Reiterative parsing of static XML.

Question

Hello:
I've a quick question. I've a class which takes data from lots of static XML files (translations, for example). "Static" in the sense that they don't change during the execution of the application but may be updated by me or other programmer if needed, for maintenance. Many of these files need to be read very frequently (there are many concurrent users). To avoid making the XML parser work in such a reiterative way, I've thought about keeping the obtained data in some kind of "temporal media".
One idea which I had was to keep the data in form of arrays or maps, and then store them as temporal object files, using the class ObjectOutputStream. However this also means to serialize, check the existence of the instantiated class, etc. I guess... Do you believe that's a good mechanism or would you choose another way?
Thank you.

Answer 1

Dude, just use a cache. In memory. If you're worried about running out of heap space, make it a LRU cache. I'll bet you can find an LRU cache implementation on the web somewhere. As usual, I'll bet Apache Jakarta Commons has one; if not, there's probably something on sourceforge.
Saving them to disk and restoring them from disk would eat up as much time as parsing the XML from disk. (Well, maybe you could come up with an efficient compressed format that would take less time to parse, but parsing is probably not the dominant time sink -- that could just as easily be the raw IO -- so gains would be small and probably not worth bothering with.)