Average Disk Queue Length

Hey guys. I've a problem with Average Disk queue length staying ta 100%. the
server is a very good box and doesn't do much on it. It also has 6gb RAM with
2gb each for 2 instances and 2 gb for Windows. 2 Xeon procs at 2.4Ghz hyper
threaded.
It is on Raid 5 and all the files are located on 1 drive. I know this is the
worst practice but this is a DEV box and not much is happening on it. There
is Replication going from instance 1 on the machine to instance2 on the
machine. But, still this should not happen. Someone Please help...Disk Queue counters are not a percentage they are a number. So either it is
not the Avg Disk Queue counter or you are misinterpreting it. Have a look
at this:
http://www.microsoft.com/sql/techinfo/administration/2000/perftuning.asp
Performance WP's
http://www.swynk.com/friends/vandenberg/perfmonitor.asp Perfmon counters
http://www.sql-server-performance.com/sql_server_performance_audit.asp
Hardware Performance CheckList
http://www.sql-server-performance.com/best_sql_server_performance_tips.asp
SQL 2000 Performance tuning tips
http://www.support.microsoft.com/?id=q224587 Troubleshooting App
Performance
http://msdn.microsoft.com/library/default.asp?url=/library/en-us/adminsql/ad_perfmon_24u1.asp
Disk Monitoring
--
Andrew J. Kelly SQL MVP
"Tejas Parikh" <TejasParikh@.discussions.microsoft.com> wrote in message
news:1EED4616-0587-484B-AB7F-AFDF03065F79@.microsoft.com...
> Hey guys. I've a problem with Average Disk queue length staying ta 100%.
> the
> server is a very good box and doesn't do much on it. It also has 6gb RAM
> with
> 2gb each for 2 instances and 2 gb for Windows. 2 Xeon procs at 2.4Ghz
> hyper
> threaded.
> It is on Raid 5 and all the files are located on 1 drive. I know this is
> the
> worst practice but this is a DEV box and not much is happening on it.
> There
> is Replication going from instance 1 on the machine to instance2 on the
> machine. But, still this should not happen. Someone Please help...|||Disk queue length is not measured in percent. Are you sure that's what you
have? That said, the avg disk queue length should not be > 2x #of spindles.
With all of your files on the same drive, it's not surprising that you have
this problem. RAID5 compounds this, as well. Of course, replication adds
to the burden, depending on the volume of data coming across, and this will
be exacerbated by snapshot synchronization.
Can you purchase some disks and create a couple of RAID10 sets?
--
Tom
----
Thomas A. Moreau, BSc, PhD, MCSE, MCDBA
SQL Server MVP
Columnist, SQL Server Professional
Toronto, ON Canada tom@.cips.ca
www.pinpub.com
"Tejas Parikh" <TejasParikh@.discussions.microsoft.com> wrote in message
news:1EED4616-0587-484B-AB7F-AFDF03065F79@.microsoft.com...
> Hey guys. I've a problem with Average Disk queue length staying ta 100%.
> the
> server is a very good box and doesn't do much on it. It also has 6gb RAM
> with
> 2gb each for 2 instances and 2 gb for Windows. 2 Xeon procs at 2.4Ghz
> hyper
> threaded.
> It is on Raid 5 and all the files are located on 1 drive. I know this is
> the
> worst practice but this is a DEV box and not much is happening on it.
> There
> is Replication going from instance 1 on the machine to instance2 on the
> machine. But, still this should not happen. Someone Please help...|||I understand these problems and yes, I had written it wrong. It's a counter
an dis not measured in percent. The only reason i posted this question here
is because it should not be so high even with all the existing WORST
practices followed. But, I didn't have much of a choice since it's a dev box
and not much s going on with this server here which is why the Avg disk queue
length should not be high... Is there anything else i should do to find why
it's so high?
thank you in advance.
"Tom Moreau" wrote:
> Disk queue length is not measured in percent. Are you sure that's what you
> have? That said, the avg disk queue length should not be > 2x #of spindles.
> With all of your files on the same drive, it's not surprising that you have
> this problem. RAID5 compounds this, as well. Of course, replication adds
> to the burden, depending on the volume of data coming across, and this will
> be exacerbated by snapshot synchronization.
> Can you purchase some disks and create a couple of RAID10 sets?
> --
> Tom
> ----
> Thomas A. Moreau, BSc, PhD, MCSE, MCDBA
> SQL Server MVP
> Columnist, SQL Server Professional
> Toronto, ON Canada tom@.cips.ca
> www.pinpub.com
> "Tejas Parikh" <TejasParikh@.discussions.microsoft.com> wrote in message
> news:1EED4616-0587-484B-AB7F-AFDF03065F79@.microsoft.com...
> > Hey guys. I've a problem with Average Disk queue length staying ta 100%.
> > the
> > server is a very good box and doesn't do much on it. It also has 6gb RAM
> > with
> > 2gb each for 2 instances and 2 gb for Windows. 2 Xeon procs at 2.4Ghz
> > hyper
> > threaded.
> > It is on Raid 5 and all the files are located on 1 drive. I know this is
> > the
> > worst practice but this is a DEV box and not much is happening on it.
> > There
> > is Replication going from instance 1 on the machine to instance2 on the
> > machine. But, still this should not happen. Someone Please help...
>
>|||Maybe these will help:
http://www.microsoft.com/sql/techinfo/administration/2000/perftuning.asp
Performance WP's
http://www.swynk.com/friends/vandenberg/perfmonitor.asp Perfmon counters
http://www.sql-server-performance.com/sql_server_performance_audit.asp
Hardware Performance CheckList
http://www.sql-server-performance.com/best_sql_server_performance_tips.asp
SQL 2000 Performance tuning tips
http://www.support.microsoft.com/?id=q224587 Troubleshooting App
Performance
http://msdn.microsoft.com/library/default.asp?url=/library/en-us/adminsql/ad_perfmon_24u1.asp
Disk Monitoring
--
Andrew J. Kelly SQL MVP
"Tejas Parikh" <TejasParikh@.discussions.microsoft.com> wrote in message
news:DD9DE915-894E-4723-ADA7-BDFA25491DBE@.microsoft.com...
>I understand these problems and yes, I had written it wrong. It's a counter
> an dis not measured in percent. The only reason i posted this question
> here
> is because it should not be so high even with all the existing WORST
> practices followed. But, I didn't have much of a choice since it's a dev
> box
> and not much s going on with this server here which is why the Avg disk
> queue
> length should not be high... Is there anything else i should do to find
> why
> it's so high?
> thank you in advance.
> "Tom Moreau" wrote:
>> Disk queue length is not measured in percent. Are you sure that's what
>> you
>> have? That said, the avg disk queue length should not be > 2x #of
>> spindles.
>> With all of your files on the same drive, it's not surprising that you
>> have
>> this problem. RAID5 compounds this, as well. Of course, replication
>> adds
>> to the burden, depending on the volume of data coming across, and this
>> will
>> be exacerbated by snapshot synchronization.
>> Can you purchase some disks and create a couple of RAID10 sets?
>> --
>> Tom
>> ----
>> Thomas A. Moreau, BSc, PhD, MCSE, MCDBA
>> SQL Server MVP
>> Columnist, SQL Server Professional
>> Toronto, ON Canada tom@.cips.ca
>> www.pinpub.com
>> "Tejas Parikh" <TejasParikh@.discussions.microsoft.com> wrote in message
>> news:1EED4616-0587-484B-AB7F-AFDF03065F79@.microsoft.com...
>> > Hey guys. I've a problem with Average Disk queue length staying ta
>> > 100%.
>> > the
>> > server is a very good box and doesn't do much on it. It also has 6gb
>> > RAM
>> > with
>> > 2gb each for 2 instances and 2 gb for Windows. 2 Xeon procs at 2.4Ghz
>> > hyper
>> > threaded.
>> > It is on Raid 5 and all the files are located on 1 drive. I know this
>> > is
>> > the
>> > worst practice but this is a DEV box and not much is happening on it.
>> > There
>> > is Replication going from instance 1 on the machine to instance2 on the
>> > machine. But, still this should not happen. Someone Please help...
>>|||See Andrew's reply. Also, post the numbers you get for the read and write
queue lengths.
--
Tom
----
Thomas A. Moreau, BSc, PhD, MCSE, MCDBA
SQL Server MVP
Columnist, SQL Server Professional
Toronto, ON Canada tom@.cips.ca
www.pinpub.com
"Tejas Parikh" <TejasParikh@.discussions.microsoft.com> wrote in message
news:DD9DE915-894E-4723-ADA7-BDFA25491DBE@.microsoft.com...
>I understand these problems and yes, I had written it wrong. It's a counter
> an dis not measured in percent. The only reason i posted this question
> here
> is because it should not be so high even with all the existing WORST
> practices followed. But, I didn't have much of a choice since it's a dev
> box
> and not much s going on with this server here which is why the Avg disk
> queue
> length should not be high... Is there anything else i should do to find
> why
> it's so high?
> thank you in advance.
> "Tom Moreau" wrote:
>> Disk queue length is not measured in percent. Are you sure that's what
>> you
>> have? That said, the avg disk queue length should not be > 2x #of
>> spindles.
>> With all of your files on the same drive, it's not surprising that you
>> have
>> this problem. RAID5 compounds this, as well. Of course, replication
>> adds
>> to the burden, depending on the volume of data coming across, and this
>> will
>> be exacerbated by snapshot synchronization.
>> Can you purchase some disks and create a couple of RAID10 sets?
>> --
>> Tom
>> ----
>> Thomas A. Moreau, BSc, PhD, MCSE, MCDBA
>> SQL Server MVP
>> Columnist, SQL Server Professional
>> Toronto, ON Canada tom@.cips.ca
>> www.pinpub.com
>> "Tejas Parikh" <TejasParikh@.discussions.microsoft.com> wrote in message
>> news:1EED4616-0587-484B-AB7F-AFDF03065F79@.microsoft.com...
>> > Hey guys. I've a problem with Average Disk queue length staying ta
>> > 100%.
>> > the
>> > server is a very good box and doesn't do much on it. It also has 6gb
>> > RAM
>> > with
>> > 2gb each for 2 instances and 2 gb for Windows. 2 Xeon procs at 2.4Ghz
>> > hyper
>> > threaded.
>> > It is on Raid 5 and all the files are located on 1 drive. I know this
>> > is
>> > the
>> > worst practice but this is a DEV box and not much is happening on it.
>> > There
>> > is Replication going from instance 1 on the machine to instance2 on the
>> > machine. But, still this should not happen. Someone Please help...
>>