In addition to high profile projects like Gemini and Madison, SQL Server 2008 R2 will include several other enhancements for business intelligence and data warehousing.
We talked with Microsoft's Herain Oberoi, group product manager with the SQL Server business group, about some of these changes, including improvements to Reporting Services and the new Master Data Services BI utility.
Oberoi is scheduled to present at PASS Summit 2009 in a session entitled Overview of Business Intelligence capabilities in SQL Server 2008 & R2.
It sounds like there are some big changes to Reporting Services in SQL Server 2008 R2. Can you outline what some of those improvements are?
Herain Oberoi: In every release we do, we invest in business intelligence as a big theme. So in this release the big theme that we focused on is of course this notion of managed self-service.
I spoke before about our PowerPivot for Excel and SharePoint , but we also have existing tools like Reporting Services that we are improving, with the idea in mind being "How do we make it easier for end-users to build reports and interact with reports?"
So some of the key improvements we have coming out for instance is this notion of grab-and-go reporting or reusable report parts, which basically allows an end-user to not have to build reports from scratch. Instead they can go into Report Builder and [Microsoft SharePoint] and then look at a set of pre-built report parts -- much like you would when putting together a dashboard with pre-built Web parts. You could pull a chart that someone else built and a table that a third person built, and with all these things together -- without writing much code -- assemble a report together and publish that back out. And then when you publish that back out even that gets shared with rest of the community.
So we are really moving from this notion of an individual building a report from scratch to a more collaborative version of building reports, which is a report that's really a sum of parts from different pieces of information. So how do you allow end-users to pull that stuff in? That's going to be a big focus for a while.
Master Data Services is a new feature that was created from software that was originally a Stratature product before Microsoft acquired it. What can you tell us about Master Data Services and what it provides to SQL Server 2008 R2 users?
Oberoi: Mater Data Services is new and we haven't been speaking too much to it, but it is sort of a big deal. This is because if you think about SQL Server as a data platform, being more than just a database, you know we don't just provide Reporting Services and Analysis Services and Integration Services.
In [SQL Server 2008 R2] we're actually producing a whole new service and that's Master Data Services. And so, as you mentioned before, it came from an acquisition of Stratature that we did over a year ago.
It enables a sort of consistency of information across systems in organizations. If I have two operational systems, I have an ERP system in location A and a slightly different system in location B, and I have some information in it, like say product information, and I have to go to update the product information in system A. How do you make sure that you know the version of the product in system B is the same? And how do you keep those in sync?
So that's where Master Data Services comes in.
So if you are going to build a BI solution, you don't want to be guessing which system you go to, you want that master data to reference, or to say "This is what my single version of the truth is and that's where I'm going to get it." So there's this notion of operational master data, which is keeping systems in sync, and then there's this notion of analytic master data, which is not just keeping my traditional line of business systems in sync, but also keeping both things in sync with my data warehouse.
And so what Master Data Services provides is a few things. It's a centralized hub, it's a portal, and it provides a set of capabilities to run a process. Those three things together then let you keep both the operational systems and your analytic systems in sync. And it's not just about the systems themselves, it's about the people who interact with the systems, too.
So say you have an end-user who defines sales in a given month in a certain way. If that definition changes from a business perspective, then that end-user can go into the data management portal and change that definition. And now your master data system is also up-to-date based on what business rules are changing.
So what the set of services ends up providing is, again, a single version of truth, which for business becomes extremely important because then you are making decisions on consistent information and the most up-to-date information.
So it really works well with the rest of the BI platform that we have, which is part of why we added it into the R2 release.
It's also been written that there will be some new tools available for gathering data and managing workloads for people when it comes to server consolidation. Is that correct?
Oberoi: Yes, that's right. In any given release we are making advancements and making the whole platform easier to manage and deploy.
So back in the [SQL Server 2008] release, we released a set of capabilities around policy-based management, where you were defining what the policy was to manage your different database instances as opposed to going into the tool and doing everything manually.
So we've sort of taken that one step further in this release to what we call application and multi-server management. So it's not just doing that for a single server or instance, but it's really about being able to create a map or view of all the licenses across a specific network, and then manage those in a consistent way.
So then instead of having to deal with an individual database or database instance one by one, you can now do this notion of multi-server management, where you can create these policies and publish them out more broadly, and then you can monitor your instances more broadly as well. Then across this database farm that [you] have, you know which ones you have at capacity, which ones you need to add more storage space to, which ones are running out of memory. And then you can do it in a more consolidated view as opposed to doing it as a one off thing.
So that capability essentially helps with a consolidation scenario because [what you have to begin with] is a more centralized view of your system across an organization.
Check out more preview coverage of PASS Summit 2009.