When I got back from SQL Rally 2012 a few weeks ago in Dallas, I shoved the notes into my laptop bag and took off for vacation. I still want to share a few of the highlights of my trip. These were some excellent speakers and I hope to see them again at future events.
Sessions:
Tuning Analysis Services Performance by John Welch
Presentation
John did an excellent job of explaining the basics of the processing options in SSAS. What was interesting was his experience splitting up jobs to increase performance. Process Full will process all the dimensions and facts of a cube, but is usually the slowest option. By splitting up the process job into Process Dimensions, Process Data and then Process Index there may be faster processing. Also covered in the discussion was the SSAS memory allocation and how the ‘High’ memory limit is really just suggestion to SSAS, not a real limit. John also recommended the book ‘Getting Naked’ a consulting model used and recommended by Slalom Consulting.
The Report Part Library by Jessica Moss
Presentation
Jessica did an excellent demonstration of publishing and using the Report Parts feature of SSRS to build a reusable library. Jessica is always a great speaker and is co-author of the book I’m reading now ‘Professional SQL Server 2012 Integration Services’.
Introduction to Data Quality Services by Tim Mitchell
Presentation
Tim explained the different approaches from data cleansing to data matching. Data cleansing will process the data against a known set of entities and domain rules, such as address validation. Data matching groups data like fuzzy grouping to allow further processing. Good demo and definitely something to go try at home.
Building a Write-Back Application with Analysis Services by Bryan Smith
Presentation
This is a task that I have seen requested by clients so it was very cool to see in action. Bryan explained and demonstrated writing back to a cube by defining a write back partition. One caveat is that you can only write back to sums, not to count type members. At high levels, you need to define an allocation method. An interesting note was writing back to dimensions to write back to attributes.
Data Mining (It’s not the size of your data – it’s what you do with it) by Adam Jorgensen and Devin Knight
Presentation
This was entirely a demonstration session about the new data mining tools available from http://www.sqlserverdatamining.com/ssdm/. They had volunteers come up to the podium and walk through simple data mining exercises to show how easy the tools have become. These guys are always entertaining and informative.
Business Intelligence SQL Azure: SSIS & SSRS Focus by Jen Stirrup
Presentation
Jen first explained how to get a free Azure trial and walked through the simple setup process. She gave a fun demo as she loaded data about the ‘Jedi’ religion in the UK and showed how ‘Jedi Knight’ ended up becoming the 4th largest religion. Jen discussed data table requirements and development options for moving SSRS and SSIS to Azure. Personally, I think every session would be better with an English accent!
Digging into SSRS 2012 with Sharepoint 2010 by Adam Saxton
Presentation
Adam covered the new options available such as Data Alerts, PowerView and SharePoint shared service. Adam did a good job of covering inter-version support between SQL and SharePoint. He also covered some setup gotchas.
Making Heads or Tails of Analysis Services Storage by Greg Galloway
Presentation
Greg’s talk was a very in depth coverage of the storage options. What I learned is that I need to learn more about these options in SQL 2012. Greg recommended Teo Lachev’s book Applied Analysis Services:Tabular Model and the Microsoft SQL Server 2012 Analysis Services: The BISM Tabular Model by Russo, Ferrari and Webb.
By the way, Greg is the coordinator for BIDS Helper which has been updated to 1.6 with support for SQL 2012. Go get it!
Advanced DDL Constraints by Joe Celko
Presentation
This was a community pick and my favorite talk because Joe Celko is entertaining and very informative. Mr. Celko started by explaining the need for DBA’s and developers to use the SQL standards. The demonstration included check constraints to validate data at the table level. He started an interesting discussion when he said that views should not be prefixed with vw_ unless it is for a view of Volkswagon cars. He encouraged everyone to take time to read the ISO-11179 rules for the data element names. Read the standards here in your spare time: http://standards.iso.org/
Thanks to all the speakers for their time and effort. Also many thanks to the sponsors who make these events possible.
I hope some of these speakers will submit for SQL Saturday #125 in Oklahoma City, OK on August 25th, 2012. The deadline for submissions is June 30th.
If you are keeping track of the books from these authors, here is Joe Celko’s book: