August 3, 2009 at 11:12 am
I want to have one main server which holds all the data which is connected to two other servers which have the same data replicated as long as they are connected.
When the connection is lost the two sub server should go on with the data entry and sync back to main server when the connection is established.
What is the best possible solution in sqlserver.
What are the problems that may arise as all the three servers will be updated while they are not connected.
August 4, 2009 at 12:20 pm
That's a business continuity project and just to get a decent scope of work much more information would be needed.
_____________________________________
Pablo (Paul) Berzukov
Author of Understanding Database Administration available at Amazon and other bookstores.
Disclaimer: Advice is provided to the best of my knowledge but no implicit or explicit warranties are provided. Since the advisor explicitly encourages testing any and all suggestions on a test non-production environment advisor should not held liable or responsible for any actions taken based on the given advice.August 4, 2009 at 9:28 pm
This is related to newly created municipal corp. data gathering project.
The municipal corp. is created from 3 small towns and so there are 3 centers for data collection.
But in future the data will be available from any point.
There are no lease lines and the only option is VPN through broadband (this can be a slow connection).
The broadband connection are not that good over here, hence the need for syncing when connected (some times for 1-2 days they may be disrupted).
As the data will be gathered form all three centers and processed as required from any of the three centers I think merge replication will be the best bet.
August 5, 2009 at 6:24 am
You are correct, merge replication might be a good option but unreliable network is a major caveat.
a couple of follow up questions:
1- What volume of data is expected to be collected in each point in a daily basis?
2- Does data from -lets say sites #2 and #3 has to be up-to-the-minute when somebody queries it from site #1?
3- What's the physical distance in between sites #1, #2 and #3?
_____________________________________
Pablo (Paul) Berzukov
Author of Understanding Database Administration available at Amazon and other bookstores.
Disclaimer: Advice is provided to the best of my knowledge but no implicit or explicit warranties are provided. Since the advisor explicitly encourages testing any and all suggestions on a test non-production environment advisor should not held liable or responsible for any actions taken based on the given advice.August 5, 2009 at 9:16 am
These are thing we have to decide as this is still in primary stage and we can still constrain the user for now like
1. Updation of specific data on specific server to keep data consistency.
2. Reporting to be done only after synced data or on a backup database which is synced till yesterday.
But once the thing is operational then may be this will be difficult but the amount of data entered on day to day basis will be very less.
About size I am not sure yet but will let you now once the table designs are ready after preliminary working.
August 5, 2009 at 9:26 am
... how about the physical distance in between the sites?
Any ball-park idea about the size of the databases?... like most likely less than 10 Gig... between 10 to 50 Gig... 50-100 Gig... larger than 100 Gig?
_____________________________________
Pablo (Paul) Berzukov
Author of Understanding Database Administration available at Amazon and other bookstores.
Disclaimer: Advice is provided to the best of my knowledge but no implicit or explicit warranties are provided. Since the advisor explicitly encourages testing any and all suggestions on a test non-production environment advisor should not held liable or responsible for any actions taken based on the given advice.August 5, 2009 at 9:38 am
50-100 GB would be correct slot
Viewing 7 posts - 1 through 6 (of 6 total)
You must be logged in to reply to this topic. Login to reply