How Can Cisco PerfMon be Used to Improve Application Performance Visibility?

Short for Performance Monitor, the Cisco PerfMon API gives clients access to platform, service, and application performance counters allowing them to monitor in real-time the Cisco Unified Communications Manager (Cisco Unified CM) software and hardware.

What this means is that by monitoring packet flows in the network an admin can anticipate network issues before they begin, detect bottlenecks, and actively improve performance. Performance monitoring is important for real-time applications such as streaming video or VoIP which carries data that expires quickly and is especially sensitive to packet loss.

Application Performance and Cisco PerfMon

PerfMon is a useful performance tool in scenarios like:

  • To determine peak hours of contact at call centers and to cover with staff accordingly
  • To analyze and store the health of Cisco Unified CM in order to forecast budget, upgrades and replacements

PerfMon reports on several key metrics to make data useful in these types of scenarios:

  • Source and Destination IP Addresses
  • Port Numbers
  • Jitter
  • Packet Loss
  • Transport Event Packet Loss
  • Round Trip Time

Configuring PerfMon Example

Configuring PerfMon Example

Image from Cisco Performance Monitor

 

Using this example topology, H1 will generate traffic that R1 will collect PerfMon metrics about and then export to S1.

A basic PerfMon configuration could track both TCP and RTP traffic by creating two flow records. Then by combining these flow records with a flow exporter (which exports flow record data to a destination) data can be directed at a flow monitor. Two flow monitors could be used, one for TCP traffic, and one for RTP traffic. By setting a class-map, the routers will be told what traffic to collect. Two class-maps are created, one each for TCP and RTP. These class-maps are then added to a overall policy-map that is applied to an interface, such as a GigabitEthernet.

 

Visual representation configuration linkages in PerfMon

Image from Cisco Performance Monitor

PerfMon can become complex, and to help admins reduce the complexity, Cisco Easy Performance Monitor (Easy PerfMon or ezPM) has been created to work with Application and Visibility controls. Easy PerfMon trades the full flexibility of the traditional PerfMon in favor of configuration models called “profiles” that represent typical deployment scenarios. This allows admins to quickly set up their systems, but still retain much of the PerfMon functionality.

Related Terms

Cisco Application Visibility and Control (AVC)

The Cisco Application Visibility and Control (AVC) solution deploys a holistic approach for managing quality of service (QoS) technologies. It intelligently prioritizes traffic for critical applications while reducing or preventing traffic from noncritical or unwanted applications in an attempt to improve network and application performance over a wide area network (WAN).

 

Network Performance Management

Network performance management is the practice of monitoring a network and mitigating performance issues in order to maintain a required network performance level. Network monitoring plays a pivotal role in network performance management. It helps to identify performance bottlenecks by routinely reporting on the performance and quality service level of network devices such as routers. It is also used to ensure compliance with Service Level Agreements (SLA).

 

Bandwidth Utilization

Bandwidth utilization refers to the amount of bandwidth consumed on a network or network segment and the breakdown of its composite traffic. Understanding how bandwidth is utilized in a network is of critical importance to a network manager who is chiefly responsible for ensuring peak network performance. By using sophisticated network monitoring software, they can track traffic throughout large and complex networks and determine how bandwidth is being utilized. For instance, software solutions can monitor bandwidth utilization by port, and even by direction, which is extremely useful in pinpointing problems or to justify additional bandwidth. They can then enact policies that prioritize traffic for high demand applications.