Method and apparatus for extracting and visualizing execution patterns from web services

• US 7,873,728 B2
• Filed: 06/30/2008
• Issued: 01/18/2011
• Est. Priority Date: 11/15/2004
• Status: Active Grant

• Current Assignee: Kyndryl Incorporated
• Original Assignee: International Business Machines Corporation
• Inventors: Morar, John, Lei, Michelle, De Pauw, Wim, Krasikov, Sophia
• Primary Examiner(s): Najjar; Saleh
• Assistant Examiner(s): Kim; Edward J
• Application Number:
  US12/164,726
• Publication Number:
  US 20090006615A1
• Time in Patent Office: 932 Days
• Field of Search: 709/223, 709/224
• US Class Current: 709/224
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G06Q 40/00 Finance; Insurance; Tax strategies; Processing of corporate or income taxes

Abstract

One embodiment of the present method and apparatus for extracting and visualizing execution patterns from Web Services collects data relating to one or more transactions executed in accordance with one or more Web Services-based applications (e.g., messaging systems or other services). One or more patterns is then extracted from the collected data, where the patterns are representative transactions commonly executed in accordance with the Web Services configuration under analysis.

Claims

1. A method for analyzing a Web Services configuration, said method comprising:

collecting data relating to one or more messages exchanged in accordance with said Web Services configuration; and

extracting one or more patterns from said data, said one or more patterns being representative of one or more transactions executed in accordance with said Web Services configuration, wherein said extracting comprises;

reducing one or more repetitive sequences occurring in said one or more transactions to produce one or more reduced sequences;

representing each of said one or more transactions, including said one or more reduced sequences, as a graph, such that a plurality of graphs is produced; and

identifying commonalities among said plurality of graphs,

wherein at least one of;

the collecting, the extracting, the reducing, or the representing is performed using a processor.

2. The method of claim 1, wherein said identifying comprises:

dividing said graphs into one or more sub-sets, where all graphs within a given sub-set share structural similarities; and

dividing at least one of said one or more sub-sets into two or more sub-sub-sets, where all graphs within a given sub-sub-set share at least one of;

at least one node attribute or at least one edge attribute.

3. The method of claim 2, wherein all graphs within a given sub-set are isomorphic.

4. The method of claim 2, wherein said at least one node attribute is at least one of:

a service point, a port type, a machine, node performance data or a function of a service point, a port type, a machine or node performance data.

5. The method of claim 2, wherein said edge attribute is at least one of:

an operation, a message name, a message type, an error type, performance data, message content or a function of an operation, a message name, a message type, an error type, performance data or message content.

6. The method of claim 2, wherein said at least one node attribute or at least one edge attribute is deemed to match across graphs in an associated sub-sub-set in accordance with a predefined matching criterion.

7. The method of claim 6, wherein said predefined matching criterion requires said at least one node attribute or at least one edge attribute to be identical across said graphs in said associated sub-sub-set.

8. The method of claim 6, wherein said predefined matching criterion requires said at least one node attribute or at least one edge attribute to be similar within a defined degree across said graphs in said associated sub-sub-set.

9. The method of claim 2, wherein said at least one node attribute or at least one edge attribute is deemed to match across graphs in said associated sub-sub-set in accordance with an automatically discovered matching criterion.

10. The method of claim 2, further comprising:

defining at least one pattern in accordance with at least one of said two or more sub-sub-sets.

11. The method of claim 10, wherein said defining comprises:

identifying a suitable representation for all transactions represented within a given sub-sub-set; and

defining said suitable representation as said at least one pattern.

17. The method of claim 10, wherein said defining comprises:

creating a new transaction that is as similar as possible to all transactions represented within a given sub-sub-set, when all attributes of all transactions represented within the given sub-sub-set are not identical; and

defining said new transaction as said at least one pattern.

18. The method of claim 17, wherein the creating comprises:

selecting a node, an edge, or an attribute as a starting point for a graph representing the new transaction, where the node, the edge, or the attribute that is selected was used to match all transactions within the given sub-sub-set; and

identifying a common denominator for each node, edge, or attribute that was not used to match all transactions within the given sub-sub-set, wherein the common denominator is used to extend the graph representing the new transaction.

19. The method of claim 10, further comprising:

adding to said at least one pattern information that aggregates data from all transactions within the given sub-sub-set.

20. The method of claim 11, wherein all attributes of all transactions represented within the given sub-sub-set are identical.

12. A non-transitory computer readable medium containing an executable program for analyzing a Web Services configuration, where the program performs steps comprising:

collecting data relating to one or more messages exchanged in accordance with said Web Services configuration; and

extracting one or more patterns from said data, said one or more patterns being representative of one or more transactions executed in accordance with said Web Services configuration, wherein said extracting comprises;

reducing one or more repetitive sequences occurring in said one or more transactions to produce one or more reduced sequences;

representing each of said one or more transactions, including said one or more reduced sequences, as a graph, such that a plurality of graphs is produced; and

identifying commonalities among said plurality of graphs.

13. The non-transitory computer readable medium of claim 12, wherein said identifying comprises:

dividing said graphs into one or more sub-sets, where all graphs within a given sub-set share structural similarities; and

dividing at least one of said one or more sub-sets into two or more sub-sub-sets, where all graphs within a given sub-sub-set share at least one of;

at least one node attribute or at least one edge attribute.

14. The non-transitory computer readable medium of claim 13, further comprising:

defining at least one pattern in accordance with at least one of said two or more sub-sub-sets.

15. The non-transitory computer readable medium of claim 14, wherein said defining comprises:

identifying a suitable representation for all transactions represented within a given sub-sub-set; and

defining said representation as a pattern.

16. Apparatus for analyzing a Web Services configuration, the apparatus comprising:

means for collecting data relating to one or more messages exchanged in accordance with said Web Services configuration; and

means for extracting one or more patterns from said data, said one or more patterns being representative of one or more transactions executed in accordance with said Web Services configuration, wherein said means for extracting comprises;

means for reducing one or more repetitive sequences occurring in said one or more transactions to produce one or more reduced sequences;

means for representing each of said one or more transactions, including said one or more reduced sequences, as a graph, such that a plurality of graphs is produced; and

means for identifying commonalities among said plurality of graphs,

wherein at least one of;

the means for collecting, the means for extracting, the means for reducing, or the means for representing comprises a processor.