System and method for network performance and server application performance monitoring and for deriving exhaustive performance metrics

1. In a system having one or more distributed probes and one or more analyzers coupled to a network, a method for network performance and server application performance monitoring in at least one of the probes, comprising:

installing configuration information into the probe;

detecting the arrival at the probe of a flow instance of interest based on the configuration information;

determining with the probe whether the arriving flow instance is a new flow instance;

if the arriving flow instance is a new flow instance, establishing with the probe a state associated with the flow instance;

capturing, with the probe, relevant information associated with the flow instance;

exporting the relevant information based on a configuration;

determining whether an end of the flow instance has been reached;

if the end of the flow instance has been reached, destroying the state and exporting the relevant information;

if the end of the flow instance has not been reached, continuing to monitor the flow instance;

installing configuration information in one or more of the analyzers; and

providing at least one of flow information and statistics from one or more of the probes to one or more of the analyzers;

wherein one or more of the analyzers use flow matching to uniquely associate events observed at various of the probes with a single flow instance, and perform a set of operations on data based on the configuration information;

wherein the flow matching includes:

determining a flow instance to be matched;

determining one or more down stream probes where a corresponding flow instance is capable of appearing;

determining a time window in a way that considers characteristics of the flow instance; and

searching the flow instances falling into the time window at the down stream probes;

wherein if any records characterized by the same flow identifier and sequence number are found, determining that a matching flow instance has been obtained.

2. The method of claim 1 wherein the relevant information associated with the flow instance includes one or more types of information chosen from the group consisting of: timestamps, probe identifiers, flow identifiers, sequence numbers, flow information and flow statistics.

3. The method of claim 1 wherein the flow instance includes one or more flow instances chosen from the group consisting of a single packet, a stream of packets that constitute a message and a stream of packets that constitute a session.

4. The method of claim 1 further comprising constructing a record from the captured flow information.

5. The method of claim 4 further comprising exporting the record to one or more of the analyzers, wherein the trigger for record export is based on the configuration.

6. The method of claim 1 wherein the configuration information includes one or more types of information chosen from the group consisting of flows that need to be monitored, flow instances that need to be captured, sets of performance metrics that need to be measured, and the trigger which governs when the information is exported.

7. The method of claim 1 wherein the configuration information installed on the analyzer includes one or more types of information chosen from the group consisting of flows that need to be monitored, flow instances that need to be captured, and sets of performance metrics that need to be measured.

8. The method of claim 1 , wherein the flow information pertains to on-going monitoring activities.

9. The method of claim 1 , wherein the flow information pertains to one or more flows that have terminated.

10. The method of claim 1 wherein one or more of the analyzers produce performance reports.

11. The method of claim 10 wherein the performance reports include QoS metrics and derived statistics.

12. The method of claim 11 wherein the QoS metrics include one or more metrics selected from the group of CS Msg Jitter, SC Msg Jitter, CS Strm Jitter, SC Strm Jitter, CS ERT SS, SC ERT SS, CS ERT ES, SC ERT ES, CS ERT SE, SC ERT SE, CS ART SS, SC ART SS, CS ART ES, SC ART ES, CS ART SE, SC ART SE, Conn Establish, Conn Graceful Term, Conn Timeout Term, CS Conn Window, SC Conn Window, CS Hop Count, and SC Hop Count.

13. The method of claim 11 wherein the derived statistics include one or more statistics selected from the group consisting of Number of Data Points, Frequency, Maximum, Minimum, Range, Arithmetic Mean, Root Mean Square, Variance, Standard Deviation, Trending.

14. The method of claim 1 further comprising, by matching flow instances, associating various events with the flow instance of interest, and deriving one or more network or server application QoS metrics.

15. A system having one or more distributed probes and one or more analyzers coupled to a network, a set of processor readable instructions executable on a processor of one or more of the probes, the instructions embodying a method for flow instance capturing in at least one of the probes, comprising:

installing configuration information into the probe;

detecting the arrival at the probe of a flow instance of interest based on the configuration information;

determining with the probe whether the arriving flow instance is a new flow instance;

if the arriving flow instance is a new flow instance, establishing with the probe a state associated with the flow instance;

capturing, with the probe, relevant information associated with the flow instance;

determining whether an end of the flow instance has been reached;

if the end of the flow instance has been reached, destroying the state and exporting the relevant information; and

if the end of the flow instance has not been reached, continuing to monitoring the flow instance;

wherein flow matching is performed and includes:

determining a flow instance to be matched;

determining one or more down stream probes where a corresponding flow instance is capable of appearing;

determining a time window in a way that considers characteristics of the flow instance; and

searching the flow instances falling into the time window at the down stream probes;

wherein if any records characterized by the same flow identifier and sequence number are found, determining that a matching flow instance has been obtained.

16. A processor readable medium having embodied therein a set of a set of processor readable instructions executable on a processor of one or more probes in a system having one or more distributed probes and one or more analyzers coupled to a network, the instructions embodying a method for flow instance capturing in at least one of the probes, the method comprising:

installing configuration information into the probe;

detecting the arrival at the probe of a flow instance of interest based on the configuration information;

determining with the probe whether the arriving flow instance is a new flow instance;

if the arriving flow instance is a new flow instance, establishing with the probe a state associated with the flow instance;

capturing, with the probe, relevant information associated with the flow instance;

determining whether an end of the flow instance has been reached;

if the end of the flow instance has been reached, destroying the state and exporting the relevant information; and

if the end of the flow instance has not been reached, continuing to monitoring the flow instance;

wherein flow matching is performed and includes:

determining a flow instance to be matched;

determining one or more down stream probes where a corresponding flow instance is capable of appearing;

determining a time window in a way that considers characteristics of the flow instance; and

searching the flow instances falling into the time window at the down stream probes;

wherein if any records characterized by the same flow identifier and sequence number are found, determining that a matching flow instance has been obtained.

17. A method, comprising:

detecting a flow instance;

determining whether the flow instance is a new flow instance;

if the arriving flow instance is a new flow instance, establishing a state associated with the flow instance; and

capturing flow information associated with the flow instance;

wherein the flow matching is utilized to associate events with the flow instance;

wherein the flow matching includes:

determining a flow instance to be matched,

determining one or more locations where a matching flow instance is capable of appearing,

determining a time window associated with the flow instance to be matched, and searching a plurality of flow instances associated with the time window at the one or more locations;

wherein, if data characterized by the same flow identifier and sequence number are found, indicating that a matching flow instance has been identified.