Depending on how dynamic the content is, you can parse the directory into an "index.html" file in a once-in-a-while process. The "slow" is because the Web Server (Apache) is processing the
directory contents on the fly, instead of attempting to simply dump the contents of a similarly sized file.
I would probably write this as a nightly Perl task - customized. On the other hand, you could also do something like this:
+++
rm htdocs/index.html
wget -o tmp.html
http://mp3.thirtythreeandathird.net/
mv tmp.html htdocs/index.html
+++
It would give you the same result, but will not have to wait for Apache's dynamic processing every time.