Overview of CoS on Trio MPCs/MICs Features

This topic covers aspects of Class of Service (CoS) configuration for the Trio Modular Port Concentrator (MPC) and Modular Interface Card (MIC), with the emphasis on differences between the Trio family and other families of interface types. Included on Trio MPCs/MICs is support for shaping rates at the queue level, configurable bandwidth profiles with percentages, dynamic bandwidth allocation among different services, scheduler node scaling, and delay buffer allocation. To configure these features, include the relevant statements at the [edit class-of-service] hierarchy level and apply them if necessary at other hierarchy levels such as the [edit interfaces] hierarchy level.

You can configure Differentiated Service Code Point (DSCP) marking for Generic Routing Encapsulation (GRE) encapsulated packets that aligns with the service provider core CoS policy for an Multicast VPN (MVPN). To configure this feature, include the DSCP rewrite-rule dscp dscp-rule-name statement with the DSCP values at the [edit class-of-service] hierarchy level and then apply the rewrite rule to the core-facing multicast interface at the [edit class-of-service interfaces] hierarchy level.

When the delay buffers are oversubscribed by configuration (that is, the user has configured more delay-buffer memory than the system can support), then the configured Weighted Random Early Discard (WRED) profiles are implicitly scaled to drop packets more aggressively from the relatively full queues. This creates buffer space for packets in the relatively empty queues and provides a sense of fairness among the delay buffers. There is no configuration needed for this feature.

When load balancing on the Trio MPCs/MICs, you should configure odd and even numbered MICs in the form interface-fpc/odd | even/ports. For example, if one link is xe-1/0/0, the other should be xe-1/1/0. If you do not configure odd and even load balancing, the system RED-drops packets when sending at line rate.

On the Trio MPCs/MICs, in contrast to the Enhanced Intelligent Queuing (IQE) and Enhanced Intelligent Queuing 2 (IQ2E) PICs, the guaranteed rate is set to zero at the logical interface (unit) level in oversubscribed per-unit scheduling mode so that all queues operate only in the excess region. The output rate of each queue is based on their weight. For more information on hierarchical schedulers, see Configuring Hierarchical Schedulers for CoS.