On Wed, 15 Aug 2001, Jihong A. Kim wrote:
> Q1)
> Is it okay to distribute IP address unevenly among clients or servers and
> run the polymix test? ( For instance, I would give one client machine 300
> IP addresses and the other 500 IP address, and run the test. )
It is OK to run tests with any IP address distribution. You just
cannot call the resulting workload a PolyMix-3 because it violates
PolyMix-3 rules. Whether your results will be any different compared
to a compliant PolyMix setup is impossible to say in general. In most
cases, I would not expect any difference.
> Q2)
> Do client machines and server machines have to move together at the
> same time from one phase to the next throughout, the Polymix-3 test?
Clients and servers might get out of sync. This is a bug. Fortunately,
this bug should not affect the traffic seen by the device under test
as long as clients stay in sync. Server-side phase schedule is not
used for anything in current Polygraph versions.
If you use --idle_tout on the server side (and you should), the server
will terminate after the client goes away. You can also send the
server process and INT signal to stop it nicely.
We may have fixed the bug in Polygraph 2.7. However, since we do not
have an easy-to-reproduce test case, I am not sure.
HTH,
Alex.
This archive was generated by hypermail 2b29 : Mon Feb 06 2006 - 12:00:19 MST