Resource locks

hi,
I have a T-sql as part of ETL. The code selects all data
from table x (with 3 million rows) and populates table y
(which already has 1.5 million rows). There are no where
conditions no calculations - simple insert into table x
(col1,col2,...coln) select col1,col2...coln from table y.
When I run sp_lock I see a lot locks of type EXT and mode
X. Can I used hint TABLOCKX.
I noticed that if I used hint TABLOCKX for the table
inserted into then number of locks reduces drastically. If
I use hint TABLOCKX in the table selected from there is no
impact.
Anybody encountered similar issues? Any input will be
useful..
Thx,
DeepaThat means you are getting extent locks which are OK for this type operation
but if you don't have other users accessing it would be best to use a table
level lock but don't need to use TABLOCKX. Instead try TABLOCK but if you
have users in the table you are selecting from they may prevent this from
happeing.
--
Andrew J. Kelly SQL MVP
"Deepa" <anonymous@.discussions.microsoft.com> wrote in message
news:01cc01c3b51f$f2276eb0$a301280a@.phx.gbl...
> hi,
> I have a T-sql as part of ETL. The code selects all data
> from table x (with 3 million rows) and populates table y
> (which already has 1.5 million rows). There are no where
> conditions no calculations - simple insert into table x
> (col1,col2,...coln) select col1,col2...coln from table y.
> When I run sp_lock I see a lot locks of type EXT and mode
> X. Can I used hint TABLOCKX.
> I noticed that if I used hint TABLOCKX for the table
> inserted into then number of locks reduces drastically. If
> I use hint TABLOCKX in the table selected from there is no
> impact.
> Anybody encountered similar issues? Any input will be
> useful..
> Thx,
> Deepa
>|||Thank you Andrew. I am new to sqlserver and your help is
appreciated.
Now, this table has six indexes and dropped them before
the insert and recreate them after.
What I notice is for a brief moment the number of locks
jumps to 30,000+ before reducing to around 200 incase of
tablock and 4000 when used without tablock. Why is there a
spike initially.
Also I read that each lock resource uses 96k. My box has
2GB memory which means I can have upto (2*1024*1024*1024)/
(96*1024) which comes to about 21800 then how can the lock
resource grow to 30,000?
i encounter resource lock issues only at the time that the
o.s backup is happening. My systems people are not able to
advice me. Would you know if an o.s. backup would be heavy
on memory. I'm not a windows person either...
Many thanks,
Deepa|||Its 96 bytes Deepa and not 96KB :-)
"Deepa" <anonymous@.discussions.microsoft.com> wrote in message
news:002b01c3b5dd$f18aef40$a501280a@.phx.gbl...
> Thank you Andrew. I am new to sqlserver and your help is
> appreciated.
> Now, this table has six indexes and dropped them before
> the insert and recreate them after.
> What I notice is for a brief moment the number of locks
> jumps to 30,000+ before reducing to around 200 incase of
> tablock and 4000 when used without tablock. Why is there a
> spike initially.
> Also I read that each lock resource uses 96k. My box has
> 2GB memory which means I can have upto (2*1024*1024*1024)/
> (96*1024) which comes to about 21800 then how can the lock
> resource grow to 30,000?
> i encounter resource lock issues only at the time that the
> o.s backup is happening. My systems people are not able to
> advice me. Would you know if an o.s. backup would be heavy
> on memory. I'm not a windows person either...
> Many thanks,
> Deepa
>
>|||Yes as Hassan pointsout it is 96 bytes and not KB so you ae not as short on
memory as you think. The reason you see it spike initialy is that sql sever
starts out with row or page locks and will escalate to a table lock if it
can after a while. You say it is an OS backup, are you sure they aren't
using a sql plug-in to do sql backups as well? If you are doing a sql
backup it will take locks when it reads the data. If it is strictly an OS
backup they should eliminate the sql erver files from the backup as they are
useless from a sql server point of view and only can cause issues when
accessing the sql files.
--
Andrew J. Kelly SQL MVP
"Deepa" <anonymous@.discussions.microsoft.com> wrote in message
news:002b01c3b5dd$f18aef40$a501280a@.phx.gbl...
> Thank you Andrew. I am new to sqlserver and your help is
> appreciated.
> Now, this table has six indexes and dropped them before
> the insert and recreate them after.
> What I notice is for a brief moment the number of locks
> jumps to 30,000+ before reducing to around 200 incase of
> tablock and 4000 when used without tablock. Why is there a
> spike initially.
> Also I read that each lock resource uses 96k. My box has
> 2GB memory which means I can have upto (2*1024*1024*1024)/
> (96*1024) which comes to about 21800 then how can the lock
> resource grow to 30,000?
> i encounter resource lock issues only at the time that the
> o.s backup is happening. My systems people are not able to
> advice me. Would you know if an o.s. backup would be heavy
> on memory. I'm not a windows person either...
> Many thanks,
> Deepa
>
>|||Thank you both. Yes it is 96 bytes and not 96kb.. my bad!
I have been doing tests with drop index / populate /
recreate index and testing is in progress but this is
likely to resolve the error.
Its a standard o.s backup. The files being backed up are
the backup files created by my maintenance plans - no the
datafiles.
many thanks for the response.
>--Original Message--
>Yes as Hassan pointsout it is 96 bytes and not KB so you
ae not as short on
>memory as you think. The reason you see it spike
initialy is that sql sever
>starts out with row or page locks and will escalate to a
table lock if it
>can after a while. You say it is an OS backup, are you
sure they aren't
>using a sql plug-in to do sql backups as well? If you
are doing a sql
>backup it will take locks when it reads the data. If it
is strictly an OS
>backup they should eliminate the sql erver files from the
backup as they are
>useless from a sql server point of view and only can
cause issues when
>accessing the sql files.
>--
>Andrew J. Kelly SQL MVP
>
>"Deepa" <anonymous@.discussions.microsoft.com> wrote in
message
>news:002b01c3b5dd$f18aef40$a501280a@.phx.gbl...
>> Thank you Andrew. I am new to sqlserver and your help is
>> appreciated.
>> Now, this table has six indexes and dropped them before
>> the insert and recreate them after.
>> What I notice is for a brief moment the number of locks
>> jumps to 30,000+ before reducing to around 200 incase of
>> tablock and 4000 when used without tablock. Why is
there a
>> spike initially.
>> Also I read that each lock resource uses 96k. My box has
>> 2GB memory which means I can have upto
(2*1024*1024*1024)/
>> (96*1024) which comes to about 21800 then how can the
lock
>> resource grow to 30,000?
>> i encounter resource lock issues only at the time that
the
>> o.s backup is happening. My systems people are not able
to
>> advice me. Would you know if an o.s. backup would be
heavy
>> on memory. I'm not a windows person either...
>> Many thanks,
>> Deepa
>>
>
>.
>

Resource Limits ?

In Oracle & Sybase you can set login resource limits to restrict, time of day, number of i-o's, cpu use etc for any login, at the login method. All actions done from a resource limit constrained login inherit the constraints.

Is there any way to implment this with any MS SQL Server version from 200-2005/SP2?

Before anyone answers, please do not suggest set rowcount etc.
thanks.I don't see this kind of feature available yet in any of the versions of SQL Server, however workaround are the only possible methods like the one you have suggested.sql

resource limit was reached

Error:
The OLE DB provider "SQLNCLI" for linked server reported an error. Execution
terminated by the provider because a resource limit was reached.
Situation:
Large query is pulled from one server to another (data warehouse).
Suspect:
DTC on the Windows server 2003 where the source server is located (SQL 2k)
does not have DTC enabled:
Looking for:
A work-around as we are in a situation where the server cannot yet be
rebooted.
Question:
Is there a way to configure named pipes to effectively handle the resource
limitation on linked servers?
--
Regards,
JamieAs a workaround, can you try using TCP\IP provider for linking?
--
Kevin G. Boles
Indicium Resources, Inc.
SQL Server MVP
kgboles a earthlink dt net
"thejamie" <thejamie@.discussions.microsoft.com> wrote in message
news:A4DB5E7F-56F3-46AE-9858-0F67546B82ED@.microsoft.com...
> Error:
> The OLE DB provider "SQLNCLI" for linked server reported an error.
> Execution
> terminated by the provider because a resource limit was reached.
> Situation:
> Large query is pulled from one server to another (data warehouse).
> Suspect:
> DTC on the Windows server 2003 where the source server is located (SQL 2k)
> does not have DTC enabled:
> Looking for:
> A work-around as we are in a situation where the server cannot yet be
> rebooted.
> Question:
> Is there a way to configure named pipes to effectively handle the resource
> limitation on linked servers?
> --
> Regards,
> Jamie|||I think I may be misunderstanding. I'm unaware of TCP/IP configurations
other than what I already use. Which it may be that you misunderstood my
question, thinking I am using named pipes (not so - have named pipes turned
off).
Thought is - Will named pipes offer a temporary solution to overcome the
resource limitation imposed by the WinSvr2003 R2 without the DTC turned on?
Dunno enough to guess whether there is something in it that will work.
BACKGROUND:
I currently use TCP/IP and aliasing and I suspect the linked servers use
that already to communicate.
When we access a view on a SQL 2000 server using SQL 2005 on another server
(both are WinSvr 2003 R2 64 bit).
There is a problem with the DTC (I did not mention this one - I am waiting
for a server reboot to fix it - http://support.microsoft.com/kb/329332 or
http://support.microsoft.com/?kbid=873160 The operation could not be
performed because the OLE DB provider 'SQLOLEDB' was unable to begin a
distributed transaction. OLE DB error trace [OLE/DB Provider 'SQLOLEDB'
ITransactionJoin::JoinTransaction returned 0x8004d00a]. [OLE/DB provider
returned message: New transaction cannot enlist in the specified transaction
coordinator. ]).
The one housing SQL 2000 does not yet have DTC setup for network access.
The situation does not allow us to reboot (short of an emergency of which
this is not - instead each view must be re-written to get around the resource
issue - usually just create a sub-query in place of the view).
Herein lies the problem: After rewriting a bunch of failing views we
discovered that if a view is altered and consumes more resources, the
borderline ones we use for data-warehousing archive information at night can
fail for no other reason than that - view times out - resource limit
reached... blah blah). I hate it when a job fails.
Regards,
Jamie
"TheSQLGuru" wrote:
> As a workaround, can you try using TCP\IP provider for linking?
> --
> Kevin G. Boles
> Indicium Resources, Inc.
> SQL Server MVP
> kgboles a earthlink dt net
>
> "thejamie" <thejamie@.discussions.microsoft.com> wrote in message
> news:A4DB5E7F-56F3-46AE-9858-0F67546B82ED@.microsoft.com...
> > Error:
> > The OLE DB provider "SQLNCLI" for linked server reported an error.
> > Execution
> > terminated by the provider because a resource limit was reached.
> > Situation:
> > Large query is pulled from one server to another (data warehouse).
> > Suspect:
> > DTC on the Windows server 2003 where the source server is located (SQL 2k)
> > does not have DTC enabled:
> > Looking for:
> > A work-around as we are in a situation where the server cannot yet be
> > rebooted.
> >
> > Question:
> > Is there a way to configure named pipes to effectively handle the resource
> > limitation on linked servers?
> > --
> > Regards,
> > Jamie
>
>|||Misread first post. My bad. I don't know that there is a way to cut DTC
out of the loop, regardless of what connection mechanism you are using. I
don't have any other suggestions for possible workarounds for your issue
either. :(
--
Kevin G. Boles
Indicium Resources, Inc.
SQL Server MVP
kgboles a earthlink dt net
"thejamie" <thejamie@.discussions.microsoft.com> wrote in message
news:AF702DC6-3A07-4D85-9713-8C5406099EED@.microsoft.com...
>I think I may be misunderstanding. I'm unaware of TCP/IP configurations
> other than what I already use. Which it may be that you misunderstood my
> question, thinking I am using named pipes (not so - have named pipes
> turned
> off).
> Thought is - Will named pipes offer a temporary solution to overcome the
> resource limitation imposed by the WinSvr2003 R2 without the DTC turned
> on?
> Dunno enough to guess whether there is something in it that will work.
> BACKGROUND:
> I currently use TCP/IP and aliasing and I suspect the linked servers use
> that already to communicate.
> When we access a view on a SQL 2000 server using SQL 2005 on another
> server
> (both are WinSvr 2003 R2 64 bit).
> There is a problem with the DTC (I did not mention this one - I am waiting
> for a server reboot to fix it - http://support.microsoft.com/kb/329332 or
> http://support.microsoft.com/?kbid=873160 The operation could not be
> performed because the OLE DB provider 'SQLOLEDB' was unable to begin a
> distributed transaction. OLE DB error trace [OLE/DB Provider 'SQLOLEDB'
> ITransactionJoin::JoinTransaction returned 0x8004d00a]. [OLE/DB provider
> returned message: New transaction cannot enlist in the specified
> transaction
> coordinator. ]).
> The one housing SQL 2000 does not yet have DTC setup for network access.
> The situation does not allow us to reboot (short of an emergency of which
> this is not - instead each view must be re-written to get around the
> resource
> issue - usually just create a sub-query in place of the view).
> Herein lies the problem: After rewriting a bunch of failing views we
> discovered that if a view is altered and consumes more resources, the
> borderline ones we use for data-warehousing archive information at night
> can
> fail for no other reason than that - view times out - resource limit
> reached... blah blah). I hate it when a job fails.
>
> --
> Regards,
> Jamie
>
> "TheSQLGuru" wrote:
>> As a workaround, can you try using TCP\IP provider for linking?
>> --
>> Kevin G. Boles
>> Indicium Resources, Inc.
>> SQL Server MVP
>> kgboles a earthlink dt net
>>
>> "thejamie" <thejamie@.discussions.microsoft.com> wrote in message
>> news:A4DB5E7F-56F3-46AE-9858-0F67546B82ED@.microsoft.com...
>> > Error:
>> > The OLE DB provider "SQLNCLI" for linked server reported an error.
>> > Execution
>> > terminated by the provider because a resource limit was reached.
>> > Situation:
>> > Large query is pulled from one server to another (data warehouse).
>> > Suspect:
>> > DTC on the Windows server 2003 where the source server is located (SQL
>> > 2k)
>> > does not have DTC enabled:
>> > Looking for:
>> > A work-around as we are in a situation where the server cannot yet be
>> > rebooted.
>> >
>> > Question:
>> > Is there a way to configure named pipes to effectively handle the
>> > resource
>> > limitation on linked servers?
>> > --
>> > Regards,
>> > Jamie
>>

resource limit was reached

Error:
The OLE DB provider "SQLNCLI" for linked server reported an error. Execution
terminated by the provider because a resource limit was reached.
Situation:
Large query is pulled from one server to another (data warehouse).
Suspect:
DTC on the Windows server 2003 where the source server is located (SQL 2k)
does not have DTC enabled:
Looking for:
A work-around as we are in a situation where the server cannot yet be
rebooted.
Question:
Is there a way to configure named pipes to effectively handle the resource
limitation on linked servers?
Regards,
Jamie
As a workaround, can you try using TCP\IP provider for linking?
Kevin G. Boles
Indicium Resources, Inc.
SQL Server MVP
kgboles a earthlink dt net
"thejamie" <thejamie@.discussions.microsoft.com> wrote in message
news:A4DB5E7F-56F3-46AE-9858-0F67546B82ED@.microsoft.com...
> Error:
> The OLE DB provider "SQLNCLI" for linked server reported an error.
> Execution
> terminated by the provider because a resource limit was reached.
> Situation:
> Large query is pulled from one server to another (data warehouse).
> Suspect:
> DTC on the Windows server 2003 where the source server is located (SQL 2k)
> does not have DTC enabled:
> Looking for:
> A work-around as we are in a situation where the server cannot yet be
> rebooted.
> Question:
> Is there a way to configure named pipes to effectively handle the resource
> limitation on linked servers?
> --
> Regards,
> Jamie
|||I think I may be misunderstanding. I'm unaware of TCP/IP configurations
other than what I already use. Which it may be that you misunderstood my
question, thinking I am using named pipes (not so - have named pipes turned
off).
Thought is - Will named pipes offer a temporary solution to overcome the
resource limitation imposed by the WinSvr2003 R2 without the DTC turned on?
Dunno enough to guess whether there is something in it that will work.
BACKGROUND:
I currently use TCP/IP and aliasing and I suspect the linked servers use
that already to communicate.
When we access a view on a SQL 2000 server using SQL 2005 on another server
(both are WinSvr 2003 R2 64 bit).
There is a problem with the DTC (I did not mention this one - I am waiting
for a server reboot to fix it - http://support.microsoft.com/kb/329332 or
http://support.microsoft.com/?kbid=873160 The operation could not be
performed because the OLE DB provider 'SQLOLEDB' was unable to begin a
distributed transaction. OLE DB error trace [OLE/DB Provider 'SQLOLEDB'
ITransactionJoin::JoinTransaction returned 0x8004d00a]. [OLE/DB provider
returned message: New transaction cannot enlist in the specified transaction
coordinator. ]).
The one housing SQL 2000 does not yet have DTC setup for network access.
The situation does not allow us to reboot (short of an emergency of which
this is not - instead each view must be re-written to get around the resource
issue - usually just create a sub-query in place of the view).
Herein lies the problem: After rewriting a bunch of failing views we
discovered that if a view is altered and consumes more resources, the
borderline ones we use for data-warehousing archive information at night can
fail for no other reason than that - view times out - resource limit
reached... blah blah). I hate it when a job fails.
Regards,
Jamie
"TheSQLGuru" wrote:

> As a workaround, can you try using TCP\IP provider for linking?
> --
> Kevin G. Boles
> Indicium Resources, Inc.
> SQL Server MVP
> kgboles a earthlink dt net
>
> "thejamie" <thejamie@.discussions.microsoft.com> wrote in message
> news:A4DB5E7F-56F3-46AE-9858-0F67546B82ED@.microsoft.com...
>
>
|||Misread first post. My bad. I don't know that there is a way to cut DTC
out of the loop, regardless of what connection mechanism you are using. I
don't have any other suggestions for possible workarounds for your issue
either.
Kevin G. Boles
Indicium Resources, Inc.
SQL Server MVP
kgboles a earthlink dt net
"thejamie" <thejamie@.discussions.microsoft.com> wrote in message
news:AF702DC6-3A07-4D85-9713-8C5406099EED@.microsoft.com...[vbcol=seagreen]
>I think I may be misunderstanding. I'm unaware of TCP/IP configurations
> other than what I already use. Which it may be that you misunderstood my
> question, thinking I am using named pipes (not so - have named pipes
> turned
> off).
> Thought is - Will named pipes offer a temporary solution to overcome the
> resource limitation imposed by the WinSvr2003 R2 without the DTC turned
> on?
> Dunno enough to guess whether there is something in it that will work.
> BACKGROUND:
> I currently use TCP/IP and aliasing and I suspect the linked servers use
> that already to communicate.
> When we access a view on a SQL 2000 server using SQL 2005 on another
> server
> (both are WinSvr 2003 R2 64 bit).
> There is a problem with the DTC (I did not mention this one - I am waiting
> for a server reboot to fix it - http://support.microsoft.com/kb/329332 or
> http://support.microsoft.com/?kbid=873160 The operation could not be
> performed because the OLE DB provider 'SQLOLEDB' was unable to begin a
> distributed transaction. OLE DB error trace [OLE/DB Provider 'SQLOLEDB'
> ITransactionJoin::JoinTransaction returned 0x8004d00a]. [OLE/DB provider
> returned message: New transaction cannot enlist in the specified
> transaction
> coordinator. ]).
> The one housing SQL 2000 does not yet have DTC setup for network access.
> The situation does not allow us to reboot (short of an emergency of which
> this is not - instead each view must be re-written to get around the
> resource
> issue - usually just create a sub-query in place of the view).
> Herein lies the problem: After rewriting a bunch of failing views we
> discovered that if a view is altered and consumes more resources, the
> borderline ones we use for data-warehousing archive information at night
> can
> fail for no other reason than that - view times out - resource limit
> reached... blah blah). I hate it when a job fails.
>
> --
> Regards,
> Jamie
>
> "TheSQLGuru" wrote:

Resource Isolation of Multiple SQL Instances

Hello,
I would like to know (without experimentation) if by using two instances of
SQL Server I can isolate a first database which requires high repeatability
in query update times from another much larger second database which may
consume both more CPU and memory resources.
Assume: (let me know also if this cannot be done)
(a) that the server has dual processors each SQL instance is assigned to its
own CPU.
(b) that I have allocated separate memory to each instance.
(I have read that the overall performance for a single instance is better
than multiple instances on the same server. This fact may be irrelavent if
I can live with a degraded overall performance but guarantee that the
instances are adequately isolated and queries remain repeatable on the first
database.)
Thanks in advance,
Ted HannahTed,
I think you've got the hang of it.
One thing to consider is that the DBMS relies on the OS. So, when the DBMS
has a lot to do, then the OS will place a heavier load as well. And, in your
scenario, you only have one OS. This means that your "OLTP" database will
still be somewhat affected by your "reporting" database (I'm guessing where
regarding OLTP and reporting). I don't think I've ever seen any tests on how
much this "bleeding" one can expect, but I'd thought I at least mention it.
My guess is that it won't be that much - main resource usage would be inside
each SQL Server.
--
Tibor Karaszi
"Ted Hannah" <tedhannah@.hotmail.com> wrote in message
news:eGVF6eyoDHA.2404@.TK2MSFTNGP12.phx.gbl...
> Hello,
> I would like to know (without experimentation) if by using two instances
of
> SQL Server I can isolate a first database which requires high
repeatability
> in query update times from another much larger second database which may
> consume both more CPU and memory resources.
> Assume: (let me know also if this cannot be done)
> (a) that the server has dual processors each SQL instance is assigned to
its
> own CPU.
> (b) that I have allocated separate memory to each instance.
> (I have read that the overall performance for a single instance is better
> than multiple instances on the same server. This fact may be irrelavent
if
> I can live with a degraded overall performance but guarantee that the
> instances are adequately isolated and queries remain repeatable on the
first
> database.)
>
> Thanks in advance,
> Ted Hannah
>|||Thanks - I would be very interested if anyone has measured this with some
real test. It would save some time. Speculation, unfortunately, won't be
good enough.
TED
"Tibor Karaszi" <tibor.please_reply_to_public_forum.karaszi@.cornerstone.se>
wrote in message news:_22qb.32395$mU6.93437@.newsb.telia.net...
> Ted,
> I think you've got the hang of it.
> One thing to consider is that the DBMS relies on the OS. So, when the DBMS
> has a lot to do, then the OS will place a heavier load as well. And, in
your
> scenario, you only have one OS. This means that your "OLTP" database will
> still be somewhat affected by your "reporting" database (I'm guessing
where
> regarding OLTP and reporting). I don't think I've ever seen any tests on
how
> much this "bleeding" one can expect, but I'd thought I at least mention
it.
> My guess is that it won't be that much - main resource usage would be
inside
> each SQL Server.
> --
> Tibor Karaszi
>
> "Ted Hannah" <tedhannah@.hotmail.com> wrote in message
> news:eGVF6eyoDHA.2404@.TK2MSFTNGP12.phx.gbl...
> > Hello,
> >
> > I would like to know (without experimentation) if by using two instances
> of
> > SQL Server I can isolate a first database which requires high
> repeatability
> > in query update times from another much larger second database which may
> > consume both more CPU and memory resources.
> > Assume: (let me know also if this cannot be done)
> > (a) that the server has dual processors each SQL instance is assigned to
> its
> > own CPU.
> > (b) that I have allocated separate memory to each instance.
> >
> > (I have read that the overall performance for a single instance is
better
> > than multiple instances on the same server. This fact may be irrelavent
> if
> > I can live with a degraded overall performance but guarantee that the
> > instances are adequately isolated and queries remain repeatable on the
> first
> > database.)
> >
> >
> > Thanks in advance,
> > Ted Hannah
> >
> >
>|||IO contention will continue to be a problem unless the databases are on
different PCI Busses, Controllers,and disks.
There are not any internal OS bottlenecks that I have seen that would
prevent you from successfully isolating workloads like this, but in a large
hetergeneous environment, it makes sense to keep the DSS and OLAP stuff on
different machines than OLTP. If you're looking at one server for two apps
and can seperate the IO channels, memory, and CPU affinity, then I think you
won't have any issues.
Kevin Connell, MCDBA
----
The views expressed here are my own
and not of my employer.
----
"Ted Hannah" <tedhannah@.hotmail.com> wrote in message
news:u6IQbu#oDHA.2424@.TK2MSFTNGP10.phx.gbl...
> Thanks - I would be very interested if anyone has measured this with some
> real test. It would save some time. Speculation, unfortunately, won't be
> good enough.
> TED
> "Tibor Karaszi"
<tibor.please_reply_to_public_forum.karaszi@.cornerstone.se>
> wrote in message news:_22qb.32395$mU6.93437@.newsb.telia.net...
> > Ted,
> >
> > I think you've got the hang of it.
> > One thing to consider is that the DBMS relies on the OS. So, when the
DBMS
> > has a lot to do, then the OS will place a heavier load as well. And, in
> your
> > scenario, you only have one OS. This means that your "OLTP" database
will
> > still be somewhat affected by your "reporting" database (I'm guessing
> where
> > regarding OLTP and reporting). I don't think I've ever seen any tests on
> how
> > much this "bleeding" one can expect, but I'd thought I at least mention
> it.
> > My guess is that it won't be that much - main resource usage would be
> inside
> > each SQL Server.
> >
> > --
> > Tibor Karaszi
> >
> >
> > "Ted Hannah" <tedhannah@.hotmail.com> wrote in message
> > news:eGVF6eyoDHA.2404@.TK2MSFTNGP12.phx.gbl...
> > > Hello,
> > >
> > > I would like to know (without experimentation) if by using two
instances
> > of
> > > SQL Server I can isolate a first database which requires high
> > repeatability
> > > in query update times from another much larger second database which
may
> > > consume both more CPU and memory resources.
> > > Assume: (let me know also if this cannot be done)
> > > (a) that the server has dual processors each SQL instance is assigned
to
> > its
> > > own CPU.
> > > (b) that I have allocated separate memory to each instance.
> > >
> > > (I have read that the overall performance for a single instance is
> better
> > > than multiple instances on the same server. This fact may be
irrelavent
> > if
> > > I can live with a degraded overall performance but guarantee that the
> > > instances are adequately isolated and queries remain repeatable on the
> > first
> > > database.)
> > >
> > >
> > > Thanks in advance,
> > > Ted Hannah
> > >
> > >
> >
> >
>

Resource In SQL Reporting Services

Can someone help me

is SQL reporting services supports the Resource .

HI, Nezo:

Would you please elaborate more about your requirement? Do you want to use some resource files?

|||

Hi Nizar,

I think what you want is to use some Name/Value pairs, in which value can be accessed using name. I dont think there is anything like Resource files as we have in visual studio.

The only way I can think of to do something like this is by using custom assemblies(http://msdn2.microsoft.com/en-us/library/ms153561.aspx) , you should be able to access properties in assemblies.

|||

thanks alot

but i need drop down List which contain Language like (English , Arabic)

when someone choose language the report change the Title and Alignment to Left if English And to Right when arabic

if that possible in SQL reporting or not

please help me

|||

I dont see any direct functionality provided in SSRS to do this, what you can do is to pass the selected language(from list) to the report and set the TextAlign property of Title using Expressions.

Resource Governor(searching.....)

Hi,

Unable to find the resource governor.

If it's available in the June CTP can tell me where to locate it and if its not there whn can we possibly get it.

It is not part of the June CTP. It will be available in a future CTP.

Thanks,|||Thanks Marcel.

|||I am unable to find the Resource Governor in Jul-CTP as well. Can you tell when can i have a glimpse of this new feature.

sql