Groups > comp.sys.ibm.pc.hardware.storage > #2275

Re: IDE or AHCI ?

From	Arno <me@privacy.net>
Newsgroups	comp.sys.ibm.pc.hardware.storage
Subject	Re: IDE or AHCI ?
Date	2012-01-05 08:47 +0000
Message-ID	<9ml6dmFajhU1@mid.individual.net> (permalink)
References	(2 earlier) <hPWdnT07Memqm5nSnZ2dnUVZ7oadnZ2d@lyse.net> <4f046249$1@news.bnb-lp.com> <4aGdnUAOPIbQ9pnSnZ2dnUVZ8tadnZ2d@lyse.net> <9mjn1tFocqU3@mid.individual.net> <WZudnapbZd6iUpnSnZ2dnUVZ7vydnZ2d@lyse.net>

Show all headers | View raw

David Brown <david.brown@removethis.hesbynett.no> wrote:
> On 04/01/12 20:19, Arno wrote:
>> David Brown<david@westcontrol.removethisbit.com>  wrote:
>>> On 04/01/2012 15:29, Yousuf Khan wrote:
>>>> On 04/01/2012 2:53 AM, David Brown wrote:
>>>>> NCQ won't make a significant difference in Linux, since it has always
>>>>> had good algorithms to order disk accesses to minimise head movement. It
>>>>> will sometimes make things worse, such as when the OS wants to enforce a
>>>>> particular order (for transactions to filesystem journals, for example).
>>>>> And NCQ doesn't help windows much either - after all, it only applies
>>>>> when you do more than one thing at a time.
>>>>
>>>> I find that it doesn't even help even when multitasking. I monitor the
>>>> disk subsection of the Resource Monitor regularly, and very often when
>>>> the disk is busy the Disk Queue Length is over 1.00 (meaning more than 1
>>>> process is actively waiting on the disk) and the Active Time is pegged
>>>> near 100%. Nothing that can be done about it till SSD's are more
>>>> affordable.
>>>>
>>>> Yousuf Khan
>>
>>> NCQ can only really help if you have multiple outstanding transactions,
>>> and the OS itself hasn't ordered them appropriately.  Since the OS
>>> (Windows or Linux) /does/ order transactions, NCQ will only help if the
>>> OS is doing a bad job.  The disk knows a bit more than the OS regarding
>>> disk ordering (since it knows the full 3D geometry, rather than just a
>>> linear LBA number), but on the other hand it knows nothing about which
>>> processes are waiting for disk access, or the priorities of said
>>> accesses, and it knows nothing about barrier writes.  I don't know how
>>> Windows handles write barriers, but on Linux they are important to
>>> ensure the integrity of critical disk accesses such as journalling -
>>> they ensure that everything that was supposed to be written earlier
>>> /has/ been written.  NCQ totally screws this up, and means that the OS's
>>> IO subsystem must ensure the disk queue is completely empty before
>>> sending the barrier write, and wait for it to finish completely before
>>> sending anything else.  If the disk handles transactions in the order
>>> they are given, then such writes can be buffered better.
>>
>> Well, yes. Write-barriers are getting more and more important
>> on Linux, with filesystems deferring more and using jourmalling
>> more agressively. Looks like NCQ is basically obsolete.
>>
>> If I remember correctly, it is a thing that was brought on in
>> SCSI disks a long time ago, when it still had merit. Not so
>> anymore, just one more TLA that can be trhown at customers
>> to make them think they are getting more for their money.
>>
>> Arno

> The feature introduced on SCSI was TCQ - Tagged Command Queuing.  It was 
> more flexible, and more useful - each command sent to the disk could be 
> tagged as "head of queue" (do this command with highest priority), 
> "ordered" (enforcing the order of the tagged commands) and "simple" (do 
> in whatever order the disk wants).  NCQ is pretty much TCQ "simple".  If 
> SATA supported the "ordered" mode of TCQ, it would be very useful.  I 
> don't know what happens when mixing different tag types in TCQ, but I 
> think if "order" had top priority for writes, "head of queue" had top 
> priority for reads, and "simple" had lowest priority for both, then 
> you'd have a system that would improve speed in almost all cases as well 
> as being easy to make write-barrier safe.

> I assume that SAS supports TCQ.

Ah, yes. TCQ it was indeed.

Arno

-- 
Arno Wagner, Dr. sc. techn., Dipl. Inform., CISSP -- Email: arno@wagner.name
GnuPG:  ID: 1E25338F  FP: 0C30 5782 9D93 F785 E79C  0296 797F 6B50 1E25 338F
----
Cuddly UI's are the manifestation of wishful thinking. -- Dylan Evans

Thread

IDE or AHCI ? Lynn McGuire <lmc@winsim.com> - 2012-01-03 12:30 -0600
  Re: IDE or AHCI ? Arno <me@privacy.net> - 2012-01-03 21:05 +0000
    Re: IDE or AHCI ? "arno fucker" <af@arnofucker.com> - 2012-01-04 12:41 +1100
  Re: IDE or AHCI ? "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-01-04 12:58 +1100
  Re: IDE or AHCI ? Yousuf Khan <bbbl67@spammenot.yahoo.com> - 2012-01-04 00:07 -0500
    Re: IDE or AHCI ? David Brown <david@westcontrol.removethisbit.com> - 2012-01-04 08:53 +0100
      Re: IDE or AHCI ? "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-01-04 20:32 +1100
        Re: IDE or AHCI ? David Brown <david@westcontrol.removethisbit.com> - 2012-01-04 14:44 +0100
          Re: IDE or AHCI ? "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-01-05 07:37 +1100
            Re: IDE or AHCI ? David Brown <david.brown@removethis.hesbynett.no> - 2012-01-04 22:56 +0100
              Re: IDE or AHCI ? "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-01-05 09:48 +1100
              Re: IDE or AHCI ? Ed Light <nobody@nobody.there> - 2012-01-04 21:40 -0800
              Re: IDE or AHCI ? Mike Tomlinson <mike@jasper.org.uk> - 2012-01-05 05:37 +0000
                Re: IDE or AHCI ? "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-01-05 17:37 +1100
      Re: IDE or AHCI ? Yousuf Khan <bbbl67@spammenot.yahoo.com> - 2012-01-04 09:29 -0500
        Re: IDE or AHCI ? David Brown <david@westcontrol.removethisbit.com> - 2012-01-04 16:04 +0100
          Re: IDE or AHCI ? Arno <me@privacy.net> - 2012-01-04 19:19 +0000
            Re: IDE or AHCI ? David Brown <david.brown@removethis.hesbynett.no> - 2012-01-04 23:14 +0100
              Re: IDE or AHCI ? Arno <me@privacy.net> - 2012-01-05 08:47 +0000
          Re: IDE or AHCI ? "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-01-05 07:51 +1100
        Re: IDE or AHCI ? Arno <me@privacy.net> - 2012-01-04 19:16 +0000
        Re: IDE or AHCI ? "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-01-05 07:41 +1100
          Re: IDE or AHCI ? Yousuf Khan <bbbl67@spammenot.yahoo.com> - 2012-01-05 14:44 -0500
            Re: IDE or AHCI ? "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-01-06 17:35 +1100
              Re: IDE or AHCI ? Krypsis <krypsis@optusnet.com.au> - 2012-01-07 00:05 +1100
                Re: IDE or AHCI ? Arno <me@privacy.net> - 2012-01-06 20:04 +0000
                Re: IDE or AHCI ? Mike Tomlinson <mike@jasper.org.uk> - 2012-01-06 21:13 +0000
                Re: IDE or AHCI ? cjt <cheljuba@prodigy.net> - 2012-01-06 20:42 -0600
                Re: IDE or AHCI ? David Brown <david.brown@removethis.hesbynett.no> - 2012-01-07 13:48 +0100
                Re: IDE or AHCI ? Arno <me@privacy.net> - 2012-01-07 14:09 +0000
                Re: IDE or AHCI ? Mike Tomlinson <mike@jasper.org.uk> - 2012-01-07 14:36 +0000
                Re: IDE or AHCI ? Krypsis <krypsis@optusnet.com.au> - 2012-01-08 02:11 +1100
                Re: IDE or AHCI ? Mike Tomlinson <mike@jasper.org.uk> - 2012-01-09 12:06 +0000
                Re: IDE or AHCI ? Arno <me@privacy.net> - 2012-01-09 16:27 +0000
                Re: IDE or AHCI ? Mike Tomlinson <mike@jasper.org.uk> - 2012-01-09 17:36 +0000
                Re: IDE or AHCI ? Mike Tomlinson <mike@jasper.org.uk> - 2012-01-09 17:41 +0000
                Re: IDE or AHCI ? Winniethepooh@100acrewoods.net (GMAN) - 2012-01-09 19:07 +0000
                Re: IDE or AHCI ? David Brown <david.brown@removethis.hesbynett.no> - 2012-01-09 20:12 +0100
                Re: IDE or AHCI ? Mike Tomlinson <mike@jasper.org.uk> - 2012-01-09 21:00 +0000
                Re: IDE or AHCI ? Yousuf Khan <bbbl67@spammenot.yahoo.com> - 2012-01-09 11:31 -0500
                Re: IDE or AHCI ? David Brown <david.brown@removethis.hesbynett.no> - 2012-01-07 17:10 +0100
                Re: IDE or AHCI ? Mike Tomlinson <mike@jasper.org.uk> - 2012-01-10 14:18 +0000
                Re: IDE or AHCI ? Yousuf Khan <bbbl67@spammenot.yahoo.com> - 2012-01-09 11:23 -0500
                Re: IDE or AHCI ? Arno <me@privacy.net> - 2012-01-09 16:30 +0000
                Re: IDE or AHCI ? David Brown <david.brown@removethis.hesbynett.no> - 2012-01-09 20:16 +0100
                Re: IDE or AHCI ? cjt <cheljuba@prodigy.net> - 2012-01-09 17:13 -0600
                Re: IDE or AHCI ? "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-01-07 08:11 +1100
                Re: IDE or AHCI ? Krypsis <krypsis@optusnet.com.au> - 2012-01-08 01:23 +1100
                Re: IDE or AHCI ? "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-01-08 06:43 +1100
                Re: IDE or AHCI ? JW <none@dev.null> - 2012-01-09 05:23 -0500
              Re: IDE or AHCI ? Yousuf Khan <bbbl67@spammenot.yahoo.com> - 2012-01-09 10:24 -0500
                Re: IDE or AHCI ? "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-01-10 09:34 +1100
                Re: IDE or AHCI ? Yousuf Khan <bbbl67@spammenot.yahoo.com> - 2012-01-09 20:04 -0500
                Re: IDE or AHCI ? "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-01-10 15:33 +1100
            Re: IDE or AHCI ? Arno <me@privacy.net> - 2012-01-06 09:48 +0000
              Re: IDE or AHCI ? Yousuf Khan <bbbl67@spammenot.yahoo.com> - 2012-01-09 11:35 -0500
                Re: IDE or AHCI ? David Brown <david.brown@removethis.hesbynett.no> - 2012-01-09 20:23 +0100
                Re: IDE or AHCI ? Arno <me@privacy.net> - 2012-01-10 11:31 +0000
                Re: IDE or AHCI ? Yousuf Khan <bbbl67@spammenot.yahoo.com> - 2012-01-20 06:54 -0500
      Re: IDE or AHCI ? Arno <me@privacy.net> - 2012-01-04 19:14 +0000

csiph-web