Method and apparatus for dynamically reconfiguring a server system

• US 7,739,364 B2
• Filed: 06/27/2002
• Issued: 06/15/2010
• Est. Priority Date: 02/13/2002
• Status: Active Grant

• Current Assignee: Telefonaktiebolaget LM Ericsson
• Original Assignee: Telefonaktiebolaget LM Ericsson
• Inventors: Danne, Anders, Blomme, Max, Willehadson, Stefan
• Primary Examiner(s): Pwu; Jeffrey
• Assistant Examiner(s): Musa; Abdelnabi O
• Application Number:
  US10/504,086
• Publication Number:
  US 20050066014A1
• Time in Patent Office: 2,910 Days
• Field of Search: 709/230, 709/220, 709/221, 709/203, 709/219, 709/238, 709/218, 709/223
• US Class Current: 709/223
• CPC Class Codes: G06F 2209/5016 Session
• CPC Class Codes: G06F 9/505 considering the load

Abstract

A method and apparatus for dynamically reconfiguring a server system. When a new server configuration is applied, a previous server configuration is saved. When receiving a service request, a first server is selected by using a first scheduling algorithm according to the new configuration. If the selected server is incorrect, a second server is selected by using a second scheduling algorithm according to the saved previous server configuration. User data associated with the received service request is moved from the second server, if correct, to the first server. In this way, stored data will be gradually and dynamically reorganized from the previous configuration to the new configuration in a relatively simple way.

Claims

1. A method of dynamically reconfiguring a server system comprising a plurality of servers for handling incoming service requests involving one or more storing tasks, each server locally storing different user data at that server in a first server configuration, comprising the steps of:

A)—

applying a second different server configuration where a number of the servers is different from the first server configuration,

B)—

saving the first server configuration, wherein the following further steps are executed by a service request scheduling unit when receiving a service request;

C)—

selecting a server for handling a received service request by using a first scheduling algorithm according to the second server configuration with a client or session identity of the received service request as input,

D)—

selecting a server by using a second scheduling algorithm according to the first server configuration saved in step B) with the client or session identity as input, if the server selected in step C) was incorrect by not having locally stored user data associated with the received service request, and

E)—

moving user data associated with the received service request from the server selected in step D), using the second scheduling algorithm, if the server selected in step D) is correct having locally stored the user data associated with the received service request, to the server selected in step C) using the first scheduling algorithm.

2. A method according to claim 1, comprising the further step of selecting a server by using a third scheduling algorithm according to a saved even earlier configuration, if the server selected in step D) using the second scheduling was also incorrect.

3. A method according to claim 1, wherein the first and second scheduling algorithms are hashing algorithms.

4. A method according to claim 3, wherein using the hashing algorithms includes deriving a hash number from a user ID and calculating a server ID number from the derived a hash number.

5. A method according to claim 4, comprising wherein the hashing algorithm is:

server ID=hash(user ID)modulo n 

 

(1)

where n is the number of possible servers, and the modulo operator providing an integer between 0 and n−

1.

6. A method according to claim 5, wherein the second server configuration involves a changed number of servers, and that in the second scheduling algorithm, n is changed accordingly from the first scheduling algorithm.

7. A method according to claim 1, wherein the service requests involve storing tasks.

8. A method according to claim 7, wherein the storing tasks include any one or more of:

storing new data, updating already stored data, and retrieving stored data.

9. A method according to claim 1, wherein step A) of applying a second server configuration includes storing new configuration data in a central administrator, which can be accessed by the servers.

10. A method according to claim 9, wherein new configuration data is pushed from the central administrator to one or more servers.

11. A method according to claim 9, wherein new configuration data is retrieved from the central administrator by sending a request from a server.

21. A computer program product comprising a software code stored on a tangible medium for method of claim 1 in a server system.

12. An apparatus for reconfiguring a server system, the server system comprising a plurality of servers each being capable of performing at least one storing task, and a scheduling unit configured to select servers for handling incoming service requests involving one or more storing tasks, each server locally storing different user data, wherein the scheduling unit is configured to perform the following:

apply a second different server configuration where a number of the servers is different from the first server configuration,

save the first server configuration,

select a first server for handling a received service request by using a first scheduling algorithm according to the second server configuration with a client or session identity of the received service request as input, and

select a second server by using a second scheduling algorithm according to the first server configuration, if the first server was incorrect by not having locally stored user data associated with the received service request, and

wherein the apparatus further comprises means for moving user data associated with the received service request from the second server, if the second server is correct having locally stored user data associated with the received service request, to the first server.

13. An apparatus according to claim 12, wherein the scheduling unit is configured to select a server by using a scheduling algorithm according to an even earlier configuration, if the second server was also incorrect.

14. An apparatus according to claim 12, wherein the scheduling unit is configured to use hashing algorithms as scheduling algorithms.

15. An apparatus according to claim 14, wherein the scheduling unit is configured to derive a hash number from a user ID and calculate a server ID number from the derived a hash number.

16. An apparatus according to claim 15, wherein the scheduling unit is configured to use the following algorithm:

server ID=hash(user ID)modulo n 

 

(1)

where n is the number of possible servers, and the modulo operator providing an integer between 0 and n−

1.

17. An apparatus according to claim 12, further comprising a central administrator for storing new configuration data, which can be accessed by the servers.

18. An apparatus according to claim 17, further comprising a local administrator in each server for receiving new configuration data from the central administrator.

19. An apparatus according to claim 17, wherein the central administrator comprises means for pushing new configuration data to one or more servers.

20. An apparatus according to claim 18, wherein the local administrator comprises means for sending a request for new configuration data to the central administrator.