Method and system to bundle message over a network

• US 20040117469A1
• Filed: 12/16/2002
• Published: 06/17/2004
• Est. Priority Date: 12/16/2002
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: International Business Machines Corporation
• Inventors: Krishnapuram, Raghuram, Kummamuru, Krishna, Kumar, Manoj
• Granted Patent: US 7,299,273 B2
• Time in Patent Office: Days
• US Class Current: 709/223
• CPC Class Codes: G06F 2209/505 Clust
• CPC Class Codes: G06F 9/5061 Partitioning or combining of resources

Abstract

The invention describes a method and system to optimize network bandwidth and obtain greater efficiency in transmission of messages/data in, a client-server network. The invention proposes the use of clustering of client requests and the data items in such a manner so as to optimize the network transmission as well as reduce the cost of processing involved in sending and picking/pruning the data items at server and client end respectively.

Claims

1. A method for minimizing the cost of processing client requests in a client-server network, comprising the steps of:

selecting a range of values for the number of clusters into which incoming client requests are to be clustered,

simultaneously clustering requests and requesting clients for each selected value,

calculating the cost of processing each cluster of requests and requesting clients, and

determining the cluster combination that provides minimum cost of processing.

2. The method as claimed in claim 1 further comprising the step of assigning the clusters of items to the respective clusters of clients.

3. The method as claimed in claim 1, wherein the step of selecting a range of values for the number of clusters is a configurable manual process.

4. The method as claimed in claim 1, wherein the step of clustering for each value in the chosen range for the number of clusters, comprises the steps of:

randomly initializing the fuzzy membership values of requests and requesting clients,

iteratively calculating the fuzzy membership values until the sum of changes of membership values is less than a configurable threshold value, and

defuzzifying the fuzzy membership values to obtain actual clusters of items and clients.

5. In a client-server network a system for minimizing the cost of servicing client requests comprising:

a selection means for selecting a range of values for the number of clusters into which incoming client requests are to be clustered,

a clustering means for simultaneously clustering requests and requesting clients for each selected value,

a calculating means for establishing the total processing cost for each selected value, and

a determining means for establishing the minimum cost of processing.

6. The system as claimed in claim 5 further comprising a means for assigning the clusters of items to the respective clusters of clients.

7. The system as claimed in claim 5, wherein the requesting clients are request generating computing devices connected to one or more computing devices acting as servers servicing requests made by the clients.

8. The system as claimed in claim 5, wherein the requests include information, messages and data required by the clients.

9. The system as claimed in claim 7, wherein the said means reside on one or a combination of server devices comprising:

a system bus,

a communications unit connected to the system bus,

a memory including a set of instructions connected to the system bus, and

a control unit executing the instructions in the memory.

10. The system as claimed in claim 7, wherein the clients and server computing devices are interconnected through any suitable computer network including Ethernet, Internet, LAN, WAN, and MAN using any desired network topology including ring, bus and star.

11. A computer program product comprising computer readable program code stored on a computer readable storage medium embodied therein for minimizing the total cost of processing client requests in a client-server network, comprising:

computer readable program code means configured for selecting a range of values for the number of clusters into which client requests are clustered prior to processing,

computer readable program code means configured for simultaneously clustering requests and clients for each selected value,

computer readable program code means configured for calculating the total cost of processing for each selected value for the number of clusters, and

computer readable program code means configured for determining the minimum cost of processing.

12. The computer program product as claimed in claim 11 further comprising computer readable program code means configured for assigning the clusters of requests to the respective clusters of clients.

13. The computer program product as claimed in claim 11, wherein the computer readable program code means of clustering for each value in the selected range for the number of clusters, comprise:

computer readable program code means configured for randomly initializing the fuzzy membership values of items and clients,

computer readable program code means configured for iteratively calculating the fuzzy membership values until the sum of changes of membership values is less than a configurable threshold value, and

computer readable program code means configured for defuzzifying the fuzzy membership values to obtain actual clusters of items and clients.