Michael Peters wrote:
Michael Peters wrote:

But these benchmarks (http://www.debian-administration.org/articles/388) say the following:

For quick operations on large file tree, choose Ext3 or XFS. Benchmarks from other authors have supported the use of ReiserFS for operations on large number of small files. However, the present results on a tree comprising thousands of files of various size (10KB to 5MB) suggest than Ext3 or
  XFS may be more appropriate for real-world file server operations

But they both say don't use ext2 :)

This may be a tangent, but my understanding is that the only real difference between ext2 and ext3 is the journaling, which is related to safety in the event of unclean shutdown rather than everyday performance. If anything, in fact, ext3 performs a little worse than ext2 because of the requirement to keep the journal (which means more writes to the disk for updates). Otherwise, all the optimization features such as dir_index are, I think, available for ext2 as well as ext3. I have noticed that for SSD drives (e.g. the Asus Eee PC, which I have), people recommend using ext2, since it's less likely to result in the write fatigue that those drives experience over time (you only get so many writes). And for laptops, ext2 results in fewer io writes. Finally, I have noticed my iowait times go down since I moved from using ext3 to ext2 on the server (previously I always used ext3, but for a recent rebuild I switched to ext2 to see how it did).

Of course I may be wrong about all this, but my experience seems to favor ext2 over ext3, at least for performance. Since I back everything up on the server anyway (using RAID0, a necessity), I am more concerned with performance than unclean shutdowns. In any case the server is in a datacenter with UPS, so that is not so likely, though it did happen once and I didn't lose any data even then.

Neil

Reply via email to