Traffic shaping in Cisco IOS/Queuing-only test
From CT3
In the first test in the traffic shaping test series we’ve verified the behavior of the MQC traffic shaping when the offered load does not exceed the traffic contract (specified with the shape rate class configuration command).
Contents |
Test result summary
The test verifies that Cisco IOS performs regular output interface queuing when the classified traffic does not exceed the traffic rate configured with the shape command.
Test setup
The following traffic shaping configuration was used on the test router:
Traffic shaping configuration on the test router
policy-map SHAPE class UDP shape average 128000 bandwidth 96 class class-default bandwidth 32
The traffic generator was sending two UDP streams: a 112 kbps UDP stream to IP address 172.16.0.12 …
UDP stream in class UDP
$ flood.pl --size=200 --bandwidth=112 172.16.0.12 Flooding 172.16.0.12 random port with 200-byte packets Interpacket delay 14.2857142857143 msec total IP bandwidth 112 kbps Break with Ctrl-C
… and a 64 kbps UDP stream to IP address 172.16.0.13
UDP stream in the default class
$ flood.pl --size=200 --bandwidth=64 172.16.0.13 Flooding 172.16.0.13 random port with 200-byte packets Interpacket delay 15 msec total IP bandwidth 64 kbps Break with Ctrl-C
Test results
The show policy-map interface command clearly shows that the router does not perform any traffic shaping activity (highlighted). The throughput of the UDP class is approximately 96 kbps (115 kbps offered rate minus 22 kbps drop rate) due to the output interface congestion.
The printout also shows the traffic shaping parameters (sustain bits = Bc, excess bits = Be, interval = Tc)
Test#show policy-map interface ser 0/1/0
Serial0/1/0
Service-policy output: SHAPE
Class-map: UDP (match-all)
33601 packets, 6854604 bytes
30 second offered rate 115000 bps, drop rate 22000 bps
Match: access-group name UDP
Traffic Shaping
Target/Average Byte Sustain Excess Interval Increment
Rate Limit bits/int bits/int (ms) (bytes)
128000/128000 1984 7936 7936 62 992
Adapt Queue Packets Bytes Packets Bytes Shaping
Active Depth Delayed Delayed Active
- 0 33601 6854604 38 7752 no
Queueing
Output Queue: Conversation 41
Bandwidth 96 (kbps)Max Threshold 64 (packets)
(pkts matched/bytes matched) 33600/6854400
(depth/total drops/no-buffer drops) 63/2248/0
Class-map: class-default (match-any)
39223 packets, 4076832 bytes
30 second offered rate 68000 bps, drop rate 37000 bps
Match: any
Queueing
Output Queue: Conversation 42
Bandwidth 32 (kbps)Max Threshold 64 (packets)
(pkts matched/bytes matched) 14058/1462032
(depth/total drops/no-buffer drops) 64/7567/0
The show queueing command displays the state of the interface output queue. The output queue on the Test router has two reserved WFQ conversations, belonging to class UDP and the default class. The third conversation is dynamic (outbound CDP packet).
Test#show queueing int ser 0/1/0
Interface Serial0/1/0 queueing strategy: fair
Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 59449
Queueing strategy: Class-based queueing
Output queue: 128/1000/64/44234 (size/max total/threshold/drops)
Conversations 3/34/32 (active/max active/max total)
Reserved Conversations 2/2 (allocated/max allocated)
Available Bandwidth 0 kilobits/sec
(depth/weight/total drops/no-buffer drops/interleaves) 64/256/2087/0/0
Conversation 42, linktype: ip, length: 104
source: 10.0.0.10, destination: 172.16.0.13, id: 0x8BA1, ttl: 127,
TOS: 0 prot: 17, source port 1044, destination port 8992
(depth/weight/total drops/no-buffer drops/interleaves) 63/86/589/0/0
Conversation 41, linktype: ip, length: 204
source: 10.0.0.10, destination: 172.16.0.12, id: 0x8BFB, ttl: 127,
TOS: 0 prot: 17, source port 1043, destination port 19678
(depth/weight/total drops/no-buffer drops/interleaves) 1/32384/0/0/0
Conversation 33, linktype: cdp, length: 326

BlogMarks
del.icio.us
digg
Facebook
LinkedIn
Newsvine
reddit
Slashdot