Re: Efficiency issues/thoughts

Jason Rennie <[email protected]>
Newsgroups gmane.mail.ifile.general
Message-ID <[email protected]>
[email protected] said:
> I've trained ifile on my entire backlog of mail; as a result, .idata
> is nearly 840 meg.

840k I hope?  I don't think an .idata file could hit 840 meg :)  My 70M of
mail yields a 508k .idata file.  Also, that may shrink some as you use
ifile.  Fewer words expire when e-mails are processed by category (like
the knowledge_base.mh script does).

[email protected] said:
> One way of speeding up the invocation would be to have ifile run as a
> daemon, and have command-line invocations just speak to the running
> daemon.   

This has been discussed in the past, but hasn't been implemented.  It
would be some work and would add complexity to ifile.  Certainly worth
pursuing if we determine that it is necessary, but there's at least one
thing to try that's much easier that might satisfy you.

If you are currently using the --query-insert option (-Q) to filter new
messages, try using the --query option (-q) and build .idata (e.g. run
knowledge_base.mh via cron) once a day to pick up new messages.  This will
also speed up refiles---there's no need to hook them into ifile since
they'll be picked up with the daily .idata build.

The --query-insert option reads and writes the .idata file.  The read is 
quite fast, but the write slows things down a bit.  The --query option 
just reads the parts of .idata that it needs.

Jason D. M. Rennie
MIT AI Lab
[email protected]
(617) 253-5339
http://www.ai.mit.edu/~jrennie/
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.