Hi, At Mon, 14 Sep 2009 12:27:31 +0200, Reinoud Zandijk wrote:
> > I think it should be considered along with ext3 approach. > > What is is the ext3 approach? What is different compared to ext2? And is it > really such a big improvement? Will it complicate directory reading/writing > more? ext3 approach is a hashed index for file names, stored on disk. So it always costs constant time to lookup the block filename stored. The point is that this hashed index extension preserve backward compatibility, which means that code which does not know about hashed index still be able to read the directory entry as usual. The improvement of this was significant compared with ext2. I re-put up the following benchmark results. http://plaza18.mbn.or.jp/~moriban/linux/img/FileSystemBenchmarkResults-01-03.png # There may be more reliable benchmark result, but I cound't find it. There are some documents for further information if you are interested in. - A Directory Index for Ext2 http://ols.fedoraproject.org/OLS/Reprints-2002/phillips-reprint.pdf - [RFC] Ext2 Directory Index - File Structure http://lwn.net/2001/0412/a/index-format.php3 - [rfc] Near-constant time directory index for Ext2 http://search.luky.org/linux-kernel.2001/msg00117.html thanks, regards, -- Jiro SEKIBA <[email protected]> _______________________________________________ users mailing list [email protected] https://www.nilfs.org/mailman/listinfo/users
