Master Data Services miscellaneous
As we head into the US holiday season, I wanted to catch up on some things I've wanted to blog for some time.
First of all, it's good to see some blogging activity around MDS.
As is the usual case around the first release of a product, it takes some time for blog activity to get momentum. But, we're beginning to see some action now as folks pick up the November CTP of SQL Server R2 and start to put it through it's paces.
This one by Nick Barclay is a series of seven introducing MDS and how to build a model and starts to get into the data model a bit as well.
For those just picking up the product for the first time, this is an excellent quick start guide.
Intrepid blogger and SSIS/MSBI geek Jamie Thomson has also been digging into the product and started blogging and he promises more to come.
Roland Lenz gives a pretty good summary of MDS and talks a little about scenarios in which it can be used.
Chris Webb blogged about John McAllister's recent presentation at PASS.
For those that missed it, I have a short entry here that talks at a high level about the product and Master Data Management in general.
Shaun Ryan has also started to blog and has one about creating hierarchies in MDS here.
The MDS team will be blogging as well. Look for those blogs to show up soon and I'll link here.
Matt Anderson has already created the first entry: Installing and Configuring Master Data Services 2008 R2 November CTP
Hooo, that's a mouth full. Fortunately, it takes about as long to say it as it does to do it! As soon as we can get the blog up and running, he'll be posting that entry with more to follow.
There are various comments about what's new in the November CTP MDS relative to the original product Microsoft acquired. Some have said it's basically the same thing. I'd like to give you all some background on that. For the benefit of those who are familiar with the Stratature product or the technology preview we released over a year ago, I'll briefly list some of the more important improvements. Later, I'll drop some hints as to what's coming next. But keep it between me and you.
The team has been doing some great work around improving the performance of the product. We've improved the schema, optimized stored procedures, and changed the way we manage key elements such as business rules to be more efficient. This work has resulted some great performance improvements. In some cases, the improvements are orders of magnitude faster.
We've done a load of work on security, accessibility, globalization and localization (21 languages), better setup and configuration.
While the user interface is mostly the same, we have done some changes to improve the experience including fixing bugs and swapping the control library for better management.
The perenial problem with building models is how to deploy them from the machine where they were developed without a lot of difficulty. And what's more, how do you build a model template and use it over and over without said efforts? Model deployment. With Model Deployment, you can build an entire model, including data on a development machine, deploy it to a test lab for testing and then easily deploy to the live production machines without ever getting your hands dirty. It's also useful for building pre-developed models for solution specialists or for sharing across the enterprise. While competitor products can sometimes take weeks to setup, configure and build the first model, I've done a demo where I setup, configured, deployed a model and built a derived hierarchy in roughly 10 minutes. That's simpler, easier and faster. That's Microsoft software.
We've added a comprehensive API to the product with both bulk and message oriented functions. Then we rewrote the user interface to go through the API instead of directly to the database/stored procedures. Anything you can do through the UI and more, you can now do through the API. It is complete and comprehensive.
So, while the product looks pretty much the same, much of the work we've been doing is to build the MDS platform into something that will support a rich set of experiences, integrations and additional scenarios.
It's a little early to be talking about the next version don't you think? OK, well, let's just say we're going to improve the user interface, integration and out of box experience and a few other goodies.