Remote Distributor question

sql2k sp3
BoxA Publisher.
BoxB Subscriber. To be used for reporting.
I want the processing drain of Replication to be done on
BoxB. So I was thinking of putting the Distributor on BoxB
and using a Push Subscription.
Do any of you gurus see any problems with my idea?
TIA, Chris?
Reporting services chews up a lot of cycles. You might find it better to
keep the distributor on the publisher.
The point you start to move the distributor off the publisher is when you
start to get significant locking on your distribution database.
Hilary Cotter
Looking for a book on SQL Server replication?
http://www.nwsu.com/0974973602.html
"ChrisR" <anonymous@.discussions.microsoft.com> wrote in message
news:0a4701c46e74$e3ab90f0$a601280a@.phx.gbl...
> sql2k sp3
> BoxA Publisher.
> BoxB Subscriber. To be used for reporting.
> I want the processing drain of Replication to be done on
> BoxB. So I was thinking of putting the Distributor on BoxB
> and using a Push Subscription.
> Do any of you gurus see any problems with my idea?
> TIA, Chris?
|||>Reporting services chews up a lot of cycles.
My bad. I wont be using actual Reporting Services.
You might find it better to
>keep the distributor on the publisher.
>The point you start to move the distributor off the
publisher is when you
>start to get significant locking on your distribution
database.
>
Why is this? Why do people do it then?

>--Original Message--
>Reporting services chews up a lot of cycles. You might
find it better to
>keep the distributor on the publisher.
>The point you start to move the distributor off the
publisher is when you
>start to get significant locking on your distribution
database.
>--
>Hilary Cotter
>Looking for a book on SQL Server replication?
>http://www.nwsu.com/0974973602.html
>
>"ChrisR" <anonymous@.discussions.microsoft.com> wrote in
message[vbcol=seagreen]
>news:0a4701c46e74$e3ab90f0$a601280a@.phx.gbl...
BoxB
>
>.
>
|||Well, most people do it to reduce the load on the Publisher. The question is
when should you do it, and there is no real rule of thumb, at least as far
as I am aware.
The deciding factor for me is when you experience locking in your
distribution database.
Hilary Cotter
Looking for a book on SQL Server replication?
http://www.nwsu.com/0974973602.html
"ChrisR" <anonymous@.discussions.microsoft.com> wrote in message
news:0c4201c46e8a$e7dd03a0$a601280a@.phx.gbl...[vbcol=seagreen]
> My bad. I wont be using actual Reporting Services.
> You might find it better to
> publisher is when you
> database.
> Why is this? Why do people do it then?
> find it better to
> publisher is when you
> database.
> message
> BoxB

Remote cube processing

Hi,

I'm using Informatica 8 for ETL procedures and I would like my SASS 2005 to process a cube as the ETL ends.
Does anyone know if there's a component for Informatica that supports XMLA (or any other way to process the cube remotly) ?

Thanks in advance,

Ariel.

Hi,

you can use a command line tool that is avaliable from microsoft,

it is called as ASCMD.

Check this link for download and other information on ASCMD. : http://msdn2.microsoft.com/en-us/library/ms365187.aspx

Hope this helps

Regards

|||

Hi Vijay,

I'll try it on my system and get back here (it's not connected to the internet).

Thanks a lot.

|||Hi all,

Well, I checked my sytem and my ETL server is linux so I can't use ascmd because it works only on windows servers.
Any workaround or suggestions?

Thanks in advance,
Ariel.|||

Hi,

In which database is the data stored after the ETL?

Regards

|||Hi,

The DB is Oracle.
Infronatica is installed on Linux server.|||

Hi,

This is what you can do:

Set up Http access to your Analysis Server 2005.

Link to setup http access: http://www.microsoft.com/technet/prodtechnol/sql/2005/httpasws.mspx

From your Linux system, which contains Informatica:

After your ETL completes,

pass a pre-constructed xmla command (Http call) ,

to the Http access point (msmdpump.dll) of the Analysis Server 2005.

This xmla command would contain a "Process" batch command,

which instructs the Analysis Server to process a specific object (Cube/MeasureGroup/Partition/dimension).

Infact you can even create new partitions and then process them dynamically using XMLA.

xmla overview :

http://msdn2.microsoft.com/en-us/library/ms187178.aspx

xmla command for processing analysis server 2005 objects:

http://msdn2.microsoft.com/en-us/library/ms187199.aspx

In Analysis Server 2005 you would have the data source configured to get data from Oracle or whichever data stores you have.

Hope this helps.

Regards