The Decline of SQL Server

For many years in my career, SQL Server was the runner up. Both Oracle and DB2 were in much wider use and more popular among all sizes of companies. Serious applications used one of those platforms, according to many technologists, the DBAs on those platforms made more money, and most companies were willing to pay for the software. That changed, and over the years I’ve seen SQL Server grow close to Oracle, while DB2 has declined (and MySQL has grown). Here’s the 2014 chart from Gartner, on of all places, the MySQL site.

I see other charts and reports, showing various percentages, so it’s hard to be sure where things stand. Many of them agree that for total market share, it’s still Oracle and DB2 (mostly on larger hardware), but SQL Server rounds out the big 3. However is that the case for jobs and opportunities? I read a piece from Thomas Kejser that pointed out that the popularity of SQL Server may be declining (there’s also an update).

I see lots of jobs posted, and so many companies using SQL Server. SQL Saturdays continue to have great attendance, so I wanted to ask you this week:

Is SQL Server being used less, or are other platforms being implemented, in your experience?

Are you seeing a trend away from new, or additional, SQL Servers in your organization? I haven’t heard of any large trends, though I certainly do know that some organizations are considering other relational platforms. I think that’s healthy and expected. The technology for managing transactions is well known, and I think that all three major platforms, as well as MySQL and PostgreSQL are maturing to the point that any of them can handle the load for most applications out there.

I think it’s only natural that other platforms will be used to solve problems that they are well suited for. However just as I don’t think SQL Server will grow to dominate the RDBMS space, I don’t think that NoSQL platforms will dominate over RDBMSes in many applications.

Steve Jones

The Voice of the DBA Podcast

Listen to the MP3 Audio ( 2.9MB) podcast or subscribe to the feed at iTunes and LibSyn. 

About way0utwest

Editor, SQLServerCentral
This entry was posted in Editorial and tagged , . Bookmark the permalink.

2 Responses to The Decline of SQL Server

  1. SqlChow says:

    While organizations try and make sense of data that is worth decades, more data is being generated by their sources (web, application, in-house, etc.), and traditional databases/data warehouses have to take up this load and be the data repositories for more short-term analysis (fraud detection, spend analysis, financial year trending) or predictive analysis. I believe this makes it look, at least externally, that the traditional data-sources are dead. But, they are not!

    One thing to remember is that traditional data-sources like SQLServer have been around for a while and knowledge reserves of people who work in these technologies also mature along with the platform. As the knowledge reserves mature, we see more accurate solutions being provided in-house without the need for looking towards external sources of help. Also, as there is more maturity in a product with strong community support you tend to see a lot of high-quality articles being published. Again, this could be perceived as lack of interest but, it is just that they have multiple avenues to search and arrive at a solution rather than look at only one place.

    In my opinion: Yes, the role traditional data-sources play has changed and no there is no decline in interest (It is the thriving community that makes it possible for a beginner to find his feet fast) but, rather it is the opposite.


    • I have to agree with SqlChow that the high quality of articles and practices already in place make the need for searching the web for answers less and less necessary. In just starting a new position, I already have built a tool set of queries and practices that I immediately put into place to tell me where the current systems are configured correctly or not. I refer to the Maintenance Solution by Ola Hallengren, Chris Shaw’s UtilityDB, Glenn Berry’s SQL Server Installation Configurations, and Brent Ozar’s First Response Kit. Between these basics I have already done in one or two weeks what used to take me six months to a year to get the databases is an almost completely automated capacity so I can then look into the real problems, performance. And the best part is they update their material for every version, 2005, 2008, 2012, and 2014.


Comments are closed.