Cisco IOS XE 17.16.1a

Application performance monitoring helps you configure performance monitors using which you can view real-time, end-to-end application performance filtered by client segments, network segments, and server segments.

This information helps you optimize application performance and reduce downtime

• Cisco Catalyst 8000V Edge Software

• Cisco Catalyst 8500 Series Edge Platforms

• Cisco Catalyst 8300 Series Edge Platforms

• Cisco Catalyst 8200 Series Edge Platforms

• Cisco 1000 Series Integrated Services Routers

• Cisco 4461 Integrated Services Router

class-map match-any APP_PERF_MONITOR_APPS_0

The APP_PERF_MONITOR_APPS_0 is the class-map that contains application groups to monitor.

This parameter determines how packets are evaluated when multiple match criteria exist. A packet must match any of the match statements to be accepted. If you do not specify the match-any or match-all keyword, the default keyword of match-all is used.

match protocol attribute application-group

To configure the match criterion for a class map based on the specified application group, use the match protocol attribute application-group command.

The application-group keyword allows the configuration of applications grouped together based on the same networking application as the match criteria.

For example, Yahoo-Messenger, Yahoo-VoIP-messenger, and Yahoo-VoIPover-SIP are grouped together under the yahoo-messenger-group.

class-map match-any APP_PERF_MONITOR_FILTERS

The APP_PERF_MONITOR_FILTERS is the filter that is used by a specific monitor.

This parameter sets the context of how monitoring has to be performed.

performance monitor context APP_PM_POLICY profile application-aggregation

The APP_PM_POLICY is the name of performance monitor context . A context contains information about a traffic monitor that has to be enabled.

The application-aggregation profile is the preconfigured and default policy for SD-Routing devices to filter traffic based on your intent. To enable the performance monitor context on a specified interface, use the performance monitor context command.

exporter destination localcontroller source Null0

The exporter destination command is used to export metrics from performance monitors to the Cisco vManage.

traffic-monitor art-aggregated class-and APP_PERF_MONITOR_FILTERS interval-timeout 300 sampling- interval 100

The application monitoring framework provides two types of traffic-monitors:

The art-aggregated monitor only monitors TCP traffic. Whereas the media-aggregated monitor only monitors RTP traffic. Both these monitors have default filters.

In addition to the default filter, the key word class-and specifies a customized filter using which different application groups can be defined and therefore provides more exact scope of the traffic to be monitored.

The interval-timeout 300 command means the metrics are reported to Cisco vManage every 300 seconds. 300 seconds is the recommended value. If interval-timeout keyword is not explicitly configured, the default value is 60 seconds.

The sampling-interval 100 command means that every 100 milliseconds, a new flow of a certain application on a certain interface is analysed to monitor its metric. 100 milliseconds is the recommended value.

If sampling-interval command is not explicitly configured, the default behaviour is to monitor every flow, which leads to more processing overhead on a device.

interface Tunnel100 performance monitor context APP_PM_POLICY

You can use context to specify the interface on which the performance monitoring context has to be enabled.