I hope this is the right forum to ask this question - I couldn't find any specific SQL forums to post this so here goes...
We are hoping to use Remote Blob Storage (RBS) and are running some tests to check suitability. When I first read the documentation I had assumed that RBS was something you install, configure, and it "just works" i.e. the client applications write to SQL,
and SQL takes care of moving the files on via RBS. But after implementing a test following the sample code it looks like the application/client still needs to be RBS-aware and has to stream data into/out of SQL via an RBS client provider.
Am I missing something here? I had hoped that existing client applications would not have to change their logic i.e. they could continue to access SQL Server in exactly the same way, but would benefit by RBS moving blobs from binary/image fields to RBS-enable
provider storage. But all the samples point to the clients having to use .NET libraries to interact with the RBS functionality, and while SQL then takes care of the rest it would mean any client would have to be modified to support this.
If that's the case, I don't see the point to RBS. Yes, it's fantastic that we can store blobs outside of SQL via a common API, and yes, it is transaction-aware so blob clean-up is a much better experience, but the client has to still do the work to enable
View Complete Post