RE: Configuring Generic Traffic Shaping

From: CCIE 19999 (ccie@iprimus.com.au)
Date: Sun Aug 26 2007 - 21:18:44 ART


Lim,

VC List Rate Limit bits/int bits/int (ms) (bytes)
Active
- 28000 1988 7952 7952 284 994
- 32000 2000 8000 8000 250 1000
- 40000 2000 8000 8000 200 1000
- 48000 1992 7968 7968 166 996
- 56000 1988 7952 7952 142 994
- 64000 2000 8000 8000 125 1000
- 72000 1998 7992 7992 111 999
- 80000 2000 8000 8000 100 1000
- 88000 1980 7920 7920 90 990
- 96000 1992 7968 7968 83 996
- 104000 1976 7904 7904 76 988
- 112000 1988 7952 7952 71 994
- 120000 1980 7920 7920 66 990
- 128000 1984 7936 7936 62 992
- 256000 1984 7936 7936 31 992
- 257000 1542 6168 6168 24 771
- 384000 2400 9600 9600 25 1200
- 512000 3200 12800 12800 25 1600
- 768000 4800 19200 19200 25 2400
- 1024000 6400 25600 25600 25 3200
- 1536000 9600 38400 38400 25 4800

You may be interested in the above default values.

Regards,
Shine

-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of Toh
Soon, Lim
Sent: Monday, 27 August 2007 7:03 AM
To: Cisco certification
Subject: Configuring Generic Traffic Shaping

Hi All,

I need to configure GTS on a frame relay interface with target bit rate of
128Kbps & such that FECNs are reflected as BECNs.

I configured as follows:

!
int s 0/0/0
 traffic-shape rate 128000
 traffic-shape fecn-adapt
!

R2#sh run int s 0/0/0
!
interface Serial0/0/0
 encapsulation frame-relay
 traffic-shape rate 128000 7936 7936 1000
 traffic-shape fecn-adapt

R2#sh traffic-sh

Interface Se0/0/0
       Access Target Byte Sustain Excess Interval Increment Adapt
VC List Rate Limit bits/int bits/int (ms) (bytes)
Active
- 128000 1984 7936 7936 62 992 BECN

The router adds the other values in the running-config. Can anyone explain
how the router determines the values for sustain bits and excess bits? It
appears to me they are based on the formula CIR=Bc/Tc. But I'm only
providing the CIR value. How are the other values being calculated?

From Command Ref on DocCD,

An interval is calculated as follows:
* If the burst-size is not equal to zero, the interval is the burst-size
divided by the bit-rate.
* If the burst-size is zero, the interval is the excess-burst-size divided
by the bit-rate.

In my case, both my burst-size and excess-burst-size are unspecified.

Any comments are most welcome.

Thank you.

Confused,
Lim TS



This archive was generated by hypermail 2.1.4 : Sat Sep 01 2007 - 11:32:13 ART