Client - Server(s?)

[Scott Emery <emery@xxxxxxx>]

Hullo Jens,
	I have enjoyed using fio, I find it to be easy to set up and
run. 

	I have a question about the direction of client/server, which
revolves around a situation I would like to benchmark.  I deal with
a couple of clustered filesystems and would like to get aggregate
performance values based on I/O from multiple filesystem clients.
To that end I would like to drive a test from a single "location" to
multiple filesystem clients. 

	My understanding (?) of fio client/server semantics is that
one client drives no more than one server.  With this understanding
I would set up a fio server on each filesystem client that I want  to
test and then run fio clients from my test location in parallel to
each fio server/filesystem client.  I would then aggregate the
statistics by hand.   This is certainly a huge step up from scripting
something myself by hand. 

	It would be convenient if I could have one client/set of clients
work together to drive the fio servers and aggregate the statistics
for me.  Is this where fio is going?

Scott Emery
emery@xxxxxxx
--
To unsubscribe from this list: send the line "unsubscribe fio" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html