HAMMER update 06-Feb-2008

[Ivan Voras]

Matthew Dillon wrote:

    * Get rid of the A-list's (the recursive bitmap radix tree allocation
      model).  Poof, gone.  All of them, gone.  Have no random block
      allocation model at all.
    * Get rid of the per-cluster B-Tree's.  Instead have just one global
      B-Tree for the entire filesystem, able to access any record anywhere
      in the filesystem (currently a per-cluster B-Tree can only access
      information within that cluster).
    * Implement the filesystem as one big huge circular FIFO, pretty much
      laid down linearly on disk, with a B-Tree to locate and access data.
How will this affect parallel IO (reads, but especially writes)? Would 
having such a global structure serialize it? (I'm assuming, possibly 
wrongly, that having trees per-cluster allowed you to lock individual 
clusters).



Attachment:
signature.asc
-------------- next part --------------
A non-text attachment was scrubbed...
Name: pgp00000.pgp
Type: application/octet-stream
Size: 250 bytes
Desc: "Description: OpenPGP digital signature"
URL: <http://lists.dragonflybsd.org/pipermail/kernel/attachments/20080206/92bde83d/attachment-0020.obj>