Network Metrics

To get the clearest understanding of your network performance, Obkio understands and measures the performance metrics that matter most.

Network metrics are one of the most important elements in network monitoring, which is why Obkio continuously measures core network performance metrics between Monitoring Agents every 500ms.

Obkio measures network metrics by establishing a communication between Monitoring Agents in key network locations.

Once a Network Monitoring Session is established between two Monitoring Agents, the session will start collecting network metrics. This article will describe the network metrics that are measured for each and every session.

While sending the network performance measurement packets, the client and server agents add timing and sequence information that are analyzed by the client agent to measure the following network metrics:

• Latency refers to the measure of time it takes for data to reach its destination across a network. It is usually measured as a round trip delay, in milliseconds (ms), taking into account the time it takes for the data to get to its destination and then back again to its source.

• Jitter refers to the variation of the latency. To calculate the jitter, the current latency for a session is compared to the previous latency value. For example, if the latencies are 5ms 10ms 5ms 10ms 50ms, the jitter values will be 5ms 5ms 5ms 40ms.

• Packet Loss refers to the percentage of data packets that were sent by the client agent, and which received no response from the server. Over a period of 1 minute, 120 measurement packets are sent. If 5 packets are dropped by the network, the packet loss rate will be 4.2%.

In the Network Monitoring Session Graphs, the multiple percentiles of the latency and jitter are shown.

It’s not always easy to understand or identify the impact of a key network metric on the end-user experience. That’s why Obkio measures QoE (Quality of Experience) metric that’s based on the core network metrics introduced in the previous section but is used to represent the quality during a Voice call and the interpretation of the end-user perspective.

It allows you to figure out how good or bad the user's experience was with a VoIP call during a period of time. For example, if there’s a lot of jitter or packet loss happening, not only is the call quality affected, but it can also have an impact on the user experience. Learn more on the VoIP Quality experience metric.

In addition to the core network metrics, the network performance measurement packets also monitor some network anomalies such as Packet Reordering, Packet Duplication and QoS DSCP Mismatch.

A more detailed article about network metrics is available in our Blog: How to Measure Network Performance: 5 Network Metrics.

We also have a full article about How to Measure Packet Loss.