How to work with millions of files when xargs does not work

In some cluster I need to enter in some directory and list and select some files. The problem is that probably there are even millions of very small files. If I do

ls -l

It is very inefficient. But if I try a supposed to be better alternative like

find . -name "*.mol2" | xargs ls

It takes minutes and I do not get any answer ...

I wonder if there are better/faster methods for situations like this one. Of course I could tell the person who generated all these files something, but thi开发者_如何学Gos is out of the scope of the question. Thanks

The obvious would be:

find . -name "*.mol2"

But with millions of files it will still take time - at least the first time. The second time the dir will hopefully be cached.

For the outside scope: Create subdirs maybe named as substrings, so foobarbaz123.mpl2 will go into foo/bar/baz/foobarbaz123.mpl2

How about ls -al *.mol2 .*mol2?