It’s 2011 already and the first T-SQL Tuesday for the year has arrived.
I have come up with a list that I want covered by my team this year, but my main goals are two: Allowing me to sleep properly and investing time on number crunching.
What does “allow me to sleep” mean?
Last year we started our migration project of all production database servers to High Availability solutions. We shall continue this year and finalize it. On the same token, homogenize our environment so every single business unit runs on SQL Server 2008 R2. We really spent time figuring out processes that cannot run in SQL Server 2005 (e.g. change data capture) and I need to make sure it does not happen anymore.
On the storage side, there are 2 projects running in parallel. One is to allow growth on a scalable solution which is already in place (stop concerning about hitting volume limits), and the second one to reduce data file footprint. The first project is trivial as we can hookup additional trays and expand volumes. The second one is more complex as we need to identify tables that hold historical data, partition them (if applicable), phase out old data and archive without impacting production applications.
On the monitoring side, we are creating a new project to see if we shall stick with Management Data Warehouse or get a 3rd party tool to give us bird’s eye view of our systems at the DB level, define KPIs, and report based on them. We got proper HW and Service level monitoring in place already.
What does “focus on number crunching” mean?
This has 2 phases.
The first one is to work on a Master Metadata repository project and figure out relationships between different business entities. Then work with Business Owners and Data Analysts to figure out what should we report on, how far back should we do so, and what KPIs should be defined. This can allow us to continue building our Data Warehouse, extending ETLs, and allowing business unit A to correlate its data with Business Unit B.
This project is also tied with data file footprint reduction as we will be able to rollup information that shall be available on our DSS without the need of all the transactional data.
The second phase is analytics. This can be tied with data mining and statistics and will definitely involve some of the information we have stored in our cubes. We are still figuring this one out.
That is it for me on the technical side. On the managerial and personal? I would write about them on future blog posts.
What are your resolutions for 2011?