-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 08/23/2015 06:48 PM, Simon Slavin wrote: > Paolo is using Linux which does not do read-ahead optimization like > some versions of Windows. Therefore if he really is using an SSD > then fragmentation is not an issue.
You are confusing things. The data structures used in a copy on write filesystem (as both btrfs and zfs are) are what gets fragmented. This is not like a traditional filesystem that will update existing data structures. It is more analoguous to garbage collection. I promise you that it really does become an issue, even with an SSD. When a file is across ten thousand fragments, latency and throughput suffer. > I have seen fragmentation make something take twice as long. You are talking about traditional fragmentation, not a degenerate state for copy on write based filesystems. > My suspicion here is that there's a design fault in ZFS. Unless zfs does automatic defragging, it will have exactly the same problems as btrfs. This is inherent in how the data structures are laid out, and that nothing is modified in place. Roger -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iEYEARECAAYFAlXakBgACgkQmOOfHg372QTepACffpEZ/tozxJKv0bKgZQ0D0wIL HqUAn3ES+b+xr/c8h7I/lqJs1zhQRVrg =+S02 -----END PGP SIGNATURE-----
