What is Active Service Assurance?
Active assurance involves the creation and measurement of synthetic traffic in a network. Typically, active assurance probes are deployed at the edge of a network and are programmed to trigger different types of traffic to a central location.??
This traffic maybe be low level test traffic, often TWAMP or ICMP based, or certain probes can simulate end users and trigger web browsing, YouTube, Netflix and so on.? ?Active assurance is particularly suited to the creation of “Birth Certificates”, which certify a new connection is ready for service, ahead of users connecting.
Active testing may also be able to speed test connections (during quiet times), to help validate speed claims.
What is Passive assurance?
As the name implies, passive assurance is a “listen only” mode of assurance, that silently analyses end user traffic.? Passive assurance tends to be more centrally located in a network and can be deployed on a network tap, i.e. not inline, but instead receiving a copy of user traffic.?
Since passive assurance sees actual user traffic, deployments can be very large.? Passive assurance is best suited for real user experience measurement. ?
AppLogic Networks and Allot are examples of passive assurance providers, some network equipment providers like Nokia and Ericsson also have similar offerings. Each provider has different features, the advantages of passive assurance in this article, is based on AppLogic Networks feature set. Some other vendors do not offer all these benefits.
Disadvantages compared
Whether or not the following are actual disadvantages will depend on the network type and use cases required. For example, an enterprise will have different requirements to a large B2C ISP.
Active Assurance
Passive Assurance
领英推荐
Which is Best, Passive or Active Assurance?
Spoiler alert – neither is, since they are mostly complementary, with neither having a full capability. Let’s dig a little deeper into the key capabilities, remember the big difference is what traffic is inspected. ?Active = synthetically generated traffic. Passive = end user traffic.
Active
This is able to work even when there is no user traffic, in fact, in-life, this is typically when it is used, to try and ensure that user traffic is not impacted. For example, test every hour when the connection is detected as idle, or perhaps during the night.? Active assurance can theoretically see potential problems before a user is impacted. ?
Active assurance is predictable, tests can be scheduled and trends observed easily. Active assurance is great at seeing connection up / down issues quickly, especially when using a low level protocol such as TWAMP.
Passive
A passive based approach can achieve many of the same results as active. Passive of course relies on end user traffic, but this opens up additional and important possibilities. Being able to measure and understand a user’s behaviour has two key benefits:
AppLogic Networks passive assurance can be trained and then alert to anomalies in user traffic, such as unusual drops in quality. ?Having a direct measurement of the user experience a network is delivering to end users is a powerful KQI and helps protect a brand’s reputation.
Passive testing does not add any additional load to a network.
Summary
Passive and Active Assurance complement each other, providing many similar, but some unique use cases. ?Active assurance is ideal to provide turn up testing and “Up / down” connectivity status very well.? Passive testing gives a view of the actual user and can include additional use cases based on that:
1.???? Marketing segmentation
2.???? Churn analysis and prediction
Together, Active and Passive assurance can give a full 360deg view of the service a network is providing, from birth, through to in life and to marketing and churn prevention.
Helping Telcos and Enterprises enable Experience First Networking
3 个月Good article, Tim. Very well summarized
Obsessing over the Customer Experience of ISPs
3 个月Kevin Penn Ian Sandman Nick Watson Daren Lawrence