Open Source Databases

All indications are the next significant growth segment for open source software will be in databases. This follows the trend of open source operating systems (Open Solaris and Linux).

Two open source databases of note are Hadoop and MySQL.
Hadoop is not for everyone. It is a very powerful open source software focused on highly scalable distributed computing. It implements the MapReduce distributing computing metaphor in use at some very large computer powerhouses. In general, I don’t believe it will be of immediate use to the average enterprise, it is for the big guys with high end problems.  My recommendation is that all CTOs at least download it at home and try it out just for familiarity (I’m running Hadoop on my home systems now so I can kick the tires and will be writing more about it in coming posts). But I don’t recommend every enterprise everywhere adopt it.
MySQL, on the other hand, should be of interest to any enterprise, big or small.  I’m a MySQL user and really enjoy it. I’m not alone in that regard. MySQL has over 11 million installations and is the driver behind most major web technologies today. It is the database for a variety of development platforms including popular software bundles like LAMP, BAMP, MAMP, SAMP, and WAMP Popular websites using MySQL include Facebook, Zappos, Cox Communications, NASA, Flickr, Wikipedia, Google and YouTube. The Obama campaign was also run with technology based on MySQL.
How much does MySQL cost? It is available for free under the GNU General Public License, which is a great way to get and use software. Enterprises like support, and support costs money. How much will support for MySQL cost? I don’t know, since I’ve never required enterprise support, but from what I understand the cost is about 20% the cost of support for proprietary systems. MySQL lacks some features of the higher end high cost enterprise systems, but at such a reduced cost it will increasingly be the alternative of choice for solutions that don’t require every feature of a massive ERP-type capability.
Additionally, MySQL can result in better reliability and more uptime, which should also be factored into your TCO calculations.
In your engineering trades you will likely find that MySQL will run more calculations per second on lower cost hardware, and, adminstration/services costs are also significantly lower.
So, those are cost reasons to move to MySQL. Other, perhaps more important reasons include:
  • It is easy to learn and easy to administrate
  • It helps prevent vendor lock-in and companies that will try to place you over the barrel
  • Security is built in and in my opinion there will continue to be fewer vulnerabilities in MySQL because of its open source model
  • There are very large numbers of developers supporting MySQL, so it is easy to find highly qualified developers and administrators.
The big providers like Oracle, Sybase, Microsoft and IBM continue to roll out improvements and advanced features and and they have powerful capabilities that will likely be with us for a long long time. But my recommendation is that every CTO check out MySQL and use it everywhere you can. It will help you deliver more functionality faster and for a much more economical cost.
Comments?
About BobGourley

Bob Gourley is the editor of CTOvision.com and is the founder and Chief Technology Officer (CTO) of Crucial Point LLC, a technology research and advisory firm. Bob was named one of the top 25 most influential CTOs in the globe by Infoworld in 2007, and selected for AFCEAs award for meritorious service to the intelligence community in 2008. He was named by Washingtonian as one of DC’s “Tech Titans” in 2009. Bob was named one of the “Top 25 Most Fascinating Communicators in Government IT” by the Gov2.0 community GovFresh.

Post comment as twitter logo facebook logo
Sort: Newest | Oldest

Boy did I get something really wrong here. I'm glad I suggested all download Hadoop, but I definitely should have seen the fact that Hadoop is becoming useful almost everywhere. It is hard to learn and implement, but folks who have data and need to make sense of it should be working to get it up and running now, and that seems to be everyone.