White Paper: Implementing post-deployment changes to Analysis Services databases

By Steve Pontello

When you create an application in PerformancePoint Planning, PerformancePoint Planning Server creates and maintains a SQL Server Analysis Services (SSAS) database for each model site in the application. Sometimes, you might find that you need to make changes to the SSAS databases. This can occur in several situations. For example, one customer might want to add named sets to the database, or incorporate formatting changes to comply with reporting requirements. Alternatively, a customer might want to improve query performance by designing and creating aggregations.

The following white paper by Steve Pontello describes a method that you can use to make post-deployment modifications to the SSAS databases. The solution implements a single Data Manipulation Language (DML) trigger that is applied to the DeployInfo table of an application database that requires post-deployment modifications. The trigger definition relies on the use of a standardized naming convention to create a SQL Server Agent job that applies the modifications. The white paper includes relevant examples and sample code.

Click the following link to read the white paper.

White paper: Implementing post-deployment changes to an Analysis Services database

Steve Pontello is an Architect with BI Global Practices at Microsoft. Steve has more than 17 years of experience in systems development, ranging from management to design. For the past 11 years, Steve has worked exclusively on the development and implementation of Business Intelligence systems for medium to large organizations. Since joining Microsoft in 2006, he has been helping customers use Microsoft Business Intelligence software to build reliable, accurate, and scalable OLAP solutions.

You can contact Steve with questions at stevepon@microsoft.com, or call his office at 208.319.2230.

Top of Page

Share Facebook Facebook Twitter Twitter Email Email

Was this information helpful?

Great! Any other feedback?

How can we improve it?

Thank you for your feedback!