From: Michael Snyder (msnyder@xxxxxxxxxxxxxxxxxxxxxx)
Date: Sun Aug 11 2002 - 14:24:39 GMT-3
I stand corrected. So if it does run out of tokens and queues the
packet, that queued packet can't be sent till the next time interval
anyway. Thus the limiting function is still preserved.
I was thinking it was just a general realtime interface queue. But if
that was the case, then it would defeat the whole purpose of GTS.
Thanks,
This email thread was time well spent for me.
Michael
P.S. Did anyone else catch the inconstancy of the feature name GTS, and
it's limiting functions? To me, CQ is a shaper, and GTS is a policer.
-----Original Message-----
From: Balaji Siva [mailto:bsivasub@cisco.com]
Sent: Sunday, August 11, 2002 12:08 PM
To: msnyder@revolutioncomputer.com
Cc: ccielab@groupstudy.com
Subject: RE: QOS Question from Routopia LAB 3
mike,
why not ? shaping would allow only 10 % per time interval...it doesn't
matter which time interval i send it
..here is the question again
"Ping traffic can only utilize 10% of available bandwidth."
how are you violating the rule with shaping ?
Balaji
-----Original Message-----
From: Michael Snyder [mailto:msnyder@ldd.net]
Sent: Sunday, August 11, 2002 1:03 PM
To: 'Balaji Siva'
Cc: ccielab@groupstudy.com
Subject: RE: QOS Question from Routopia LAB 3
I forgot about the queueing if it ran out of tokens. I was only
thinking of the limiting function.
So if it will queue packets without a token, and send the packet anyway,
then my GTS answer won't work.
This archive was generated by hypermail 2.1.4 : Sat Sep 07 2002 - 19:48:23 GMT-3