Why UNIX I/O is so slow (was VAX vs SUN 4 performance)

[Larry McVoy]

In article <8124 at brl-smoke.ARPA> gwyn at brl.arpa (Doug Gwyn (VLD/VMB) <gwyn>) writes:
>Such considerations should lead to the conclusion that each type of
>filesystem may need its own access algorithms (perhaps in an I/O
>processor).  This is easy to arrange via the File System Switch.

Do the wizards have a preference (based on logic, not religion, one presumes)
between the file system switch and the vnode method of virtualizing file
systems?  Anyone looked into both?
-- 
Larry McVoy	laidbak!lm at sun.com	1-800-LAI-UNIX x286