Geolocation-based load balancing

• US 8,929,896 B2
• Filed: 02/24/2012
• Issued: 01/06/2015
• Est. Priority Date: 02/24/2012
• Status: Active Grant

• Current Assignee: Empire Technology Development LLC (Allied Inventors Management, LLC)
• Original Assignee: Empire Technology Development LLC (Allied Inventors Management, LLC)
• Inventors: Kurabayashi, Shuichi, Yoshida, Naofumi, Takano, Kosuke
• Primary Examiner(s): TRAN, CONGVAN
• Application Number:
  US13/640,888
• Publication Number:
  US 20130225176A1
• Time in Patent Office: 1,047 Days
• Field of Search: 455436-439, 455/450, 455/453, 370/229, 370/230, 370/230.1, 370/252, 370/253, 370/338, 370351-356
• US Class Current: 455/437
• CPC Class Codes: H04W 36/023 Buffering or recovering information during reselection
• CPC Class Codes: H04W 64/006 with additional information processing, e.g. for direction or speed determination

Abstract

Technologies are generally described for geolocation-based load balancing in a cloud computing environment hosting a mobile service. In some examples, a method may include calculating a probability of movement of a user device from a first cell to a second cell, and calculating an amount of cached data to copy from a second cache server corresponding to the second cell to a first cache server corresponding to the first cell based at least in part on the calculated probability.

Claims

1. A method comprising:

calculating a probability of movement of a user device from a first cell to a second cell; and

calculating an amount of cached data to copy from a second cache server corresponding to the second cell to a first cache server corresponding to the first cell based at least in part on the calculated probability,

wherein the first cache server caches data regarding the first cell, and the second cache server caches data regarding the second cell.

2. The method of claim 1, further comprising:

receiving a request for geographical data from the user device; and

extracting at least one of an identifier and geolocation information associated with the user device from the received request,

wherein the calculating the probability of movement is based on the at least one of the extracted identifier and geolocation information.

3. The method of claim 2, wherein the request includes an HTTP (Hypertext Transfer Protocol) request, and wherein the extracting at least one of the identifier and the geolocation information includes analyzing a header of the HTTP request.

4. The method of claim 3, wherein the HTTP request includes an HTTP GET request or an HTTP POST request, and the identifier includes at least one of a cookie and an identification text string in the header of the HTTP GET request or the HTTP POST request.

5. The method of claim 2, wherein the geolocation information includes GPS (Global Positioning System) information.

6. The method of claim 1, wherein the calculating the probability of movement of the user device is based at least in part on a moving speed of the user device.

7. The method of claim 1, wherein the first cache server and the second cache server are in-memory cache servers.

8. The method of claim 1, wherein the first cache server and the second cache server respectively cache from a database server geographical data on the first cell and geographical data on the second cell.

9. The method of claim 1, wherein the amount of cached data to copy increases with the calculated probability of movement.

10. The method of claim 1, wherein the calculating amount of cached data to copy is based at least in part on a distance between the first cell and the second cell.

11. The method of claim 1, wherein the first cell and the second cell are substantially a same size with each other.

12. The method of claim 1, wherein respective sizes of the first cell and the second cell vary depending on at least one of total number of cache servers and memory capacity of the cache servers.

13. The method of claim 1, further comprising:

copying the calculated amount of cached data from the second cache server to the first cache server.

14. A load balancer, comprising:

a device information extractor configured to extract, from a stream received from a user device, at least one of an identifier and geolocation information associated with the user device;

a geolocation mesh manager configured to manage an association between cache servers and real-space cells,

wherein each cache server caches, from a database server, information regarding a cell allocated thereto;

a movement probability calculator configured to calculate a probability of movement of the user device from a first real-space cell to a second real-space cell; and

a copy factor calculator configured to calculate amount of cached data to copy from a second cache server corresponding to the second real-space cell to a first cache server corresponding to the first real-space cell based at least in part on the calculated probability of movement.

15. The load balancer of claim 14, wherein the stream stems from an HTTP (Hypertext Transfer Protocol) request from the user device, and

wherein the device information extractor is further configured to analyze a header of the HTTP request to extract at least one of the identifier and the geolocation information associated with the user device.

16. The load balancer of claim 14, wherein the geolocation information includes GPS (Global Positioning System) information.

17. The load balancer of claim 14, wherein the first cache server and the second cache server are in-memory cache servers.

18. The load balancer of claim 14, wherein the first cache server and the second cache server respectively cache, from the database server, geographical data regarding the first real-space cell and geographical data on the second real-space cell.

19. The load balancer of claim 14, wherein the movement probability calculator is further configured to calculate the probability based at least in part on moving speed of the user device.

20. The load balancer of claim 14, wherein the copy factor calculator is further configured to calculate the amount of cache copy from the second cache server to the first cache server based at least in part on distance between the first real-space cell and the second real-space cell.

21. The load balancer of claim 14, wherein sizes of the first real-space cell and the second real-space cell vary depending on at least one of total number of cache servers and memory capacity of the cache servers.

22. A non-transitory computer-readable storage medium that stores a program for implementing load balancing between a database server storing geographical data and a Web application server, the program comprising:

a geolocation mesh managing module configured to manage association between real-space cells and cache servers, each of which caches from the database server at least a portion of the geographical data associated with a corresponding one of the real-space cells; and

a copy factor calculating module configured to calculate amount of cached data to copy from a second cache server corresponding to a second real-space cell to a first cache server corresponding to a first real-space cell based at least in part on a calculated probability of movement of a user device from the first real-space cell to the second real-space cell and a distance between the first real-space cell and the second real-space cell.

23. The non-transitory computer-readable storage medium of claim 22, wherein the program further comprises:

an HTTP (Hypertext Transfer Protocol) header analyzing module configured to analyze an HTTP request from the user device and extract from the HTTP request at least one of an identifier and geolocation information associated with the user device.

24. The non-transitory computer-readable storage medium of claim 22, wherein the first cache server and the second cache server are in-memory cache servers.