Dynamics CRM 2011 Bulk Update

Question

Running Dynamics CRM 2011 rollout 3. Need to update millions of customer records periodically (delta updates). Using standard update (one by one) takes a few weeks. Also we don\

Answer 1

I realize this is post is over 2 years old, but I can add to it in case someone else reads it and has a similar need.

Peter Majeed's answer is on target in that CRM processes requests one record at a time. There is no bulk edit that works the way you are looking for. I encourage you not to touch the DB directly if you need/want Microsoft support.

If you are looking at periodic updates of millions of records, you have a few options. Consider using Scribe or develop your own custom import utility or script using the CRM SDK.

Scribe is probably going to be your best option since it is cost effective for data imports and will allow you to easily update and insert from the same file.

If you write your own .Net/SDK based utility, I'd suggest making it multithreaded and either programmatically break up your input file in memory or on disk and have each thread work with its own subset of the data - that is, of course, if the order of execution does not have to be chronological according to the contents of the input file. If you can divide and conquer the input file over multiple threads, you can reduce the overall execution time considerably. Also, if your corporate policy allows you to have access to one of the CRM Servers and you can place your code directly on the server and execute it from there - you can eliminate the network latency between a workstation running the code and the CRM web services.

Last but not least, if this large volume of import data is coming from another system, you can write a CRM plug-in to run on the Retrieve and RetrieveMultiple messages (events) in CRM for your specific entity, programmatically retrieve the desired data from the other system (and if the other system is unavailable - just use the cached copy in CRM), and keep CRM up to date in real-time or on a 'last-cached-on' basis. This is certainly more coding effort, but it potentially eliminates the need for a large synchronization job to be run every few weeks.