On Thu, 21 Jan 1999, Carlos Maltzahn wrote:
> Isn't most of the UFS overhead inode management and the fact that you use
> individual files? With mmap you have _one_ inode that is constantly used
> and therefore cached. You don't have to worry about file system level
> fragmentation. So even though a Squid FS would be nice, I disagree with
> you that it is necessary to significantly reduce disk I/O.
By SquidFS most people mean a few huge database style files or raw
partitions managed internally by Squid, not a yet another UFS. Virtually no
inode overhead and open/close calls.
> I'd be very interested in finding out more about this. Who is working on
> squid-FS?
AFAIK, there are a lot of ideas and promises, but no steady development at
the moment. After doing Squid benchmarking, even we started thinking about
actually spending time on implementing SquidFS. That's a big progress! :)
Ideally, before implementing SquidFS, one needs to convert all store
modules to a nice API so many FS could be supported without #ifdefs and
other crap.
Alex.
Received on Tue Jul 29 2003 - 13:15:55 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:12:02 MST