Impact of traffic (packet rate) on CPU - without CAR
Goal: study of impact of traffic on the CPU utilization at the ingress interface
(FastEthernet) and output interface (ATM)
Equipment:
- cisco RSP1 (R4700) processor with 65536K/2072K bytes of memory.
R4700 CPU at 100Mhz, Implementation 33, Rev 1.0
- 1 ATM network interface, VIP2 R5K controller
- 1 FastEthernet/IEEE 802.3 interface, VIP2 controller
- IOS (tm) RSP Software (RSP-JSV-M), Version 12.0(7)T,
RELEASE SOFTWARE (fc2)
Test Description
- Topology
The two FastEthernet (duplex) intrfaces of the SmartBits are connected
back-to-back to the FastEthernet interfaces of the C7500 and C7200
The ATM connection between the routers is UBR (155 Mbps) and not shared with
other sources of traffic
- Parameters:
- EF packet rate (pack/sec)
- EF IP packet size: 64 bytes
- no CAR or other QoS features configured
- CPU utilization: monitored through command:
show controller vip tech
- Stream profile:
- traffic load: variable (constant rate), packet size constant
- 1 stram, UDP; generated by the SmartBits
- Router configurations:
Comments:
- Figure 1 shows that the CPU of the FastEthernet interface gets
saturated for a maximum rate of 39000 pack/sec. For this rate 2/3 pack/sec are dropped
by the router because of CPU saturation. 39000 pack/sec correspond to 24.33 Mbps when
the IP packet size is 64 bytes.
- Utilization of CPU on the ATM interface is much lower.
Fig.1: CPU utilization of the FastEthernet interface and ATM interface for different
packet rates (without CAR)
