> Tux2's phase tree algorithm works almost entirely in cache and is intimately
> coupled to the buffer cache system. A third metatree is added, to allow
> filesystem updating to continue without pause while the second tree is commited
> to disk, eventually replacing the first metatree using the abovementioned
> atomic write.
This sounds very very close to FFS+softupdates on FreeBSD boxes where there
is DAG representation of write-orderings and dependencies of various metadata
chunks. The orderings can be delayed to reduced the number of explict disk writes,
etc... just like FSes mounted as async and has similar performance to it and
journaled FSes too. But it maintains FS consistency at all times.
It's very close to what I expect from a journaled FS.
> Daniel
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sun Jul 23 2000 - 21:00:11 EST