Jan Harkes <jaharkes_at_cs.cmu.edu> writes:

> We used a modified Andrew benchmark to test by untarring and building a
> Coda source tree, locally, in Coda using the Coda kernel module, and
> using the Linux 9pfs kernel module with and without caching.
>
> - on local filesystem,               1m 59s
> - in Coda using Coda kernel module,  2m  9s
> - in Coda using 9p2000,             67m 39s
> - in Coda using 9p2000 with caching  3m 51s
> - in Coda using 9p2000.u,           70m 23s
> - in Coda using 9p2000.L,           ~5m

It would be good to test this on some other system than Linux, since the
point is to support systems that have a FUSE implementation but not a
coda kernel one, and to start doing this early, to guard against
linux-only things.  But its good to see some numbers.

I wonder how much of this is about when there is sync to disk.

What about using a fuse p9 client on linux, instead of the kernel
module?  Does that have a notion of caching?  Is that read caching, or
also write caching?

Do you understand where the time is going?  A factor of 2 or 3 does not
seem surprising, but 30x is.