scheduler for this stupid devices which don't have a high performance i/o-controller.

This would really improve my experience with beta distributions on my laptop which doesn't have a 2nd harddrive.

Changed in linux (Ubuntu):
status: Incomplete → Invalid
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.