Was wondering if there was any specific thought into accomodating peer to peer replication for commerce server 2007 db's. It appears that the transactions db has done away with identity fields but other cs db's still use them. I was wondering if there was a white paper or even better someone who has already implemented this. Is there anything specific to commerce server I should be aware of before attempting this or do I just need to follow the standard implementation of peer to peer replication Thanks in advance....
-- Sean

Commerce Server 2007 and Peer to Peer replication
KIWIDOGGIE
I'm wondering if a reader/writer scenario makes sense where you make changes on a server and then replicate it out (perhaps using the staging service, or otherwise).
It's a shame that replication wasn't tested. I agree with you that it's a great cheap clustering solution. I was looking at NSI DoubleTake as a route to go to for that.
Have you considered other options like partitioning
Cheers,
Colin
Mike!
Ori'
The core issue you will face is that SQL database replication has not been extensively tested by us so we will not be able to support you in the event that you have problems doing this.
The recomended approach - and the one heavily tested by us - is to use CSS (staging) for the scenario you describe. You can publish to multiple targets this way and achieve whatever scaleout you are looking for.
Samurai Sjakkie
Alexey Raga
sammy chen
Personally PTP and load balancing is the sensible mans solution. If you database architecture can support it its a completely high available solution. Each node can be taken out with out the loss of service, it also can run on cheap hardware and has no recovery time, and failover is limited to only the time it takes your load balancing to realise, whether that be your app, or a load balance appliance.
GrandpaB
Alex Yakhnin - MSFT
Nonu_k
Sean,
I'm curious - what are you trying to achieve through peer replication
Cheers,
Colin
Michael Hansen