Otto,

I got your point. We have applied the logic to the LLQ but have not the same
logic when provisioning RSVP bandwidth. As you say ip rsvp bandwidth 64
would be the correct answer. Thanks for your input.
-- 
Vik Malhi ­ CCIE #13890, CCSI #31584
Senior Technical Instructor - IPexpert, Inc.

Telephone: +1.810.326.1444
Fax: +1.810.454.0130
Mailto: [email protected]


Join our free online support and peer group communities:
http://www.IPexpert.com/communities
IPexpert - The Global Leader in Self-Study, Classroom-Based, Video-On-Demand
and Audio Certification Training Tools for the Cisco CCIE R&S Lab, CCIE
Security Lab, CCIE Service Provider Lab , CCIE Voice Lab and CCIE Storage
Lab Certifications.








From: Otto Sanchez <[email protected]>
Date: Tue, 7 Jul 2009 15:34:41 -0430
To: Vik Malhi <[email protected]>, OSL Group
<[email protected]>
Subject: RE: [OSL | CCIE_Voice] Lab 10A BW Calculations

Thanks for you answers Vik,
 
1.- As you mentioned, the bandwidth is hardcoded to 40 kbps (in a g729
call), but this is for the initial rsvp request, when the media is still not
established yet between the endpoints (worst case call). Later, when this
happens the reservation is adjusted to the current call characteristics,
i.e. 24 kbps for a g729 20ms call (L3 BW), the remaining 16 kpbs would be
available for other calls,
 
I we use 40 Kbps per call for every call in the rsvp reservation command, we
could be allowing extra calls in our calculations. For example, 4 g729 calls
x 40 Kbps = 160 Kbps, in this case rsvp would allow more than 4 simultaneous
calls, if we again consider that an established call uses only 24kbps,
 
I think this is the reason the srnd recommends to make the calculation
considering only the Nth call as the worst case, the other N-1 calls can be
calculated as normal L3 calls,
 
BR,
 
Otto
 


From: Vik Malhi [mailto:[email protected]]
Sent: Martes, 07 de Julio de 2009 12:24 p.m.
To: Otto Sanchez; OSL Group
Subject: Re: [OSL | CCIE_Voice] Lab 10A BW Calculations
 
1.- For task 10.1, we must allow two calls between hq and br1 using rsvp,
however, the rsvp profile max sessions software is set to 4 calls, is there
any reason not to use 2 in this field?.

vik: Yes you can set max sessions set to 2 in the MTP- however the MTP could
also be used as an MTP not associated with RSVP call agent so it might be
wise to either increase your max sessions within the MTP or put the MTP into
a separate MRG that is bottom of the list within the MRGL. In other words-
only use the MTP for RSVP CAC.


Later In the same task the ip rsvp bandwidth is set to 80 kbps in the Serial
interface. According to the srnd we should reserve (N-1 x L3 Call BW)+ (1
Worst Case Call L3 BW), so that if we¹re using g729 codec that would be: 1 x
24 kbps (20 ms or whatever the default packet size is in ucm) + 40 kbps (10
ms call) = 64 Kbps. Although I think the 80kbps calculation fits well when
you have very few calls (in this case you could have both ringing at the
same time), does not scale well when you have a larger number of calls as
you might be reserving extra bandwidth, so I would have used 64kbps here.

vik: the amount of bandwidth requested is hardcoded to 40kbps for a g729
call  irrespective of what is actually used.


2.- In the task 10.4, to provision the llq priority queue the value
configured is (2 x L2 Call BW) + (1 Worst case L3 Call) or 2 x 26.8 kbps +
40 kbps = 93.6 kbps rounded up to 94 kbps. Aren¹t we considering three calls
here?. I agree that srnd recommends to over provision the llq when using
rsvp, but I think this is being considered when we add the worst case
scenario call (40 kbps), in consequence my calculation here would be 26.8
Kbps (1 L2 call) + 40 Kbps (worst case call) = 66.8 Kbps rounded to 67 Kbps
for the priority command,


vik: Good find! We configured 3 calls in the LLQ whereas it should have been
2.


-- 
Vik Malhi ­ CCIE #13890, CCSI #31584
Senior Technical Instructor - IPexpert, Inc.

Telephone: +1.810.326.1444
Fax: +1.810.454.0130
Mailto: [email protected]


Join our free online support and peer group communities:
http://www.IPexpert.com/communities
IPexpert - The Global Leader in Self-Study, Classroom-Based, Video-On-Demand
and Audio Certification Training Tools for the Cisco CCIE R&S Lab, CCIE
Security Lab, CCIE Service Provider Lab , CCIE Voice Lab and CCIE Storage
Lab Certifications.







From: Otto Sanchez <[email protected]>
Date: Tue, 7 Jul 2009 06:56:46 -0430
To: OSL Group <[email protected]>
Subject: [OSL | CCIE_Voice] Lab 10A BW Calculations

Hi List,
 
After finishing lab 10A and taking a look at its proctor guide, a couple of
doubts have come into my mind, regarding the way the bandwidth values are
being calculated for rsvp there, so I hope you can give me some guidance:
 
1.- For task 10.1, we must allow two calls between hq and br1 using rsvp,
however, the rsvp profile max sessions software is set to 4 calls, is there
any reason not to use 2 in this field?.
 
Later In the same task the ip rsvp bandwidth is set to 80 kbps in the Serial
interface. According to the srnd we should reserve (N-1 x L3 Call BW)+ (1
Worst Case Call L3 BW), so that if we¹re using g729 codec that would be: 1 x
24 kbps (20 ms or whatever the default packet size is in ucm) + 40 kbps (10
ms call) = 64 Kbps. Although I think the 80kbps calculation fits well when
you have very few calls (in this case you could have both ringing at the
same time), does not scale well when you have a larger number of calls as
you might be reserving extra bandwidth, so I would have used 64kbps here.
 
2.- In the task 10.4, to provision the llq priority queue the value
configured is (2 x L2 Call BW) + (1 Worst case L3 Call) or 2 x 26.8 kbps +
40 kbps = 93.6 kbps rounded up to 94 kbps. Aren¹t we considering three calls
here?. I agree that srnd recommends to over provision the llq when using
rsvp, but I think this is being considered when we add the worst case
scenario call (40 kbps), in consequence my calculation here would be 26.8
Kbps (1 L2 call) + 40 Kbps (worst case call) = 66.8 Kbps rounded to 67 Kbps
for the priority command,
 
Please let me know if I¹m missing something in the above assumptions,
 
Thanks,
 
Otto 


_______________________________________________
For more information regarding industry leading CCIE Lab training, please
visit www.ipexpert.com


_______________________________________________
For more information regarding industry leading CCIE Lab training, please visit 
www.ipexpert.com

Reply via email to