Method for improving local descriptors in peer-to-peer file sharing

• US 7,558,789 B2
• Filed: 11/20/2006
• Issued: 07/07/2009
• Est. Priority Date: 11/20/2006
• Status: Active Grant

• Current Assignee: Ophir Frieder, Wai Gen Yee
• Original Assignee: Illinois Institute of Technology
• Inventors: Frieder, Ophir, Yee, Wai Gen
• Primary Examiner(s): Trujillo; James
• Assistant Examiner(s): Casanova; Jorge A
• Application Number:
  US11/602,039
• Publication Number:
  US 20080120295A1
• Time in Patent Office: 960 Days
• Field of Search: 707/3, 707/10, 709/201, 709/217
• US Class Current: 1/1
• CPC Class Codes: G06F 16/152 using file content signatures, e.g. hash values
• CPC Class Codes: G06F 16/1834 implemented based on peer-to-peer networks, e.g. gnutella p2p communication protocols H04L67/104
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching

Abstract

A method for improving searches in a peer-to-peer (P2P) file sharing system that includes a plurality of server computers. A content file, identified by a descriptor including at least one metadata term and a mathematical identifier that uniquely identifies the content file in one of the server computers, is selected for searching. Other server computers are searched to find one or more matching content files; one that has a descriptor with a mathematical identifier matching the mathematical identifier of the first content file. The descriptors of the matching content files are returned to the searching server computer and used to expand the local descriptor.

Claims

1. In a method for searching a peer-to-peer (P2P) file sharing system of a plurality of server computers, wherein a search query is routed to the server computers for comparing query terms to a descriptor of each of a plurality of server content files in one or more of the server computers and returning as search results a descriptor of each of matching server content files, and wherein the descriptor includes at least one metadata term and a key, the improvement comprising:

automatically searching for a matching server content file of another of the server computers according to;

T<

[N_fN_q/(N_r+1)]−

N_pT

 

where T is a user-defined threshold, N_f is a number of content files available for sharing, N_q is a number of search queries received, N_r is a number of responses returned in response to received search queries, and N_p is the number of probe queries performed, and

expanding a first descriptor of a content file of one of the server computers using a second descriptor of the matching server content file of another of the server computers.

2. The method of claim 1, wherein expanding the first descriptor increases the likelihood the first descriptor will match a query looking for the content file.

3. The method of claim 1, wherein the second descriptor includes a key matching the first descriptor.

4. The method of claim 1, wherein expanding the first descriptor comprises incorporating into the first descriptor a metadata term from the second descriptor.

5. The method of claim 1, wherein searching for the matching server content file comprises comparing a key of the first content file to a key of each of one or more descriptors of the another of the server computers.

6. The method of claim 1, further comprising automatically searching for the matching server content file when the one of the server computers, during a predetermined time period, has returned a number of responses to received search queries that is less than a predetermined number.

7. The method of claim 1, comprising automatically searching for the matching server content file as a function of the number of search queries received by the one of the server computers during a predetermined time period.

8. The method of claim 1, comprising automatically searching for the matching server content file when a number of files available for sharing reaches a predetermined number.

9. The method of claim 1, wherein the first descriptor is a descriptor of the one of the server computers that includes the fewest metadata terms or has matched the fewest received search queries over a predetermined time period.

10. The method of claim 1, wherein the first descriptor has matched the fewest received search queries over a predetermined time period, and further comprising artificially increasing a search query match count after expanding the first descriptor.

11. Hardware for implementing the method of claim 1.

12. The hardware of claim 11, comprising an Application Specific Integrated Circuit (ASIC) or a Field Programmable Gate Array (FPGA).

13. A method for improving searches in a peer-to-peer (P2P) file sharing system including a plurality of server computers, the method comprising:

one of the server computers automatically selecting a first content file in the one of the server computers according to;

T<

[N_fN_q/(N_r+1)]−

N_pT

 

where T is a user-defined threshold, N_f is a number of content files available for sharing, N_q is a number of search queries received, N_r is a number of responses returned in response to received search queries, and N_p is the number of probe queries performed, wherein the first content file includes a first descriptor comprising at least one metadata term and a key that uniquely identifies the content file;

the one of the server computers automatically searching at least one other of the server computers to find a second content file having a second descriptor including a key matching the key of the first content file;

one of the at least one other of the server computers returning the second descriptor to the one of the server computers; and

the one of the server computers automatically expanding the first descriptor using the second descriptor.

14. The method according to claim 13, wherein searching the at least one other of the server computers comprises:

routing the key of the first content file to the at least one other of the server computers; and

comparing the key of the first content file to a key of each of one or more descriptors of the at least one other of the server computers.

15. The method of claim 13, further comprising:

returning a plurality of descriptors from the at least one other of the server computers, each including a key matching the key of the first content file; and

selecting from the plurality of descriptors a metadata term that is not in the first descriptor; and

incorporating the metadata term from the plurality of descriptors into the first descriptor.

16. The method of claim 15, wherein each of the at least one of the other of the server computers that has a matching content file with a key matching the key of the first content file returns a matching content file descriptor according to a given probability.

17. The method of claim 15, further comprising selecting as the metadata term the metadata term appearing most or least frequently in the plurality of descriptors.

18. The method of claim 15, further comprising randomly selecting the metadata term.

19. The method of claim 15, further comprising:

establishing a predetermined metadata term limit for the first descriptor;

selecting from the plurality of descriptors more than one metadata term that is not in the first descriptor; and

incorporating the more than one metadata term from the plurality of descriptors into the first descriptor, wherein the predetermined metadata term limit is not exceeded.

20. The method of claim 13, further comprising automatically selecting and searching the first content file when the one of the server computers, during a predetermined time period, has returned a number of responses to received search queries that is less than a predetermined number.

21. The method of claim 13, further comprising automatically selecting and searching the first content file as a function of the number of search queries received by the one of the server computers during a predetermined time period.

22. The method of claim 13, further comprising automatically selecting and searching the first content file when a number of files on the one of the server computers available for sharing reaches a predetermined number.

23. The method of claim 13, further comprising selecting as the first content file a server content file having a descriptor that includes the fewest metadata terms or has matched the fewest received search queries over a predetermined time period.

24. The method of claim 13, further comprising:

selecting as the first content file a server content file having a descriptor that has matched the fewest received search queries over a predetermined time period; and

artificially increasing a search query match count for the first content file after expanding the first descriptor.