There's been a lot of buzz lately on open source DBs. This topic comes up now and then when one of the vendors talks a SQL guy with a pen into writing something on the usefullness of open source. I guess it works too, because here I am putting my 2-cents in on it keeping the topic alive. All the same though, the topic at hand remains whether or not open source is ready for the enterprise. I'm just going to go ah There’s been a lot of buzz lately on open source DBs. This topic comes up now and then when one of the vendors talks a SQL guy with a pen into writing something on the usefullness of open source. I guess it works too, because here I am putting my 2-cents in on it keeping the topic alive. All the same though, the topic at hand remains whether or not open source is ready for the enterprise. I’m just going to go ahead and say it… you cannot run an enterprise on open source DBs. They just aren’t there yet. So far, none of the open source DBs can handle XML, nor can they do any kind of real BI whatsoever. Quite often, as is the case with MySQL, they don’t even have a complete coding set. Ingres is the only one so far that I’ve seen that could compete with closed source, but that’s only because they used to be closed, and now CA has opened it up. Open source simply doesn’t have the development effort that closed source does. A group of guys in their garage can never match the money that closed source implementations have. Companies like Oracle, IBM, and MS put millions into their products every year, and that kind of funding will always be richer in features, and stability. That does make me wonder though… with this new MySQL worm that’s going around, will these open source implementations be able to keep up with security patches once they get popular enough for the hackers to care about them? Keeping up with attacks is a fulltime job, and my guess is they’ll have to bring on more help just to release patches. I just wonder how that’s going to work out in the end.Sure, I give the closed source vendors a hard time about being greedy and soaking us for everything they can, but in the end, you get what you pay for. I’ve heard a lot of talk in the newsgroups by DBAs who swear by open source DBs, and they may be able to run their business just fine, but you’ll never run an enterprise solely on open source. Try to get SAP up and running on MySQL, or ADP, or BAAN, or any of the other enterprise-level packages. Even your monitoring solution won’t work with any of the open source DBs. That means that HP Openview, or Quest Foglight, Tivoli, BMC, or Concord won’t run on open source DBs. That’s just the tip of the iceberg in the discussion on open source DBs. I could go on and on about the problems with encryption support, auditing, and other issues surrounding compliance. I’ve written before that their management interfaces aren’t rich enough to support enterprise-level implementations, but vendors are only just now beginning to look at writing products for them. Rich code editors, management tools, and monitoring and alerting tools are not only the bread and butter for a lot of vendors, but they’re also what make supporting a large database enterprise possible. As it stands, whatever you save in DB licensing from the big vendors, you will make up for in salary costs for more DBAs to manage it. It simply doesn’t make sense.Is open source useless? Absolutely not. It definitely has its place in smaller home-grown apps that require less robust implementation, and less security. These DBs can be quite large though. Yahoo! is running portions of their website on MySQL, and Suzuki is running MySQL as well. There are several high-profile companies running open source to lend credibility to the project. I just don’t think that they have the features it takes to put into a high-level enterprise application. As was put so succinctly in one of the SQL newsletters… you can’t make a career out of open source. Just try to find someone who specializes in MySQL. It’s practically impossible. Everyone specializes in one of the Big 4, and does open source as more of a point of curiosity. Sure, some DBAs have gotten quite good at them, but I’ve never met one who’s ready to trade in all of his Oracle or SQL Server DBs for anything open source. All that being said, I’m a fan of open source. I still think the Big 4 are too greedy, and we need open source to help even the playing field. The new mantra of DBAs should be open source for your everyday needs, Big 4 for the dress balls.That’s all I have to say about that.Comments? Databases