Performance metric of a system conveying web content

• US 10,230,590 B2
• Filed: 12/03/2013
• Issued: 03/12/2019
• Est. Priority Date: 12/03/2013
• Status: Active Grant

• Current Assignee: Telefonaktiebolaget LM Ericsson
• Original Assignee: Telefonaktiebolaget LM Ericsson
• Inventors: Da Silva, Icaro L. J., Kuivinen, Fredrik, Fu, Jing
• Primary Examiner(s): Nguyen, Thu Ha T
• Application Number:
  US15/035,398
• Publication Number:
  US 20160294640A1
• Time in Patent Office: 1,925 Days
• Field of Search: 709223, 709224
• CPC Class Codes: H04L 41/5009 Determining service level performance parameters or violations of service level contracts, e.g. violations of agreed response time or mean time between failures [MTBF]
• CPC Class Codes: H04L 41/5083 wherein the managed service relates to web hosting
• CPC Class Codes: H04L 47/193 at the transport layer, e.g. TCP related
• CPC Class Codes: H04L 47/6215 Individual queue per QOS, rate or priority
• CPC Class Codes: H04L 61/5007 Internet protocol [IP] addresses
• CPC Class Codes: H04L 67/02 based on web technology, e.g. hypertext transfer protocol [HTTP]
• CPC Class Codes: H04L 69/22 Parsing or analysis of headers

Abstract

Obtaining a performance metric in a system for conveying web content from a server node to a terminal node along one or more network nodes, involving an inspecting of a data flow transmitting said web content toward the terminal node for extracting web content records. The extracted web content records are correlated to at least one web session. The extracted web content records are associated to a performance of one or more of said nodes. The performance metric is calculated from the correlated and associated web content records for at least one web session and one or more of said nodes.

Claims

1. A method of obtaining a performance metric, in a system for conveying web content from a server node to a terminal node along one or more network nodes, the method comprising the steps of:

inspecting a data flow transmitting said web content toward the terminal node for extracting web content records;

correlating said extracted web content records to at least one web session;

associating said extracted web content records to a performance of one or more of said nodes; and

calculating the performance metric from the correlated and associated web content records for at least one web session and one or more of said nodes, wherein the performance metric is calculated by calculating a resource download time for conveying a resource to the terminal node for each of a plurality of resource requests in a same web session from records being associated to the performance of the one or more nodes involved in conveying the resource to the terminal node,

wherein the method further comprises identifying a cause of low performance based on the calculated performance metric by

computing a page loading time (PLT) overhead caused by each of a plurality of overhead sources;

comparing absolute values of the PLT overhead of each overhead source with the PLT overhead of each other overhead source; and

selecting a maximum PLT overhead associated with a particular overhead source as the cause of low performance.

2. The method of claim 1, wherein the calculating the performance metric is repeated to calculate a plurality of performance metrics, each performance metric being associated to one or more different nodes from group of the terminal node, the server node, and the one or more network nodes.

3. The method of claim 1, wherein a client processing time is calculated as one performance metric, said client processing time indicating a processing performance of the terminal node.

4. The method of claim 1, wherein a Transmission Control Protocol (TCP) setup time is calculated as one performance metric, said TCP setup time indicating a time to set up a TCP connection between the terminal node and the server node.

5. The method of claim 1, wherein a server processing time is calculated as one performance metric, said server processing time indicating a processing performance of the server node.

6. The method of claim 1, wherein the resource download time indicates a time required for conveying a resource to the terminal node along one or more involved network nodes.

7. The method of claim 1, wherein a performance metric is calculated based on two extracted records belonging to a same web session or is calculated based on each of all resource requests in one web session.

8. The method of claim 1, further comprising a step of building a list of associated performance metric values.

9. The method of claim 8, further comprising a step of analyzing the built list using a statistical procedure for calculating a reference value for a performance metric.

10. The method of claim 1, wherein the step of inspecting the data flow comprises a deep packet inspection procedure for extracting headers as said web content records.

11. A network entity for obtaining a performance metric in a system for conveying web content from a server node to a terminal node along one or more network nodes, the network entity comprising:

a processor; and

a memory coupled to the processor and storing program code that when executed by the processor cause the processor to;

inspect a data flow transmitting said web content toward the terminal node for extracting web content records;

correlate said extracted web content records to at least one web session;

associate said extracted web content records to a performance of one or more of said nodes; and

to

calculate the performance metric from the correlated and associated web content records for at least one web session and one or more of said nodes, wherein the performance metric is calculated by calculating a resource download time for conveying a resource to the terminal node for each of a plurality of resource requests in a same web session from records being associated to the performance of the one or more nodes involved in conveying the resource to the terminal node,

wherein the processor further caused to identify a cause of low performance based on the calculated performance metric by;

computing a page loading time (PLT) overhead caused by each of a plurality of overhead sources;

comparing absolute values of the PLT overhead of each overhead source with the PLT overhead of each other overhead source; and

selecting a maximum PLT overhead associated with a particular overhead source as the cause of low performance.

12. The network entity of claim 11, wherein the calculation of the performance metric is repeated to calculate a plurality of performance metrics, each performance metric being associated to one or more different nodes from group of the terminal node, the server node, and the one or more network nodes.

13. The network entity of claim 11, wherein a client processing time is calculated as one performance metric, said client processing time indicating a processing performance of the terminal node.

14. The network entity of claim 11, wherein a Transmission Control Protocol (TCP) setup time is calculated as one performance metric, said TCP setup time indicating a time to set up a TCP connection between the terminal node and the server node.

15. The network entity of claim 11, wherein a server processing time is calculated as one performance metric, said server processing time indicating a processing performance of the server node.

16. The network entity of claim 11, wherein the resource download time indicates a time required for conveying a resource to the terminal node along one or more involved network nodes.

17. The network entity of claim 11, wherein a performance metric is calculated based on two extracted records belonging to a same web session or is calculated based on each of all resource requests in one web session.

18. The network entity of claim 11, wherein the program code when executed by the processor further causes the processor to build a list of associated performance metric values.

19. The network entity of claim 18, wherein the program code when executed by the processor further causes the processor to analyze the built list using a statistical procedure for calculating a reference value for a performance metric.

20. The network entity of claim 11, wherein the inspection of the data flow comprises a deep packet inspection procedure for extracting headers as said web content records.

21. A computer program product comprising a non-transitory computer readable storage medium storing program code configured to be executed by a processor of a system which conveys web content from a server node to a terminal node along one or more network nodes, the program code causing the processor to:

inspect a data flow transmitting said web content toward the terminal node for extracting web content records;

correlate said extracted web content records to at least one web session;

associate said extracted web content records to a performance of one or more of said nodes; and

calculate a performance metric from the correlated and associated web content records for at least one web session and one or more of said nodes, wherein the performance metric is calculated by calculating a resource download time for conveying a resource to the terminal node for each of a plurality of resource requests in a same web session from records being associated to the performance of each node associated with the one or more nodes involved in conveying the resource to the terminal node,

wherein the program code further causing the processor to identify a cause of low performance based on the calculated performance metric identifying a cause of low performance by;

computing a page loading time (PLT) overhead caused by each of a plurality of overhead sources;

comparing absolute values of the PLT overhead of each overhead source with the PLT overhead of each other overhead source; and

selecting a maximum PLT overhead associated with a particular overhead source as the cause of low performance.