Garrett Wollman wrote: > <<On Thu, 26 May 2005 22:03:39 -0600, Scott Long <scottl_at_samsco.org> said: > > >>is just as much work as porting GFS, if not more, since UFS/FFS is >>closely tied to the buffer cache and block layers on BSD, and divorcing >>probably would be quite difficult. > > > For multiple writers, probably so. For the single-writer case, I > don't think so, since the readers can mount a snapshot while the > writer mounts the read-write view. Well, having a writer is pointless if the readers are stuck on a snapshot. Also, I'm not exactly sure how dirty buffers get flushed to disk in the case of a snapshot. Recall that the snapshot file only save the deltas of the old data, and flushing those deltas to disk is still under the control of the VM/buffer/cache system. You'd probably still have to have some sort of inter-computer synchronization system in place. > > All this would be a lot easier if we had a storage manager more like > ZFS's. > Well, ZFS doesn't really exist yet, so it's hard to draw a comparison ;-) ScottReceived on Fri May 27 2005 - 16:16:35 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:35 UTC