I simply learned about http://exist-db.org/ on Start flossing weekly. (http://www.twit.tv/floss97) We've got the technology appears interesting. Does other people utilize it being produced programs? Do you know the talents and weak points of the system?

We are utilizing it at our organization, where we've around 100MB of xmls. It had been somewhat unstable in older versions, however the lately launched 1.4 version is extremely stable.

The correct answer is easy to get and employ. It features a simple Eclipse wordpress plugin, however for serious XQuery development, you will need to make use of an IDE, like Oxygen XML Editor.

Within the next version, it is also likely to supoprt XQuery debugging.

I've not attempted it, but you may also try a few of the commercial xml databases like Marklogic, Tamino, IBM DB2's pureXML.

The issue at this time is they all have different apis in Java. Many of them must have support for that standard API XQJ (JSR 225) each year approximately, which'll turn it into a lot simpler to change together.

I simply also went into exist-db lately.

My website stores (by at this time) roughly 500k 1k-sized xml files inside a s3 bucket and that we generate ~5k more every day. I have been searching at from vertica to hbase to terracotta, etc... and so on to keepOrissue these details. After I recognized that my xml had been formatted for any database also it required xpath/xquery queries -- as you would expect I had been estatic!

One important point I ought to note is the fact that All my information is WRITE ONCE and browse MANY. I've simply no intends to update the xml files after they have been in the machine. Basically do it'd simply be to merge documents together instead of change them.

I'm not sure much concerning how to scale this however for at this time this can be a HUGE response to among my sites HUGE problems. I really like miracle traffic bot!

We presently do not have this being produced but which will likely change inside the week.

For those who have a lot of data in xml that should be queried it is really an awesome match!

We lately began using eXist DB from Scala within our project. Some nice aspects were that people could simply keep XML representations we already had, which we're able to also store binary data. Coupled with Scala's built-in XML support I believe this can be a nice DB alternative. I authored a little much more about our benefits of using eXist over some options here.