All of lore.kernel.org
 help / color / mirror / Atom feed
From: Daniel Pocock <daniel@pocock.com.au>
To: Martin Mokrejs <mmokrejs@fold.natur.cuni.cz>
Cc: linux-ide@vger.kernel.org
Subject: Re: very slow write to AHCI device
Date: Mon, 07 May 2012 20:11:51 +0200	[thread overview]
Message-ID: <4FA81067.80607@pocock.com.au> (raw)
In-Reply-To: <4FA81009.1030308@fold.natur.cuni.cz>

On 07/05/12 20:10, Martin Mokrejs wrote:
> Hi Daniel,
>
> Daniel Pocock wrote:
>   
>>
>> Hi,
>>
>> I'm testing a problem with the following combination:
>> - SB700/SB800 type controller in AHCI mode (in a HP Microserver N36L)
>> - Seagate Barracuda 1TB 7k2 drive with NCQ
>>     
> Is that a disk with 4k hardware sectors (Advanced Format)? Where does
> you partition start? At sector 63 or 2048? ;-)
>
>   
At 2048:

Disk /dev/sdb: 1000.2 GB, 1000204886016 bytes
255 heads, 63 sectors/track, 121601 cylinders, total 1953525168 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x484d5754

   Device Boot      Start         End      Blocks   Id  System
/dev/sdb1   *        2048      499711      248832   83  Linux
Partition 1 does not end on cylinder boundary.
/dev/sdb2          499712  1953523711   976512000   fd  Linux raid
autodetect





>> I've got a fresh 256MB partition, formatted ext4, mounted
>> barrier=1,data=ordered, write-cache enabled (hdparm -W 1 /dev/sdb) and
>> shared over NFS.
>>
>> When the NFS client writes,
>> - unpacking a source tarball, many small files, iostat reports speeds
>> under 500kBytes/sec
>> - dd conv=fsync, iostat reports about 50MB/sec
>>
>> If I set up a USB disk on the same box, with a partition formatted
>> exactly the same way, the iostat reports the write speed (unpacking the
>> same tarball) is over 5MBytes/sec - not so fast, but 10 times faster
>> than the AHCI device.
>>     
> So USB is at speed 10 times less, right?
>
>   

No - the USB is 10x faster, 5MBytes/sec, the SATA disk is giving me
barely 500kBytes/sec for the same write over NFS

> Martin
> --
> To unsubscribe from this list: send the line "unsubscribe linux-ide" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>   


  reply	other threads:[~2012-05-07 18:12 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-05-07 16:16 very slow write to AHCI device Daniel Pocock
2012-05-07 18:10 ` Martin Mokrejs
2012-05-07 18:11   ` Daniel Pocock [this message]
2012-05-07 18:27     ` Martin Mokrejs
2012-05-07 23:30       ` Daniel Pocock
2012-05-08  8:21         ` Martin Mokrejs

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=4FA81067.80607@pocock.com.au \
    --to=daniel@pocock.com.au \
    --cc=linux-ide@vger.kernel.org \
    --cc=mmokrejs@fold.natur.cuni.cz \
    /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.