MS SQL Server and Tic Toc and why I changed my mind that it’s a good idea

I am not sure MS has got it right with the release cycle for SQL server. While this debate has happened before (I was then excited that there will be newer versions of SQL available sooner) now I am older and wiser. The reason I don’t think this approach will work in the long run is because I made the below comparison.

If you have worked in databases long enough you know data is money. Where do we store money? In Banks and Bank Vaults. Now imagine you start a Bank and buy the latest and greatest version of the safe from the provider. The safe is such a big and important investment since it ties into almost all aspects of the bank such as the architecture of the building, the processes followed by employees and the overall quality of the bank. After all this effort imagine if after 2 years the provider come to you with a newer and better version of the Safe. Now naturally most people would not bother to upgrade as long as the safe is complete and does everything the bank wants from it. But what if the safe had a few minor bugs and defects that while not mission critical still prevents you from fully utilizing the safe to its fullest potential. Say for example the safe has a feature that uses facial recognition but can be easily fooled using a photo. So rather than use the feature we stick the number combination. This is what happened with SQL 2012. Columnstore indexes, AlwaysOn while good hadn’t matured enough. They matured in SQL 2014 but by then it’s too late.

This is what’s happened as far as I can see. So many client upgraded to 2012 that there doesn’t seem to be a market for 2014? But does 2012 do everything the client wants, no not really. I guess what I am saying is there needs to more time in between release to allow dev teams at MS to Mature and work out all the bugs in the new features. Else SQL upgrades will be a weird game of catchup and most client are not going to play that game long before switching.

Another reason that prompted me to make this comparison is because I did some BI on the type of trainings/ Consulting projects my clients asked and for most part till SQL 2012 the trend has been to upskill employees with the latest and greatest even if the company still used 2005 mainly because they wanted to be able to say to clients “our employees are up to date with the latest technologies” but with SQL 2014 they still haven’t gotten over the hang over from the SQL 2012 training to invest in 2014 and most likely they realize there is no point in upskilling to 2014 coz 2016 is just round the corner.

While I hope I am wrong, I get a bad feeling that the release cycle is going to spell trouble for MS SQL Server as a dependable stable platform that makes commercial sense.

A classic case of ” too soon of a good thing”