Time Gentlemen Please.

Making predictions over time is a very common use of Data Mining.  Whether you want to know what the FX rate of EUR to GBP will look like over the next 3 months or you want to know what skills to train the service desk personnel in based on current support call trends, time series data mining is going to help you a lot.

In this presentation I want to look at what MS is offering in SQL Server 2008 for Time Series Analysis.  We will look at the different algorithms available, when to use them and how.  We will also have a look at the effects of changing some of the parameters in the Mining model like AUTO_DETECT_PERIODICITY and MISSING_VALUE_SUBSTITUTION.  We will also spend a small amount of time querying the Time series model to get a handle on the possibilities outside of BIDs.

Data Mining is often seen as super complex and off-putting but I am going to try and change that view and make this session fun.

Creating an SSIS, SSAS and SSRS monitoring solution with SSIS, SSAS and SSRS

The Microsoft BI stack is now mainstream, and mission-critical implementations can be found in the largest businesses worldwide. However, the kind of monitoring tools that we take for granted with the relational engine simply don't exist yet for Integration Services, Analysis Services and Reporting Services. How do we know if our Integration Services packages are taking longer and longer to run every day, who's running that big query that's slowing Analysis Services down, or if the boss ever looks at that Reporting Services report you took three weeks to create for him? Luckily the solution is close at hand: we can use SSIS, SSAS and SSRS to build an application for monitoring SSIS, SSAS and SSRS. This session will describe how you can do this.