Best .NET3.5 Based Strategy for Creating a Disabled db Application

Question

My requirements:

Relatively small database: 25-30 tables with 4-5 thousand rows in each. 4-5 users with their copies of the local database on their laptops. Synchronization as needed or at the end of the day. Free solution, as I am doing this setup for ogranization NFP.

In the last few days I have been exploring various options. Finally, I thought about narrowing to datasets, and serializing them into XML, but it seems like this is not a good idea after all (answers to my previous questions were highly recommended).

So now my question is which approach (database and access technology) should I take? I would like to use C # and easy-to-use technology to access my database and synchronize it at the end of the day.

According to this link: http://msdn.microsoft.com/en-us/library/8fxztkff.aspx, MS offers two different approaches. I am surprised to see that they did not mention LINQ here.

So which way do you recommend?

Thanks so much for reading.

Answer 1

Add → Create → “Local Database Cache”. This should provide all the necessary synchronization (via the ADO.NET Sync Framework) running (IIRC) by SQL CE by default, which (I think) is free.

Answer 2

Check out the Microsoft Sync Framework .

Introducing Microsoft Sync Framework - a comprehensive platform for synchronizing collaboration and offline applications, services and devices supporting any type of data, any data storage, any transfer protocol and network topology.

Answer 3

If you already have SQL Server (standard or higher), you can use SQL Server Replication Services to synchronize local copies of the database with an instance of SQL Express or SQL Server Compact on the user's computer (both of which are free).

Try here for more information.

It looks like you are looking at a merge replication scenario.

Answer 4

Microsoft has a template for this: A randomly connected client .

Answer 5

In OP:

One word of warning that I will give you at the front of the Sync platform is to upgrade the database schema on all versions. I use it in the project that I am currently developing, and I start the specification "2.0", which I decided to delete it for various reasons.

I would really like you to take the time to understand how you want to deal with circuit changes as you move forward and decide whether this is how you want. It was for me to start the system, but now that I am dealing with a nightmare, I am falling prey to the NIH script.

My prediction for 2.0 is to continue using SqlCE for clients and SQL Server 2008 for COR ... but for collapsing my own Sync Fx based on heavy use of reflection and a set of DTO objects.