IDENTIFYING A POLLING COMMUNICATION PATTERN

• US 20140372590A1
• Filed: 01/23/2012
• Published: 12/18/2014
• Est. Priority Date: 01/23/2012
• Status: Active Grant

• Current Assignee: Micro Focus LLC (Open Text Corporation)
• Original Assignee: Leonid Pekel
• Inventors: Eliassaf, Ofer, Kessner, Amir, Zemer, Meidan, Keret, Oded, Pekel, Leonid
• Granted Patent: US 9,838,263 B2
• Time in Patent Office: Days
• US Class Current: 709/223
• CPC Class Codes: H04L 41/145 involving simulating, designing, planning or modelling of a network
• CPC Class Codes: H04L 43/026 using flow identification
• CPC Class Codes: H04L 43/50 Testing arrangements
• CPC Class Codes: H04L 67/02 based on web technology, e.g. hypertext transfer protocol [HTTP]
• CPC Class Codes: H04L 67/566 Grouping or aggregating service requests, e.g. for unified processing

Abstract

A method for identifying a polling communication pattern within a sequence of communication entities includes grouping the communication entities into a plurality of clusters according to a criterion. Clusters are removed from the plurality according to at least one of a time pattern analysis, cluster size, and cluster duration. The clusters remaining are identified as having a polling communication pattern.

Claims

1. A method for identifying a polling communication pattern within a sequence of communication entities, comprising:

grouping the communication entities into a plurality of clusters according to a criterion;

removing clusters from the plurality according to at least one of a time pattern analysis, a cluster size, and a cluster duration; and

identifying remaining clusters as having a polling communication pattern.

2. The method of claim 1, wherein each communication entity includes a response and a request and wherein grouping comprises grouping communication entities that share at least one of an identical request method, identical request domain, and identical number of request key value pairs into the same cluster.

3. The method of claim 1, wherein grouping comprises:

comparing a first communication entity of the plurality of communication entities with a second communication entity of the plurality of entities; and

grouping the first and second communication entities in the same cluster only if an edit distance between a request of the first communication entity and the second communication entity is within a predetermined threshold.

4. The method of claim 1, wherein removing clusters from the plurality according to a time pattern analysis comprises:

for each cluster, dividing that cluster into chains of sequential communication entities that exclude communication entities for which the time pattern analysis indicates are not polling communications; and

examining the chains of each cluster and removing a selected cluster according to the examination.

5. The method of claim 4, wherein:

examining comprises distinguishing at least one of a cluster that contains chains of differing polling communication patterns and a cluster that contains less than a predetermined threshold percentage of communication entities included in a chain; and

removing comprises removing a distinguished cluster.

6. A computer readable medium having instructions stored thereon that when executed by a processing resource implement a system comprising a cluster engine, a chain engine, a remove engine;

and an identification engine, wherein;

the cluster engine is configured to group the sequence of communication entities into a plurality of clusters according to a criterion;

the chain engine is configured to divide each cluster into chains of sequential communication entities excluding communication entities having timing patterns not indicative of polling communications;

the remove engine is configured to remove clusters from the plurality according to at least one a cluster size, a cluster duration; and

a chain analysis; and

the identification engine is configured to identify a remaining cluster as having a polling communication pattern.

7. The medium of claim 6, wherein each communication entity includes a response and a request and wherein the cluster engine is configured to group communication entities that share at least one of an identical request method, identical request domain, and identical number of request key value pairs into the same cluster.

8. The medium of claim 6, wherein the cluster engine is configured to group by:

comparing a first communication entity of the sequence of communication entities with a second communication entity of the sequence of entities; and

grouping the first and second communication entities in the same cluster only if an edit distance between a request of the first communication entity and the second communication entity is within a predetermined threshold.

9. The medium of claim 6, wherein each cluster includes a sequence of communication entities and each communication entity includes a request and a response and wherein:

for each cluster, the chain engine is configured to divide by comparing start and end times for the communication entities of that cluster, and building one or more chains of sequential communication entities that have start and end times that are indicative of polling communications;

the remove engine is configured to examine the chains of each cluster and to remove a selected cluster based on the examination.

10. The medium of claim 9, wherein the remove engine is configured to remove a selected cluster where the examination for that cluster reveals at least one of:

that the selected cluster contains chains of differing polling communication patterns; and

the selected cluster contains more than a predetermined threshold percentage of communication entities excluded from a chain.

11. A system for identifying a polling communication pattern, the system comprising a processing resource in communication with a computer readable medium, wherein the computer readable medium contains a set of instructions and wherein the processing unit is designed to carry out the set of instructions to:

group the plurality of communication entities into a plurality of clusters according to a criterion;

remove clusters from the plurality according to at least one of a time pattern analysis, a cluster size, and a cluster duration; and

identify remaining clusters as having a polling communication pattern.

12. The system of claim 11, wherein each communication entity includes a response and a request and wherein the set of instructions includes instruction to cause the processing resource to group communication entities that share at least one of an identical request method, identical request domain, and identical number of request key value pairs into the same cluster.

13. The system of claim 11, wherein the set of instructions includes instructions to cause the processing resource to group the plurality of communication entities into a plurality of clusters by:

comparing a first communication entity of the plurality of communication entities with a second communication entity of the plurality of entities; and

grouping the first and second communication entities in the same cluster only if an edit distance between a request of the first communication entity and the second communication entity is within a predetermined threshold.

14. The system of claim 11, wherein the set of instructions include instructions to cause the processor resource to remove clusters from the plurality by:

dividing each of the plurality of clusters into chains of sequential communication entities according to a time pattern analysis of the communication entities; and

examining the chains of each cluster and removing a selected cluster according to the examination.

15. The system of claim 11, wherein each cluster includes a sequence of communication entities and each communication entity includes a request and a response and wherein the set of instructions includes instructions to cause the processing resource to remove clusters by:

for each cluster, comparing start and end times for each communication entity of that cluster and identifying one or more chains of sequential communication entities that have start and end times that are indicative of a polling communication;

examining the chains to select clusters that contain chains of differing polling communication patterns or clusters that contain more than a predetermined threshold percentage of communication entities excluded from a chain; and

removing the selected clusters.