I’m really looking forward to attending PASS next week!
I’ll be presenting two separate talks this time around:
Maximize the performance of your packages with this set of best practices and data flow tuning techniques. We'll show you how to improve the operational efficiency of various Integration Services connectors, and take a look at how to efficiently load your data warehouse solutions.
This is basically a talk about how to get good performance from SSIS without having to invest too much time and effort. I’ll talk about “smart package design”, how to effectively introduce parallelism, and then jump into some of the easier tuning tweaks you can do in your packages.
Originally this talk was scheduled as one of the first sessions on Tuesday, which made it a very good segue for Thomas Kejser’s SQLCAT chalk talk on Tuning ETL and ELT. It looks like my session was moved to Thursday in the latest PASS schedule, which means I’ll be presenting after he does. I think both sessions are a good compliment to each other - While mine focuses on package design, Thomas takes a more scientific approach to performance tuning, with a pretty deep dive on SQL Server optimization. If you’re interested in performance, I highly recommend attending both sessions (there shouldn’t be much overlap).
My other session is a Chalk Talk that I’ll be co-presenting with John Welch.
Schema changes got you down? Tired of hand coding your SSIS packages for a ton of different data sources? Come see this session where we'll focus on the requirements for metadata driven ETL, and how you can build your own dynamic package generation framework using the SSIS object model. We'll talk about some of SSIS package frameworks currently used in-house at Microsoft, how to automate and test your packages, and about the open-source resources available online.
I’ll be at the SSIS booth and Ask the Experts areas throughout the conference, so be sure to come by and say hello.