Methods and apparatus for adaptive and holistic network measurements

A network monitoring system, comprising:

• a processor;
• and
• a storage device coupled to the processor and storing instructions that when executed by the processor cause the processor to perform a method, the method comprising:
• collecting, from a network device, traffic data associated with a network;
• applying a machine learning technique to analyze the traffic data to discover dependencies among a plurality of applications and services accessed by client devices within the network,
• wherein applying the machine learning technique comprises determining temporal and frequency correlations among packets associated with the plurality of applications and services, wherein determining the temporal correlation comprises determining whether packets destined for different applications and services are sent from a same client device within a predetermined interval, and wherein determining the frequency correlation comprises determining a number of client devices accessing each application or service;
• identifying, based on the discovered dependencies, a first server providing a first application and at least a second server providing a service that facilitates delivery of the first application;
• and
• performing active measurement on the first server and at least the second server, thereby facilitating monitoring of performance of the first application.

A non-transitory computer-readable storage medium storing instructions that when executed by a processor cause the processor to perform a method, the method comprising:

• collecting, from a network device, traffic data associated with a network;
• applying a machine learning technique to analyze the traffic data to discover dependencies among a plurality of applications and services accessed by client devices within the network, wherein applying the machine learning technique comprises determining temporal and frequency correlations among packets associated with the plurality of applications and services, wherein determining the temporal correlation comprises determining whether packets destined for different applications and services are sent from a same client device within a predetermined interval, and wherein determining the frequency correlation comprises determining a number of client devices accessing each application or service;
• identifying, based on the discovered dependencies, a first server providing a first application and at least a second server providing a service that facilitates delivery of the first application;
• and
• performing active measurement on the first server and at least the second server, thereby facilitating monitoring of performance of the first application.