Home > Articles > Cisco Certification > CCNP > CCNP BCRAN Self-Study: Managing Network Performance with Queuing and Compression

CCNP BCRAN Self-Study: Managing Network Performance with Queuing and Compression

Chapter Description

To help you prepare for the CCNP BCRAN exam, Brian Morgan and Craig Dennis discuss some of the advanced queuing techniques offered in Cisco IOS and compression techniques that are possible from Cisco routers.

Low-Latency Queuing

Low-Latency Queuing (LLQ) is really just an extension of CBWFQ. In fact, the only real difference between the two is how the bandwidth is allocated to the class maps in the policy map.

In the examples shown thus far, the bandwidth command was used to allocate a certain amount of bandwidth to any given class map. Remember that the bandwidth command has an effect in the queuing strategy only if the interface is congested. Thus, if the interface is not filled up, there is no real guarantee that any particular class map will get the requested amount of bandwidth.

LLQ uses the priority command instead of the bandwidth command to request bandwidth. The priority command guarantees that the requested bandwidth is available whether the interface is busy or not. Because this bandwidth is always available, the class map that uses the priority command is guaranteed low latency through the interface (thus the name, LLQ). This is also called a strict priority queue.

It is important to remember that any packets that exceed the requested bandwidth (using the priority command) when the interface is busy are discarded (policed). During low interface utilization, the class map may use more bandwidth than requested by the priority command. But the class map will get the required bandwidth at all times.

The policy-map Command

LLQ is configured the same as CBWFQ. The difference is how the bandwidth is requested in the policy map. Example 15-13 shows the use of the priority command.

Example 15-13 The policy-map Command

policy-map bcran-policy
 class bcran-class
  priority 48 

The difference here is that class-map bcran-class is absolutely guaranteed 48 kbps of bandwidth at all times, regardless of how busy the interface is.

LLQ Verification

Because LLQ is nearly identical to CBWFQ, most of the verification screens are the same. The one place where the difference can be seen is when the policy map is examined. Example 15-14 examines the policy-map built for LLC.

Example 15-14 The show policy-map Command

Router# show policy-map 
 Policy Map bcran-policy
  Class other-class
   Weighted Fair Queueing
      Bandwidth 24 (kbps) Max Threshold 64 (packets)
  Class bcran-class
   Weighted Fair Queueing
      Strict Priority
      Bandwidth 48 (kbps) Burst 1200 (Bytes)
  Class class-default
   Weighted Fair Queueing
      Flow based Fair Queueing
      Bandwidth 0 (kbps) Max Threshold 64 (packets)

This time, class-map bcran-class is shown with a strict priority of 48 kbps. This verifies that it is guaranteed the requested bandwidth under all circumstances.

As shown in Figure 15-7, the flow of packets through the LLQ is almost identical to that of the flow through the CBWFQ. The only real difference is that the priority traffic in the LLQ is always guaranteed its prescribed bandwidth. The CBWFQ is only in operation when the interface is congested.

Figure 7Figure 15-7 Low-Latency Queuing

6. Compression Techniques | Next Section Previous Section