[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: How stable is ext3fs?



Hans Wilmer wrote:
> 
> Can it be that ext3 is considerably faster than ext2 in accessing
> large amounts of files in a directory? I've collected about 32000
> mails (i. e. files because I'm using maildir) from the list in my
> debian-users folder. When switching into that folder, mutt reads the
> headers (or whatever) from the mails; when switching off the folder,
> it updates the mails.
> 
> With ext2, this took quite long, but with ext3 it seems to be much
> faster :) How comes?
> 

heh.

Mail produces "slow-growth" files.  Which means that their blocks
are sprinkled all over the disk.   If you're adding a few k per hour to
a file, the fs just about never manages to allocate the blocks
contiguously.  A while back, I had a six-month-old multi-megabyte
mailbox which had precisely *zero* contiguous blocks.  It was 100%
fragmented!

If you copy one of these files onto a new file, the new file is
magically unfragmented, and access to it is much faster.

In your case, it seems that you have many small files in the
same directory.  I bet the directory itself is fragmented, for
the same reason - slow growth.  Did you copy the files at
some stage?   That would unfrag the directory.   And the inode
table.

For the above reasons, I partition my machines with all partitions
the same size, and keep one free.  For the monthly theraputic
copy-all-files-and-switch-mountpoints speedup.

It's all a bit sad, really.

-





[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]