Hello,
I have two symmetrical databases on two SQL servers - one remote, the
other - local. The data in the remote server changes every minute. The data
in local SQL server is used for creating some reports and general testing.
Sometimes data from the local base can be irreversibly deleted.
I want to have some sort of procedure, that would do one-direction
synchronisation of the two databases. The data in the remote database should
overwrite the data in the local database, so that after the update the local
database is exact copy of the remote database. Of course, this kind of
result could be achieved by detaching the remote base and bringing over the
database files, or doing full DTS. However, detaching and transfering is not
a viable solution since the remote base is actively used and the size of
data files are considerable; And, doing full DTS would take years. There is
some need to do some comparison of record hashes etc., so that the size of
the data that have to be transfered is reduced to minimum.
Thanks,
Pavils
The first thing that comes to mind is BACKUP and RESTORE. After you restore
a database backup you will have an exact copy of the original (as of the
date and time the backup completed).
You mention that the size of the data files are "considerable." I am not
sure what that means, or if that is a limitation for you in your
environment.
You might be able to look into replication.
Another option: if you have datetime columns (such as DateUpdated or
DateInserted) stored on your tables you could create your own shipping
routines...simply find the "recent" data (whatever that means to you) and
insert it.
You could also use linked servers and INSERT INTO...SELECT...WHERE NOT
EXISTS... you will have to fill in the blanks, but they should be fairly
easy.
Keith
"Pavils Jurjans" <pavils@.mailbox.riga.lv> wrote in message
news:OsrHETCkEHA.3664@.TK2MSFTNGP12.phx.gbl...
> Hello,
> I have two symmetrical databases on two SQL servers - one remote, the
> other - local. The data in the remote server changes every minute. The
data
> in local SQL server is used for creating some reports and general testing.
> Sometimes data from the local base can be irreversibly deleted.
> I want to have some sort of procedure, that would do one-direction
> synchronisation of the two databases. The data in the remote database
should
> overwrite the data in the local database, so that after the update the
local
> database is exact copy of the remote database. Of course, this kind of
> result could be achieved by detaching the remote base and bringing over
the
> database files, or doing full DTS. However, detaching and transfering is
not
> a viable solution since the remote base is actively used and the size of
> data files are considerable; And, doing full DTS would take years. There
is
> some need to do some comparison of record hashes etc., so that the size of
> the data that have to be transfered is reduced to minimum.
> Thanks,
> Pavils
>
|||Hello Keith,
> You mention that the size of the data files are "considerable." I am not
> sure what that means, or if that is a limitation for you in your
> environment.
The idea is that I don't want to transfer every single bit of the whole
database.
> You might be able to look into replication.
Would replication allow to create a fully matching copy wit using as little
data xfer as possible?
> Another option: if you have datetime columns (such as DateUpdated or
> DateInserted) stored on your tables you could create your own shipping
> routines...simply find the "recent" data (whatever that means to you) and
> insert it.
That could be a good solution, but it requires reworking much of the current
application.
> You could also use linked servers and INSERT INTO...SELECT...WHERE NOT
> EXISTS... you will have to fill in the blanks, but they should be fairly
> easy.
Surely, I can do that, however I was hoping for some more automatic (or
wizard-style) solution.
-- Pavils.
|||"Pavils Jurjans" <pavils@.mailbox.riga.lv> wrote in message
news:uUQ1L7CkEHA.3968@.TK2MSFTNGP11.phx.gbl...[vbcol=seagreen]
> Hello Keith,
not
> The idea is that I don't want to transfer every single bit of the whole
> database.
understood
>
> Would replication allow to create a fully matching copy wit using as
little
> data xfer as possible?
That is the point of replication
[vbcol=seagreen]
and
> That could be a good solution, but it requires reworking much of the
current
> application.
Probably so (if you don't have a way to identify "new" rows.
[vbcol=seagreen]
fairly
> Surely, I can do that, however I was hoping for some more automatic (or
> wizard-style) solution.
This is not much different than my previous idea...it simply requires you to
type up some Transact-SQL. Hey, no one said being a dba/developer/IT
Professional was easy.
> -- Pavils.
>
Keith
Showing posts with label servers. Show all posts
Showing posts with label servers. Show all posts
Monday, March 26, 2012
Having an updated base every day
Hello,
I have two symmetrical databases on two SQL servers - one remote, the
other - local. The data in the remote server changes every minute. The data
in local SQL server is used for creating some reports and general testing.
Sometimes data from the local base can be irreversibly deleted.
I want to have some sort of procedure, that would do one-direction
synchronisation of the two databases. The data in the remote database should
overwrite the data in the local database, so that after the update the local
database is exact copy of the remote database. Of course, this kind of
result could be achieved by detaching the remote base and bringing over the
database files, or doing full DTS. However, detaching and transfering is not
a viable solution since the remote base is actively used and the size of
data files are considerable; And, doing full DTS would take years. There is
some need to do some comparison of record hashes etc., so that the size of
the data that have to be transfered is reduced to minimum.
Thanks,
PavilsThe first thing that comes to mind is BACKUP and RESTORE. After you restore
a database backup you will have an exact copy of the original (as of the
date and time the backup completed).
You mention that the size of the data files are "considerable." I am not
sure what that means, or if that is a limitation for you in your
environment.
You might be able to look into replication.
Another option: if you have datetime columns (such as DateUpdated or
DateInserted) stored on your tables you could create your own shipping
routines...simply find the "recent" data (whatever that means to you) and
insert it.
You could also use linked servers and INSERT INTO...SELECT...WHERE NOT
EXISTS... you will have to fill in the blanks, but they should be fairly
easy.
--
Keith
"Pavils Jurjans" <pavils@.mailbox.riga.lv> wrote in message
news:OsrHETCkEHA.3664@.TK2MSFTNGP12.phx.gbl...
> Hello,
> I have two symmetrical databases on two SQL servers - one remote, the
> other - local. The data in the remote server changes every minute. The
data
> in local SQL server is used for creating some reports and general testing.
> Sometimes data from the local base can be irreversibly deleted.
> I want to have some sort of procedure, that would do one-direction
> synchronisation of the two databases. The data in the remote database
should
> overwrite the data in the local database, so that after the update the
local
> database is exact copy of the remote database. Of course, this kind of
> result could be achieved by detaching the remote base and bringing over
the
> database files, or doing full DTS. However, detaching and transfering is
not
> a viable solution since the remote base is actively used and the size of
> data files are considerable; And, doing full DTS would take years. There
is
> some need to do some comparison of record hashes etc., so that the size of
> the data that have to be transfered is reduced to minimum.
> Thanks,
> Pavils
>|||Hello Keith,
> You mention that the size of the data files are "considerable." I am not
> sure what that means, or if that is a limitation for you in your
> environment.
The idea is that I don't want to transfer every single bit of the whole
database.
> You might be able to look into replication.
Would replication allow to create a fully matching copy wit using as little
data xfer as possible?
> Another option: if you have datetime columns (such as DateUpdated or
> DateInserted) stored on your tables you could create your own shipping
> routines...simply find the "recent" data (whatever that means to you) and
> insert it.
That could be a good solution, but it requires reworking much of the current
application.
> You could also use linked servers and INSERT INTO...SELECT...WHERE NOT
> EXISTS... you will have to fill in the blanks, but they should be fairly
> easy.
Surely, I can do that, however I was hoping for some more automatic (or
wizard-style) solution.
-- Pavils.|||"Pavils Jurjans" <pavils@.mailbox.riga.lv> wrote in message
news:uUQ1L7CkEHA.3968@.TK2MSFTNGP11.phx.gbl...
> Hello Keith,
> > You mention that the size of the data files are "considerable." I am
not
> > sure what that means, or if that is a limitation for you in your
> > environment.
> The idea is that I don't want to transfer every single bit of the whole
> database.
understood
> > You might be able to look into replication.
> Would replication allow to create a fully matching copy wit using as
little
> data xfer as possible?
That is the point of replication
> > Another option: if you have datetime columns (such as DateUpdated or
> > DateInserted) stored on your tables you could create your own shipping
> > routines...simply find the "recent" data (whatever that means to you)
and
> > insert it.
> That could be a good solution, but it requires reworking much of the
current
> application.
Probably so (if you don't have a way to identify "new" rows.
> > You could also use linked servers and INSERT INTO...SELECT...WHERE NOT
> > EXISTS... you will have to fill in the blanks, but they should be
fairly
> > easy.
> Surely, I can do that, however I was hoping for some more automatic (or
> wizard-style) solution.
This is not much different than my previous idea...it simply requires you to
type up some Transact-SQL. Hey, no one said being a dba/developer/IT
Professional was easy.
> -- Pavils.
>
--
Keith
I have two symmetrical databases on two SQL servers - one remote, the
other - local. The data in the remote server changes every minute. The data
in local SQL server is used for creating some reports and general testing.
Sometimes data from the local base can be irreversibly deleted.
I want to have some sort of procedure, that would do one-direction
synchronisation of the two databases. The data in the remote database should
overwrite the data in the local database, so that after the update the local
database is exact copy of the remote database. Of course, this kind of
result could be achieved by detaching the remote base and bringing over the
database files, or doing full DTS. However, detaching and transfering is not
a viable solution since the remote base is actively used and the size of
data files are considerable; And, doing full DTS would take years. There is
some need to do some comparison of record hashes etc., so that the size of
the data that have to be transfered is reduced to minimum.
Thanks,
PavilsThe first thing that comes to mind is BACKUP and RESTORE. After you restore
a database backup you will have an exact copy of the original (as of the
date and time the backup completed).
You mention that the size of the data files are "considerable." I am not
sure what that means, or if that is a limitation for you in your
environment.
You might be able to look into replication.
Another option: if you have datetime columns (such as DateUpdated or
DateInserted) stored on your tables you could create your own shipping
routines...simply find the "recent" data (whatever that means to you) and
insert it.
You could also use linked servers and INSERT INTO...SELECT...WHERE NOT
EXISTS... you will have to fill in the blanks, but they should be fairly
easy.
--
Keith
"Pavils Jurjans" <pavils@.mailbox.riga.lv> wrote in message
news:OsrHETCkEHA.3664@.TK2MSFTNGP12.phx.gbl...
> Hello,
> I have two symmetrical databases on two SQL servers - one remote, the
> other - local. The data in the remote server changes every minute. The
data
> in local SQL server is used for creating some reports and general testing.
> Sometimes data from the local base can be irreversibly deleted.
> I want to have some sort of procedure, that would do one-direction
> synchronisation of the two databases. The data in the remote database
should
> overwrite the data in the local database, so that after the update the
local
> database is exact copy of the remote database. Of course, this kind of
> result could be achieved by detaching the remote base and bringing over
the
> database files, or doing full DTS. However, detaching and transfering is
not
> a viable solution since the remote base is actively used and the size of
> data files are considerable; And, doing full DTS would take years. There
is
> some need to do some comparison of record hashes etc., so that the size of
> the data that have to be transfered is reduced to minimum.
> Thanks,
> Pavils
>|||Hello Keith,
> You mention that the size of the data files are "considerable." I am not
> sure what that means, or if that is a limitation for you in your
> environment.
The idea is that I don't want to transfer every single bit of the whole
database.
> You might be able to look into replication.
Would replication allow to create a fully matching copy wit using as little
data xfer as possible?
> Another option: if you have datetime columns (such as DateUpdated or
> DateInserted) stored on your tables you could create your own shipping
> routines...simply find the "recent" data (whatever that means to you) and
> insert it.
That could be a good solution, but it requires reworking much of the current
application.
> You could also use linked servers and INSERT INTO...SELECT...WHERE NOT
> EXISTS... you will have to fill in the blanks, but they should be fairly
> easy.
Surely, I can do that, however I was hoping for some more automatic (or
wizard-style) solution.
-- Pavils.|||"Pavils Jurjans" <pavils@.mailbox.riga.lv> wrote in message
news:uUQ1L7CkEHA.3968@.TK2MSFTNGP11.phx.gbl...
> Hello Keith,
> > You mention that the size of the data files are "considerable." I am
not
> > sure what that means, or if that is a limitation for you in your
> > environment.
> The idea is that I don't want to transfer every single bit of the whole
> database.
understood
> > You might be able to look into replication.
> Would replication allow to create a fully matching copy wit using as
little
> data xfer as possible?
That is the point of replication
> > Another option: if you have datetime columns (such as DateUpdated or
> > DateInserted) stored on your tables you could create your own shipping
> > routines...simply find the "recent" data (whatever that means to you)
and
> > insert it.
> That could be a good solution, but it requires reworking much of the
current
> application.
Probably so (if you don't have a way to identify "new" rows.
> > You could also use linked servers and INSERT INTO...SELECT...WHERE NOT
> > EXISTS... you will have to fill in the blanks, but they should be
fairly
> > easy.
> Surely, I can do that, however I was hoping for some more automatic (or
> wizard-style) solution.
This is not much different than my previous idea...it simply requires you to
type up some Transact-SQL. Hey, no one said being a dba/developer/IT
Professional was easy.
> -- Pavils.
>
--
Keith
Having an updated base every day
Hello,
I have two symmetrical databases on two SQL servers - one remote, the
other - local. The data in the remote server changes every minute. The data
in local SQL server is used for creating some reports and general testing.
Sometimes data from the local base can be irreversibly deleted.
I want to have some sort of procedure, that would do one-direction
synchronisation of the two databases. The data in the remote database should
overwrite the data in the local database, so that after the update the local
database is exact copy of the remote database. Of course, this kind of
result could be achieved by detaching the remote base and bringing over the
database files, or doing full DTS. However, detaching and transfering is not
a viable solution since the remote base is actively used and the size of
data files are considerable; And, doing full DTS would take years. There is
some need to do some comparison of record hashes etc., so that the size of
the data that have to be transfered is reduced to minimum.
Thanks,
PavilsThe first thing that comes to mind is BACKUP and RESTORE. After you restore
a database backup you will have an exact copy of the original (as of the
date and time the backup completed).
You mention that the size of the data files are "considerable." I am not
sure what that means, or if that is a limitation for you in your
environment.
You might be able to look into replication.
Another option: if you have datetime columns (such as DateUpdated or
DateInserted) stored on your tables you could create your own shipping
routines...simply find the "recent" data (whatever that means to you) and
insert it.
You could also use linked servers and INSERT INTO...SELECT...WHERE NOT
EXISTS... you will have to fill in the blanks, but they should be fairly
easy.
Keith
"Pavils Jurjans" <pavils@.mailbox.riga.lv> wrote in message
news:OsrHETCkEHA.3664@.TK2MSFTNGP12.phx.gbl...
> Hello,
> I have two symmetrical databases on two SQL servers - one remote, the
> other - local. The data in the remote server changes every minute. The
data
> in local SQL server is used for creating some reports and general testing.
> Sometimes data from the local base can be irreversibly deleted.
> I want to have some sort of procedure, that would do one-direction
> synchronisation of the two databases. The data in the remote database
should
> overwrite the data in the local database, so that after the update the
local
> database is exact copy of the remote database. Of course, this kind of
> result could be achieved by detaching the remote base and bringing over
the
> database files, or doing full DTS. However, detaching and transfering is
not
> a viable solution since the remote base is actively used and the size of
> data files are considerable; And, doing full DTS would take years. There
is
> some need to do some comparison of record hashes etc., so that the size of
> the data that have to be transfered is reduced to minimum.
> Thanks,
> Pavils
>|||Hello Keith,
> You mention that the size of the data files are "considerable." I am not
> sure what that means, or if that is a limitation for you in your
> environment.
The idea is that I don't want to transfer every single bit of the whole
database.
> You might be able to look into replication.
Would replication allow to create a fully matching copy wit using as little
data xfer as possible?
> Another option: if you have datetime columns (such as DateUpdated or
> DateInserted) stored on your tables you could create your own shipping
> routines...simply find the "recent" data (whatever that means to you) and
> insert it.
That could be a good solution, but it requires reworking much of the current
application.
> You could also use linked servers and INSERT INTO...SELECT...WHERE NOT
> EXISTS... you will have to fill in the blanks, but they should be fairly
> easy.
Surely, I can do that, however I was hoping for some more automatic (or
wizard-style) solution.
-- Pavils.|||"Pavils Jurjans" <pavils@.mailbox.riga.lv> wrote in message
news:uUQ1L7CkEHA.3968@.TK2MSFTNGP11.phx.gbl...
> Hello Keith,
>
not[vbcol=seagreen]
> The idea is that I don't want to transfer every single bit of the whole
> database.
understood
>
> Would replication allow to create a fully matching copy wit using as
little
> data xfer as possible?
That is the point of replication
>
and[vbcol=seagreen]
> That could be a good solution, but it requires reworking much of the
current
> application.
Probably so (if you don't have a way to identify "new" rows.
>
fairly[vbcol=seagreen]
> Surely, I can do that, however I was hoping for some more automatic (or
> wizard-style) solution.
This is not much different than my previous idea...it simply requires you to
type up some Transact-SQL. Hey, no one said being a dba/developer/IT
Professional was easy.
> -- Pavils.
>
Keithsql
I have two symmetrical databases on two SQL servers - one remote, the
other - local. The data in the remote server changes every minute. The data
in local SQL server is used for creating some reports and general testing.
Sometimes data from the local base can be irreversibly deleted.
I want to have some sort of procedure, that would do one-direction
synchronisation of the two databases. The data in the remote database should
overwrite the data in the local database, so that after the update the local
database is exact copy of the remote database. Of course, this kind of
result could be achieved by detaching the remote base and bringing over the
database files, or doing full DTS. However, detaching and transfering is not
a viable solution since the remote base is actively used and the size of
data files are considerable; And, doing full DTS would take years. There is
some need to do some comparison of record hashes etc., so that the size of
the data that have to be transfered is reduced to minimum.
Thanks,
PavilsThe first thing that comes to mind is BACKUP and RESTORE. After you restore
a database backup you will have an exact copy of the original (as of the
date and time the backup completed).
You mention that the size of the data files are "considerable." I am not
sure what that means, or if that is a limitation for you in your
environment.
You might be able to look into replication.
Another option: if you have datetime columns (such as DateUpdated or
DateInserted) stored on your tables you could create your own shipping
routines...simply find the "recent" data (whatever that means to you) and
insert it.
You could also use linked servers and INSERT INTO...SELECT...WHERE NOT
EXISTS... you will have to fill in the blanks, but they should be fairly
easy.
Keith
"Pavils Jurjans" <pavils@.mailbox.riga.lv> wrote in message
news:OsrHETCkEHA.3664@.TK2MSFTNGP12.phx.gbl...
> Hello,
> I have two symmetrical databases on two SQL servers - one remote, the
> other - local. The data in the remote server changes every minute. The
data
> in local SQL server is used for creating some reports and general testing.
> Sometimes data from the local base can be irreversibly deleted.
> I want to have some sort of procedure, that would do one-direction
> synchronisation of the two databases. The data in the remote database
should
> overwrite the data in the local database, so that after the update the
local
> database is exact copy of the remote database. Of course, this kind of
> result could be achieved by detaching the remote base and bringing over
the
> database files, or doing full DTS. However, detaching and transfering is
not
> a viable solution since the remote base is actively used and the size of
> data files are considerable; And, doing full DTS would take years. There
is
> some need to do some comparison of record hashes etc., so that the size of
> the data that have to be transfered is reduced to minimum.
> Thanks,
> Pavils
>|||Hello Keith,
> You mention that the size of the data files are "considerable." I am not
> sure what that means, or if that is a limitation for you in your
> environment.
The idea is that I don't want to transfer every single bit of the whole
database.
> You might be able to look into replication.
Would replication allow to create a fully matching copy wit using as little
data xfer as possible?
> Another option: if you have datetime columns (such as DateUpdated or
> DateInserted) stored on your tables you could create your own shipping
> routines...simply find the "recent" data (whatever that means to you) and
> insert it.
That could be a good solution, but it requires reworking much of the current
application.
> You could also use linked servers and INSERT INTO...SELECT...WHERE NOT
> EXISTS... you will have to fill in the blanks, but they should be fairly
> easy.
Surely, I can do that, however I was hoping for some more automatic (or
wizard-style) solution.
-- Pavils.|||"Pavils Jurjans" <pavils@.mailbox.riga.lv> wrote in message
news:uUQ1L7CkEHA.3968@.TK2MSFTNGP11.phx.gbl...
> Hello Keith,
>
not[vbcol=seagreen]
> The idea is that I don't want to transfer every single bit of the whole
> database.
understood
>
> Would replication allow to create a fully matching copy wit using as
little
> data xfer as possible?
That is the point of replication
>
and[vbcol=seagreen]
> That could be a good solution, but it requires reworking much of the
current
> application.
Probably so (if you don't have a way to identify "new" rows.
>
fairly[vbcol=seagreen]
> Surely, I can do that, however I was hoping for some more automatic (or
> wizard-style) solution.
This is not much different than my previous idea...it simply requires you to
type up some Transact-SQL. Hey, no one said being a dba/developer/IT
Professional was easy.
> -- Pavils.
>
Keithsql
Friday, March 23, 2012
Have two active SQL servers
Right now I am working on a project that requires two active SQL servers with
different local web servers accross geographic locations. I read an article
on Active-Active SQL servers and am seeing some possiblity in SQL 2005 for
this situation to work but in SQL 2000 is there any way to have two servers
with independent RAID arrays sync over the network to have different users
modifying records?
Thanks,
Joseph Ronzio
You are basically asking if you can use Majority Node Set (MNS) clustering.
SQL 2000 does not support it. Some third parties does have excellent
products that work. NSI is one that comes to mind, LeftHand software, etc.
Check my website for more.
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering
http://msmvps.com/clustering - Blog
"Joseph" <Joseph@.discussions.microsoft.com> wrote in message
news:772EC28E-A275-42B5-A610-3925F0C3542B@.microsoft.com...
> Right now I am working on a project that requires two active SQL servers
> with
> different local web servers accross geographic locations. I read an
> article
> on Active-Active SQL servers and am seeing some possiblity in SQL 2005 for
> this situation to work but in SQL 2000 is there any way to have two
> servers
> with independent RAID arrays sync over the network to have different users
> modifying records?
> Thanks,
> Joseph Ronzio
|||And SQL Server Clustering does not support 2 nodes modifying data in the
same physical DB (neither does Oracle or Sybase clustering).
If you need 2 nodes modifying data, look at replication.
Regards
Mike Epprecht, Microsoft SQL Server MVP
Zurich, Switzerland
IM: mike@.epprecht.net
MVP Program: http://www.microsoft.com/mvp
Blog: http://www.msmvps.com/epprecht/
"Rodney R. Fournier [MVP]" <rod@.die.spam.die.nw-america.com> wrote in
message news:#0Y8BfF5EHA.2624@.TK2MSFTNGP11.phx.gbl...
> You are basically asking if you can use Majority Node Set (MNS)
clustering.[vbcol=seagreen]
> SQL 2000 does not support it. Some third parties does have excellent
> products that work. NSI is one that comes to mind, LeftHand software, etc.
> Check my website for more.
> Cheers,
> Rod
> MVP - Windows Server - Clustering
> http://www.nw-america.com - Clustering
> http://msmvps.com/clustering - Blog
> "Joseph" <Joseph@.discussions.microsoft.com> wrote in message
> news:772EC28E-A275-42B5-A610-3925F0C3542B@.microsoft.com...
for[vbcol=seagreen]
users
>
|||So MySQL, Lotus Domino and IBM DB2 can do this but MS SQL can't? From my
reading it looks like SQL 2005 might be able to do this will it?
Thanks,
Joseph Ronzio
"Mike Epprecht (SQL MVP)" wrote:
> And SQL Server Clustering does not support 2 nodes modifying data in the
> same physical DB (neither does Oracle or Sybase clustering).
> If you need 2 nodes modifying data, look at replication.
> Regards
> --
> Mike Epprecht, Microsoft SQL Server MVP
> Zurich, Switzerland
> IM: mike@.epprecht.net
> MVP Program: http://www.microsoft.com/mvp
> Blog: http://www.msmvps.com/epprecht/
> "Rodney R. Fournier [MVP]" <rod@.die.spam.die.nw-america.com> wrote in
> message news:#0Y8BfF5EHA.2624@.TK2MSFTNGP11.phx.gbl...
> clustering.
> for
> users
>
>
|||Hi
No, SQL Server 2005 will not. There will still be one primary node that
handles all the work.
Clustering by definition, is for high availability, not load balancing.
Regards
Mike Epprecht, Microsoft SQL Server MVP
Zurich, Switzerland
IM: mike@.epprecht.net
MVP Program: http://www.microsoft.com/mvp
Blog: http://www.msmvps.com/epprecht/
"Joseph" <Joseph@.discussions.microsoft.com> wrote in message
news:E2BF74E7-ADEE-4C9F-B183-44D1D6C87122@.microsoft.com...[vbcol=seagreen]
> So MySQL, Lotus Domino and IBM DB2 can do this but MS SQL can't? From my
> reading it looks like SQL 2005 might be able to do this will it?
> Thanks,
> Joseph Ronzio
> "Mike Epprecht (SQL MVP)" wrote:
etc.[vbcol=seagreen]
servers[vbcol=seagreen]
2005[vbcol=seagreen]
|||That's only true about SQL Server!
Oracle RAC (and 10g grid) does support modifying the same shared physical DB
from any node. Basically, the same physical database is exposed to multiple
nodes with consistency controlled via distributed lock manager. I believe
Sybase has also just started to offer such a feature.
In theory at least, this approach offers both failover (HA) and scalability
(dynamic load balancing over more nodes since a query can be routed to any
node for processing).
Linchi
"Mike Epprecht (SQL MVP)" <mike@.epprecht.net> wrote in message
news:O6S7HdG5EHA.3416@.TK2MSFTNGP09.phx.gbl...
> And SQL Server Clustering does not support 2 nodes modifying data in the
> same physical DB (neither does Oracle or Sybase clustering).
> If you need 2 nodes modifying data, look at replication.
> Regards
> --
> Mike Epprecht, Microsoft SQL Server MVP
> Zurich, Switzerland
> IM: mike@.epprecht.net
> MVP Program: http://www.microsoft.com/mvp
> Blog: http://www.msmvps.com/epprecht/
> "Rodney R. Fournier [MVP]" <rod@.die.spam.die.nw-america.com> wrote in
> message news:#0Y8BfF5EHA.2624@.TK2MSFTNGP11.phx.gbl...
> clustering.
> for
> users
>
|||Ok apparently there was some confusion when I posted this I'm not talking the
same physcial database I am talking about replication where two database
servers have the same database but different copies. Can both database
servers have modification made to the database and replicate changes?
"Mike Epprecht (SQL MVP)" wrote:
> And SQL Server Clustering does not support 2 nodes modifying data in the
> same physical DB (neither does Oracle or Sybase clustering).
> If you need 2 nodes modifying data, look at replication.
> Regards
> --
> Mike Epprecht, Microsoft SQL Server MVP
> Zurich, Switzerland
> IM: mike@.epprecht.net
> MVP Program: http://www.microsoft.com/mvp
> Blog: http://www.msmvps.com/epprecht/
> "Rodney R. Fournier [MVP]" <rod@.die.spam.die.nw-america.com> wrote in
> message news:#0Y8BfF5EHA.2624@.TK2MSFTNGP11.phx.gbl...
> clustering.
> for
> users
>
>
|||You can do two-way replication, but what do you gain? Both servers not only
have to handle all the updates but they have to handle synching them with
their partner. The nodes won't split the work unless the database uses
partitioned views, something that only fits a very few scenarios.
Geoff N. Hiten
Microsoft SQL Server MVP
Senior Database Administrator
Careerbuilder.com
I support the Professional Association for SQL Server
www.sqlpass.org
"Joseph" <Joseph@.discussions.microsoft.com> wrote in message
news:41FDC5C9-93E1-4639-BDAC-BE3648F322A2@.microsoft.com...
> Ok apparently there was some confusion when I posted this I'm not talking
the[vbcol=seagreen]
> same physcial database I am talking about replication where two database
> servers have the same database but different copies. Can both database
> servers have modification made to the database and replicate changes?
> "Mike Epprecht (SQL MVP)" wrote:
etc.[vbcol=seagreen]
servers[vbcol=seagreen]
2005[vbcol=seagreen]
|||I would get the capability to have two servers in different geographic
regions with two different web farms for referencing and modifying data.
Granted there could be conflicts generated but in Lotus Domino today I don't
have that problem with our typical business records and I'd be migrating the
same application to SQL. I would just need the resources that would instruct
me how to do this for SQL.
"Geoff N. Hiten" wrote:
> You can do two-way replication, but what do you gain? Both servers not only
> have to handle all the updates but they have to handle synching them with
> their partner. The nodes won't split the work unless the database uses
> partitioned views, something that only fits a very few scenarios.
> --
> Geoff N. Hiten
> Microsoft SQL Server MVP
> Senior Database Administrator
> Careerbuilder.com
> I support the Professional Association for SQL Server
> www.sqlpass.org
> "Joseph" <Joseph@.discussions.microsoft.com> wrote in message
> news:41FDC5C9-93E1-4639-BDAC-BE3648F322A2@.microsoft.com...
> the
> etc.
> servers
> 2005
>
>
|||Have you considered Merge Replication?
Or if you want to use Transactional Replication then you can use Transactional Replication with Updating Subscribers.
Best Regards,
Uttam Parui
Microsoft Corporation
This posting is provided "AS IS" with no warranties, and confers no rights.
Are you secure? For information about the Strategic Technology Protection Program and to order your FREE Security Tool Kit, please visit http://www.microsoft.com/security.
Microsoft highly recommends that users with Internet access update their Microsoft software to better protect against viruses and security vulnerabilities. The easiest way to do this is to visit the following websites:
http://www.microsoft.com/protect
http://www.microsoft.com/security/guidance/default.mspx
different local web servers accross geographic locations. I read an article
on Active-Active SQL servers and am seeing some possiblity in SQL 2005 for
this situation to work but in SQL 2000 is there any way to have two servers
with independent RAID arrays sync over the network to have different users
modifying records?
Thanks,
Joseph Ronzio
You are basically asking if you can use Majority Node Set (MNS) clustering.
SQL 2000 does not support it. Some third parties does have excellent
products that work. NSI is one that comes to mind, LeftHand software, etc.
Check my website for more.
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering
http://msmvps.com/clustering - Blog
"Joseph" <Joseph@.discussions.microsoft.com> wrote in message
news:772EC28E-A275-42B5-A610-3925F0C3542B@.microsoft.com...
> Right now I am working on a project that requires two active SQL servers
> with
> different local web servers accross geographic locations. I read an
> article
> on Active-Active SQL servers and am seeing some possiblity in SQL 2005 for
> this situation to work but in SQL 2000 is there any way to have two
> servers
> with independent RAID arrays sync over the network to have different users
> modifying records?
> Thanks,
> Joseph Ronzio
|||And SQL Server Clustering does not support 2 nodes modifying data in the
same physical DB (neither does Oracle or Sybase clustering).
If you need 2 nodes modifying data, look at replication.
Regards
Mike Epprecht, Microsoft SQL Server MVP
Zurich, Switzerland
IM: mike@.epprecht.net
MVP Program: http://www.microsoft.com/mvp
Blog: http://www.msmvps.com/epprecht/
"Rodney R. Fournier [MVP]" <rod@.die.spam.die.nw-america.com> wrote in
message news:#0Y8BfF5EHA.2624@.TK2MSFTNGP11.phx.gbl...
> You are basically asking if you can use Majority Node Set (MNS)
clustering.[vbcol=seagreen]
> SQL 2000 does not support it. Some third parties does have excellent
> products that work. NSI is one that comes to mind, LeftHand software, etc.
> Check my website for more.
> Cheers,
> Rod
> MVP - Windows Server - Clustering
> http://www.nw-america.com - Clustering
> http://msmvps.com/clustering - Blog
> "Joseph" <Joseph@.discussions.microsoft.com> wrote in message
> news:772EC28E-A275-42B5-A610-3925F0C3542B@.microsoft.com...
for[vbcol=seagreen]
users
>
|||So MySQL, Lotus Domino and IBM DB2 can do this but MS SQL can't? From my
reading it looks like SQL 2005 might be able to do this will it?
Thanks,
Joseph Ronzio
"Mike Epprecht (SQL MVP)" wrote:
> And SQL Server Clustering does not support 2 nodes modifying data in the
> same physical DB (neither does Oracle or Sybase clustering).
> If you need 2 nodes modifying data, look at replication.
> Regards
> --
> Mike Epprecht, Microsoft SQL Server MVP
> Zurich, Switzerland
> IM: mike@.epprecht.net
> MVP Program: http://www.microsoft.com/mvp
> Blog: http://www.msmvps.com/epprecht/
> "Rodney R. Fournier [MVP]" <rod@.die.spam.die.nw-america.com> wrote in
> message news:#0Y8BfF5EHA.2624@.TK2MSFTNGP11.phx.gbl...
> clustering.
> for
> users
>
>
|||Hi
No, SQL Server 2005 will not. There will still be one primary node that
handles all the work.
Clustering by definition, is for high availability, not load balancing.
Regards
Mike Epprecht, Microsoft SQL Server MVP
Zurich, Switzerland
IM: mike@.epprecht.net
MVP Program: http://www.microsoft.com/mvp
Blog: http://www.msmvps.com/epprecht/
"Joseph" <Joseph@.discussions.microsoft.com> wrote in message
news:E2BF74E7-ADEE-4C9F-B183-44D1D6C87122@.microsoft.com...[vbcol=seagreen]
> So MySQL, Lotus Domino and IBM DB2 can do this but MS SQL can't? From my
> reading it looks like SQL 2005 might be able to do this will it?
> Thanks,
> Joseph Ronzio
> "Mike Epprecht (SQL MVP)" wrote:
etc.[vbcol=seagreen]
servers[vbcol=seagreen]
2005[vbcol=seagreen]
|||That's only true about SQL Server!
Oracle RAC (and 10g grid) does support modifying the same shared physical DB
from any node. Basically, the same physical database is exposed to multiple
nodes with consistency controlled via distributed lock manager. I believe
Sybase has also just started to offer such a feature.
In theory at least, this approach offers both failover (HA) and scalability
(dynamic load balancing over more nodes since a query can be routed to any
node for processing).
Linchi
"Mike Epprecht (SQL MVP)" <mike@.epprecht.net> wrote in message
news:O6S7HdG5EHA.3416@.TK2MSFTNGP09.phx.gbl...
> And SQL Server Clustering does not support 2 nodes modifying data in the
> same physical DB (neither does Oracle or Sybase clustering).
> If you need 2 nodes modifying data, look at replication.
> Regards
> --
> Mike Epprecht, Microsoft SQL Server MVP
> Zurich, Switzerland
> IM: mike@.epprecht.net
> MVP Program: http://www.microsoft.com/mvp
> Blog: http://www.msmvps.com/epprecht/
> "Rodney R. Fournier [MVP]" <rod@.die.spam.die.nw-america.com> wrote in
> message news:#0Y8BfF5EHA.2624@.TK2MSFTNGP11.phx.gbl...
> clustering.
> for
> users
>
|||Ok apparently there was some confusion when I posted this I'm not talking the
same physcial database I am talking about replication where two database
servers have the same database but different copies. Can both database
servers have modification made to the database and replicate changes?
"Mike Epprecht (SQL MVP)" wrote:
> And SQL Server Clustering does not support 2 nodes modifying data in the
> same physical DB (neither does Oracle or Sybase clustering).
> If you need 2 nodes modifying data, look at replication.
> Regards
> --
> Mike Epprecht, Microsoft SQL Server MVP
> Zurich, Switzerland
> IM: mike@.epprecht.net
> MVP Program: http://www.microsoft.com/mvp
> Blog: http://www.msmvps.com/epprecht/
> "Rodney R. Fournier [MVP]" <rod@.die.spam.die.nw-america.com> wrote in
> message news:#0Y8BfF5EHA.2624@.TK2MSFTNGP11.phx.gbl...
> clustering.
> for
> users
>
>
|||You can do two-way replication, but what do you gain? Both servers not only
have to handle all the updates but they have to handle synching them with
their partner. The nodes won't split the work unless the database uses
partitioned views, something that only fits a very few scenarios.
Geoff N. Hiten
Microsoft SQL Server MVP
Senior Database Administrator
Careerbuilder.com
I support the Professional Association for SQL Server
www.sqlpass.org
"Joseph" <Joseph@.discussions.microsoft.com> wrote in message
news:41FDC5C9-93E1-4639-BDAC-BE3648F322A2@.microsoft.com...
> Ok apparently there was some confusion when I posted this I'm not talking
the[vbcol=seagreen]
> same physcial database I am talking about replication where two database
> servers have the same database but different copies. Can both database
> servers have modification made to the database and replicate changes?
> "Mike Epprecht (SQL MVP)" wrote:
etc.[vbcol=seagreen]
servers[vbcol=seagreen]
2005[vbcol=seagreen]
|||I would get the capability to have two servers in different geographic
regions with two different web farms for referencing and modifying data.
Granted there could be conflicts generated but in Lotus Domino today I don't
have that problem with our typical business records and I'd be migrating the
same application to SQL. I would just need the resources that would instruct
me how to do this for SQL.
"Geoff N. Hiten" wrote:
> You can do two-way replication, but what do you gain? Both servers not only
> have to handle all the updates but they have to handle synching them with
> their partner. The nodes won't split the work unless the database uses
> partitioned views, something that only fits a very few scenarios.
> --
> Geoff N. Hiten
> Microsoft SQL Server MVP
> Senior Database Administrator
> Careerbuilder.com
> I support the Professional Association for SQL Server
> www.sqlpass.org
> "Joseph" <Joseph@.discussions.microsoft.com> wrote in message
> news:41FDC5C9-93E1-4639-BDAC-BE3648F322A2@.microsoft.com...
> the
> etc.
> servers
> 2005
>
>
|||Have you considered Merge Replication?
Or if you want to use Transactional Replication then you can use Transactional Replication with Updating Subscribers.
Best Regards,
Uttam Parui
Microsoft Corporation
This posting is provided "AS IS" with no warranties, and confers no rights.
Are you secure? For information about the Strategic Technology Protection Program and to order your FREE Security Tool Kit, please visit http://www.microsoft.com/security.
Microsoft highly recommends that users with Internet access update their Microsoft software to better protect against viruses and security vulnerabilities. The easiest way to do this is to visit the following websites:
http://www.microsoft.com/protect
http://www.microsoft.com/security/guidance/default.mspx
Have Svr name, need DBs
Hello Folks!
I have a bunch of servers, actually there are about 340.
I have the names of the servers but what I'd like to do is loop through
the servers and return all the databases on those servers.
I'd also like to find out how busy the servers are.
Got any ideas?
Thanks-In-AdvanceI wrote an SMO application for that. If you are interested, I can send
this to you. Though this is not yet perfect, it evaluates from the
needs of the community Just write me an email and I will share it. It
will eb additionally available on http://www.SQLServer2005.de (When the
site will be relaunched) soon.
HTH, Jens Suessmeyer.|||Jens, please send me the file, my email address is Bob@.BobSweeney.Net
Also, what is an SMO application? Am I supposed to know what that
acronym is?|||Assuming that you have an account that has the appropriate login and
permissions, put the names of all the servers in a text file with one on eac
h
line and run
for /F "tokens=1" %i in (servers.txt) do osql -S %i -E -Q
"sp_helpdb">>output.txt
The file output.txt will have have the listing of all the DBs on all the
servers.
For seeing how busy they are, use Performance Monitor (I think it's just
called "Performance" now) logs, or tool like MRTG.
If you posted to this forum through TechNet, and you found my answers
helpful, please mark them as answers.
"Bob" wrote:
> Hello Folks!
> I have a bunch of servers, actually there are about 340.
> I have the names of the servers but what I'd like to do is loop through
> the servers and return all the databases on those servers.
> I'd also like to find out how busy the servers are.
> Got any ideas?
> Thanks-In-Advance
>|||Slightly cleaner output:
for /F "tokens=1" %i in (servers.txt) do osql -S %i -d "master" -E -Q
"select @.@.SERVERNAME UNION ALL select catalog_name from
information_schema.schemata"
"Mark Williams" wrote:
> Assuming that you have an account that has the appropriate login and
> permissions, put the names of all the servers in a text file with one on e
ach
> line and run
> for /F "tokens=1" %i in (servers.txt) do osql -S %i -E -Q
> "sp_helpdb">>output.txt
> The file output.txt will have have the listing of all the DBs on all the
> servers.
> For seeing how busy they are, use Performance Monitor (I think it's just
> called "Performance" now) logs, or tool like MRTG.
> --
> If you posted to this forum through TechNet, and you found my answers
> helpful, please mark them as answers.
>
> "Bob" wrote:
>sql
I have a bunch of servers, actually there are about 340.
I have the names of the servers but what I'd like to do is loop through
the servers and return all the databases on those servers.
I'd also like to find out how busy the servers are.
Got any ideas?
Thanks-In-AdvanceI wrote an SMO application for that. If you are interested, I can send
this to you. Though this is not yet perfect, it evaluates from the
needs of the community Just write me an email and I will share it. It
will eb additionally available on http://www.SQLServer2005.de (When the
site will be relaunched) soon.
HTH, Jens Suessmeyer.|||Jens, please send me the file, my email address is Bob@.BobSweeney.Net
Also, what is an SMO application? Am I supposed to know what that
acronym is?|||Assuming that you have an account that has the appropriate login and
permissions, put the names of all the servers in a text file with one on eac
h
line and run
for /F "tokens=1" %i in (servers.txt) do osql -S %i -E -Q
"sp_helpdb">>output.txt
The file output.txt will have have the listing of all the DBs on all the
servers.
For seeing how busy they are, use Performance Monitor (I think it's just
called "Performance" now) logs, or tool like MRTG.
If you posted to this forum through TechNet, and you found my answers
helpful, please mark them as answers.
"Bob" wrote:
> Hello Folks!
> I have a bunch of servers, actually there are about 340.
> I have the names of the servers but what I'd like to do is loop through
> the servers and return all the databases on those servers.
> I'd also like to find out how busy the servers are.
> Got any ideas?
> Thanks-In-Advance
>|||Slightly cleaner output:
for /F "tokens=1" %i in (servers.txt) do osql -S %i -d "master" -E -Q
"select @.@.SERVERNAME UNION ALL select catalog_name from
information_schema.schemata"
"Mark Williams" wrote:
> Assuming that you have an account that has the appropriate login and
> permissions, put the names of all the servers in a text file with one on e
ach
> line and run
> for /F "tokens=1" %i in (servers.txt) do osql -S %i -E -Q
> "sp_helpdb">>output.txt
> The file output.txt will have have the listing of all the DBs on all the
> servers.
> For seeing how busy they are, use Performance Monitor (I think it's just
> called "Performance" now) logs, or tool like MRTG.
> --
> If you posted to this forum through TechNet, and you found my answers
> helpful, please mark them as answers.
>
> "Bob" wrote:
>sql
Have SQL report IP of query-er
I am trying to come up with a way to save the IP of the last computer that ran an update in our servers. I don't want to have to re-write the .net code, so I am wondering if there is a way, using functions available in SQL 2000 that I can get the IP of the machine running the update.
I know that SQL server 'knows' this information because it can log it in it's log files, but I don't know how to get at it from a stored procedure.
Thanks.
DECLARE @.Hostname VARCHAR(255)
SET @.Hostname=host_name()
CREATETABLE #Results
(
Results VARCHAR(4000)
)
DECLARE @.Commandstring VARCHAR(300)
SET @.Commandstring='ping '+ @.Hostname
INSERTINTO #Results
EXEC master..xp_cmdshell @.Commandstring
SelectDISTINCTSUBSTRING(Results,12,CHARINDEX(':',Results)-12)AS HostIpAdressfrom #Results
Where ResultsLIKE'Reply from%'
DROPTABLE #Results
Wednesday, March 21, 2012
has parallelism helped any of you ?
It seems to hurt more than help that we like to turn it off on all servers
by default. Has anyone seen advantages of a parallised query vs one that
doesnt go through one
Using SQL 2000Yes. It has helped. However, if a Query is processed with Parellelism, we
check
each of the Query and make sure that it does help indeed. (On a Case to Case
Basis)
In very rare cases, did we have to use OPTION (MAXDOP 1). But later we
realised that "Exec SPname WITH RECOMPILE" actually helped remove
Parellelism.
Gopi
"Hassan" <fatima_ja@.hotmail.com> wrote in message
news:%23nzTh5FUFHA.752@.TK2MSFTNGP10.phx.gbl...
> It seems to hurt more than help that we like to turn it off on all servers
> by default. Has anyone seen advantages of a parallised query vs one that
> doesnt go through one
> Using SQL 2000
>|||Hassan
I'd not change a deafult configiration of SQL Server instead I'd tune the
queries and see if it hurts perfomance by using parallelism try to opotimize
it perhaps by using MAXDOP(1) hint
"Hassan" <fatima_ja@.hotmail.com> wrote in message
news:%23nzTh5FUFHA.752@.TK2MSFTNGP10.phx.gbl...
> It seems to hurt more than help that we like to turn it off on all servers
> by default. Has anyone seen advantages of a parallised query vs one that
> doesnt go through one
> Using SQL 2000
>sql
by default. Has anyone seen advantages of a parallised query vs one that
doesnt go through one
Using SQL 2000Yes. It has helped. However, if a Query is processed with Parellelism, we
check
each of the Query and make sure that it does help indeed. (On a Case to Case
Basis)
In very rare cases, did we have to use OPTION (MAXDOP 1). But later we
realised that "Exec SPname WITH RECOMPILE" actually helped remove
Parellelism.
Gopi
"Hassan" <fatima_ja@.hotmail.com> wrote in message
news:%23nzTh5FUFHA.752@.TK2MSFTNGP10.phx.gbl...
> It seems to hurt more than help that we like to turn it off on all servers
> by default. Has anyone seen advantages of a parallised query vs one that
> doesnt go through one
> Using SQL 2000
>|||Hassan
I'd not change a deafult configiration of SQL Server instead I'd tune the
queries and see if it hurts perfomance by using parallelism try to opotimize
it perhaps by using MAXDOP(1) hint
"Hassan" <fatima_ja@.hotmail.com> wrote in message
news:%23nzTh5FUFHA.752@.TK2MSFTNGP10.phx.gbl...
> It seems to hurt more than help that we like to turn it off on all servers
> by default. Has anyone seen advantages of a parallised query vs one that
> doesnt go through one
> Using SQL 2000
>sql
Monday, March 19, 2012
Hardware?
Hi everyone,
First please excuse my ignorance. I am a consultant and generally work with
small business servers and I have no experience with clustering.
I have a nonprofit client that has a need for a small database (500mb or
so) and wants to use 2 Dell workstations as the servers. I know that he
should be using true server hardware, but they just can't afford it. I know
that the servers need a shared disk array and a heartbeat cable to cluster.
We are looking at a Dell Powervault 221 with 2 36gb mirrored drives. Can
someone please recommend SCSI cards for use in the workstations. The
computers only have 32bit PCI ports. Will this be a problem? I know that
the workstations will need 2 NIC per system. Am I missing anything else?
Also, they have purchased 2 copies of Windows 2003 server Enterprise. They
also have 1 copy of SQL 2000 enterprise. Do they need a second copy of SQL?
I appreciate any help that is provided.
Brian
Clustering just for the sake of having a cluster is almost as bad an idea as
buying an elephant because you need a handy way to get rid of peanuts. It
may work, but it will certainly cause more problems than it solves. A
cluster is part of a high availability solution. Building a cluster out of
low-end parts decreases the overall availability, not increases it. A
cluster is an enterprise-class solution, soon to be pushed down to
department-class, but is certainly not a workstation-grade solution. You
will actually be further from your availability goals with such a system as
you propose. IMHO, I would abandon the cluster idea for now. I would
purchase a small workgroup or department grade Dell server with some drive
redundancy. By the time you purchase the extra clustering hardware and the
workstations, you can certainly afford a basic server-grade machine. Trade
in the Enterprise SQL licenses for Standard Edition Licenses. Your system
will run better and your client will be happier.
Geoff N. Hiten
Microsoft SQL Server MVP
"Brian Z" <bzaayer@.yahoo.com> wrote in message
news:eF%23yFO2VFHA.2984@.tk2msftngp13.phx.gbl...
> Hi everyone,
> First please excuse my ignorance. I am a consultant and generally work
> with small business servers and I have no experience with clustering.
> I have a nonprofit client that has a need for a small database (500mb
> or so) and wants to use 2 Dell workstations as the servers. I know that
> he should be using true server hardware, but they just can't afford it. I
> know that the servers need a shared disk array and a heartbeat cable to
> cluster. We are looking at a Dell Powervault 221 with 2 36gb mirrored
> drives. Can someone please recommend SCSI cards for use in the
> workstations. The computers only have 32bit PCI ports. Will this be a
> problem? I know that the workstations will need 2 NIC per system. Am I
> missing anything else? Also, they have purchased 2 copies of Windows 2003
> server Enterprise. They also have 1 copy of SQL 2000 enterprise. Do they
> need a second copy of SQL?
> I appreciate any help that is provided.
> Brian
>
|||I agree 100% with Geoff. With such a small db they have lots of options to
be up and running quickly. Heck even restoring a backup on another server
should only take a minute. You might want to look into Log Shipping or with
SQL2005 you can do Database Mirroring. Both much cheaper than Clustering
and can be done with minimum hardware.
Andrew J. Kelly SQL MVP
"Geoff N. Hiten" <sqlcraftsman@.gmail.com> wrote in message
news:Outwag6VFHA.2684@.TK2MSFTNGP09.phx.gbl...
> Clustering just for the sake of having a cluster is almost as bad an idea
> as buying an elephant because you need a handy way to get rid of peanuts.
> It may work, but it will certainly cause more problems than it solves. A
> cluster is part of a high availability solution. Building a cluster out
> of low-end parts decreases the overall availability, not increases it. A
> cluster is an enterprise-class solution, soon to be pushed down to
> department-class, but is certainly not a workstation-grade solution. You
> will actually be further from your availability goals with such a system
> as you propose. IMHO, I would abandon the cluster idea for now. I would
> purchase a small workgroup or department grade Dell server with some drive
> redundancy. By the time you purchase the extra clustering hardware and
> the workstations, you can certainly afford a basic server-grade machine.
> Trade in the Enterprise SQL licenses for Standard Edition Licenses. Your
> system will run better and your client will be happier.
> Geoff N. Hiten
> Microsoft SQL Server MVP
>
> "Brian Z" <bzaayer@.yahoo.com> wrote in message
> news:eF%23yFO2VFHA.2984@.tk2msftngp13.phx.gbl...
>
First please excuse my ignorance. I am a consultant and generally work with
small business servers and I have no experience with clustering.
I have a nonprofit client that has a need for a small database (500mb or
so) and wants to use 2 Dell workstations as the servers. I know that he
should be using true server hardware, but they just can't afford it. I know
that the servers need a shared disk array and a heartbeat cable to cluster.
We are looking at a Dell Powervault 221 with 2 36gb mirrored drives. Can
someone please recommend SCSI cards for use in the workstations. The
computers only have 32bit PCI ports. Will this be a problem? I know that
the workstations will need 2 NIC per system. Am I missing anything else?
Also, they have purchased 2 copies of Windows 2003 server Enterprise. They
also have 1 copy of SQL 2000 enterprise. Do they need a second copy of SQL?
I appreciate any help that is provided.
Brian
Clustering just for the sake of having a cluster is almost as bad an idea as
buying an elephant because you need a handy way to get rid of peanuts. It
may work, but it will certainly cause more problems than it solves. A
cluster is part of a high availability solution. Building a cluster out of
low-end parts decreases the overall availability, not increases it. A
cluster is an enterprise-class solution, soon to be pushed down to
department-class, but is certainly not a workstation-grade solution. You
will actually be further from your availability goals with such a system as
you propose. IMHO, I would abandon the cluster idea for now. I would
purchase a small workgroup or department grade Dell server with some drive
redundancy. By the time you purchase the extra clustering hardware and the
workstations, you can certainly afford a basic server-grade machine. Trade
in the Enterprise SQL licenses for Standard Edition Licenses. Your system
will run better and your client will be happier.
Geoff N. Hiten
Microsoft SQL Server MVP
"Brian Z" <bzaayer@.yahoo.com> wrote in message
news:eF%23yFO2VFHA.2984@.tk2msftngp13.phx.gbl...
> Hi everyone,
> First please excuse my ignorance. I am a consultant and generally work
> with small business servers and I have no experience with clustering.
> I have a nonprofit client that has a need for a small database (500mb
> or so) and wants to use 2 Dell workstations as the servers. I know that
> he should be using true server hardware, but they just can't afford it. I
> know that the servers need a shared disk array and a heartbeat cable to
> cluster. We are looking at a Dell Powervault 221 with 2 36gb mirrored
> drives. Can someone please recommend SCSI cards for use in the
> workstations. The computers only have 32bit PCI ports. Will this be a
> problem? I know that the workstations will need 2 NIC per system. Am I
> missing anything else? Also, they have purchased 2 copies of Windows 2003
> server Enterprise. They also have 1 copy of SQL 2000 enterprise. Do they
> need a second copy of SQL?
> I appreciate any help that is provided.
> Brian
>
|||I agree 100% with Geoff. With such a small db they have lots of options to
be up and running quickly. Heck even restoring a backup on another server
should only take a minute. You might want to look into Log Shipping or with
SQL2005 you can do Database Mirroring. Both much cheaper than Clustering
and can be done with minimum hardware.
Andrew J. Kelly SQL MVP
"Geoff N. Hiten" <sqlcraftsman@.gmail.com> wrote in message
news:Outwag6VFHA.2684@.TK2MSFTNGP09.phx.gbl...
> Clustering just for the sake of having a cluster is almost as bad an idea
> as buying an elephant because you need a handy way to get rid of peanuts.
> It may work, but it will certainly cause more problems than it solves. A
> cluster is part of a high availability solution. Building a cluster out
> of low-end parts decreases the overall availability, not increases it. A
> cluster is an enterprise-class solution, soon to be pushed down to
> department-class, but is certainly not a workstation-grade solution. You
> will actually be further from your availability goals with such a system
> as you propose. IMHO, I would abandon the cluster idea for now. I would
> purchase a small workgroup or department grade Dell server with some drive
> redundancy. By the time you purchase the extra clustering hardware and
> the workstations, you can certainly afford a basic server-grade machine.
> Trade in the Enterprise SQL licenses for Standard Edition Licenses. Your
> system will run better and your client will be happier.
> Geoff N. Hiten
> Microsoft SQL Server MVP
>
> "Brian Z" <bzaayer@.yahoo.com> wrote in message
> news:eF%23yFO2VFHA.2984@.tk2msftngp13.phx.gbl...
>
Subscribe to:
Posts (Atom)