From: Chuck Larrieu (chuck@xxxxxxxxxxxxx)
Date: Thu Dec 21 2000 - 22:08:34 GMT-3
Interesting question.
Several CCIE's have advised that one should not try to read too much into
questions. On the surface, I would choose some default manner in which the
four types were getting equal treatment, and therefore 25% each. If that
meant leaving the default setting, then so be it.
If you were to ask the proctor, I have been told that the best approach is
to provide the proctor with the alternatives you come up with, and why -
i.e. what each would accomplish. The proctor would then tell you one or the
other.
I'm reminded a bit of that Monty Python bit in Search for the Holy Grail -
at the Chasm of Nicht. How far can a swallow fly? Is that an African swallow
or a European swallow. I don't know ( and if you haven't seen it, never
mind )
I would bet that in a Lab scenario, you would get something that would
require thinking out the calculating of the percentages. ;->
Chuck
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of Ron
Brewer
Sent: Thursday, December 21, 2000 5:21 AM
To: ccielab@groupstudy.com
Subject: Custom queuing question
When you see a question on a practice lab state
something like, "Make ftp, telnet, snmp, and
everything else share 25% of the bandwidth each," is
it ok to just go with the default byte count of 1500
and let each queue use that amount? Or should you
take into consideration the bandwidth of the line, for
example 64K, and divide by 8 to get 8000 bytes per
second to get a queue length of 2000 bytes per queue
(based on 1 second intervals). Or do you go further
and base the intervals on 125 msec and divide the 8000
bytes by 8 to get 1000 bytes which would leave you
with a queue length of 250 bytes per queue.
Just curious as to how detailed custom queuing should
be with a statement like the one above.
Thanks,
Ron
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 08:26:09 GMT-3