Network Performance Monitoring
It's not because it's UP that it's FAST
Network Performance Monitoring let IT pros work on issues before they affect end-users.
What is NPM?
It’s continuous monitoring by sending synthetic traffic from decentralized agents to establish a performance baseline and history.
Once the network performance monitoring (NPM) feature is setup between agents, the agents exchange UDP traffic to measure the network performance between each pair of agents. This light traffic is non-intrusive and very secure. It does not require packet capture so it respects the users' privacy and does not require any changes to device configuration.
The agents send the synthetic traffic every 500 ms. So twice a second, the agents measure the network performance between each other. Compared to other monitoring solutions, Obkio Network Performance Monitoring continuously monitors the performance instead of executing periodic checks at various intervals. This is mandatory to be able to measure intermittent issues that cause slowness in networks.
Network performance can only be measured between two points in the network. With centralized solutions, the performance is always measured between the centralized server and the remote locations. With Obkio, since the agents are distributed at strategic network locations (head office, branch office, datacenters, and clouds), the solution is decentralized to allow measurement between each pair of agents.
Establish Baseline and History
Have you ever faced a situation where a user tells you that the service was slow yesterday and you don’t have any clue as to what happened because everything is fine in your logs and device monitoring systems? This is why network performance history is required to go back in time and establish a baseline of the network performance with a 1-minute granularity.
Continuous Monitoring of Network Performance Metrics
User Quality of Experience (QoE) measured every minute
With the Live feature, get the real-time network performance, updated every 500ms in the App!
The network performance monitoring solution calculates the VoIP Quality for each network performance monitoring sessions every minute. The VoIP Quality score is calculated even if there is an ongoing call so this gives proactive monitoring over packet capture solution. Don't wait for bad user experience complaints after phone calls to start troubleshooting! This Quality of Experience (QoE) metric helps IT pros understand the impact of network performance on VoIP, which is a very performance sensitive application.
Speed tests can be run manually or by schedule between agents or groups of agents to test the connectivity bandwidth throughput. With scheduled speed tests, users can introduce load on the network path to make sure all components are working correctly. If one part of the network chokes, including equipment that are not under your control, NPM alerts will notify it. The maximum throughput depends on the hardware on which the agent is running and can go up to 10Gbps.
Traceroutes are often used by IT Pros to identify the network path to a destination and, in some cases, it helps to find where the issue is located. Looking at traceroutes when there is an issue is great but being able to compare it with a traceroute executed when everything was running fine is even better. Once a network monitoring session is configured between two agents, traceroutes are executed automatically between the agents every 10 minutes in each direction.
For private networks (ex: MPLS VPN Networks) with QoS enabled, the QoS Monitoring feature monitors the network performance for each class of service (ex: Data vs Priority). Also, the QoS marking inside the IP DSCP header is tested for end-to-end propagation. If a device rewrites the IP DSCP field, the agents will raise an alarm. This feature makes sure your QoS is always active on the network.