In a paper I developed some theory; some of the applications require extensive computations that are not part of the paper. I wrote a Mathematica notebook. I want to publish a PDF and .nb version somewhere to refer to from the paper. arXiv.org seems a good choice, but they won't accept the .nb file. I do not want to put this through peer review. Where to publish it?
You should upload the notebook along with the sources of the actual article to the arXiv, of course.
The official (but rather hard to find) advice on this from the arXiv is to place your code in a directory called /aux/. (This is problematic for windows users.)
You can see an example of this in my recent paper on the extended Haagerup subfactor. A footnote in the text of the article explains that the code is available along with the source download. It would also be appropriate to use the comments field in an arxiv submission to explain that source code is available in the /aux/ directory.
The problem with storing the files on your personal website is that it is not clear how long that website will remain valid. What happens if you move to a new institution and your old account gets deactivated? The arXiv solution pointed out by Scott Morrison is better.
Here is another possibility: Some journals that publish an electronic version in addition to (or instead of) a paper version will allow you to include ancillary files with the electronic version of an article. If you'll excuse the self-promotion, see this journal's website, for example.