Load-Balancing In Replication Engine of Directory Server

• US 20110131288A1
• Filed: 12/02/2009
• Published: 06/02/2011
• Est. Priority Date: 12/02/2009
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Patel, Virag S.
• Granted Patent: US 8,832,215 B2
• Time in Patent Office: Days
• US Class Current: 709/208
• CPC Class Codes: G06F 9/505 considering the load
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/1019 Random or heuristic server selection
• CPC Class Codes: H04L 67/1036 Load balancing of requests to servers for services different from user content provisioning, e.g. load balancing across domain name servers

Abstract

An embodiment of this invention provides a new LDAP server control to automate load balancing for replication based directory infrastructure. This solution is applicable to both master-master and master-replica based replication topologies. If one or more of the replication servers or the master servers are available for processing, the request is assigned to them based on output of a weighted non-uniform random number generator. This LDAP server control allows gaining very high performance while doing multiple parallel heavy LDAP operations. This LDAP control prevents the exposure of the replication directory server details to the outer world and therefore proves helpful in easing the security concerns.

Claims

1. A method for automatic load-balancing in a plurality of directory servers, the method comprising:

receiving a request at a first directory server among a plurality of directory servers, wherein each of the plurality of directory servers being configured to operate as one of a master directory server or a replication directory server; and

identifying whether the received request is one of a read request or a write request.

2. The method as claimed in claim 1, further comprising

detecting availability status of each of the plurality of directory servers.

3. The method as claimed in claim 2, further comprising

grouping the replication directory servers to replication server groups;

and

grouping the master directory servers to master server groups.

4. The method as claimed in claim 1, wherein in case of receiving the read request, further comprising

first selecting a first replication server group among the replication server groups based on source Internet protocol address of the read request.

5. The method as claimed in claim 3, further comprising

first assigning the read request to a first replication server that is available and belongs to the first replication server group.

6. The method as claimed in claim 5, wherein the first assigning is based on output of a first weighted non-uniform random number generator, if one or more of the replication servers belong to the first replication server group are available for processing the read request.

7. The method as claimed in claim 6, further comprising

repeating the first selecting and the first assigning steps until the read request is accepted for processing by one of the replication servers or a first configured timeout has reached.

8. The method as claimed in claim 4, further comprising

second selecting a first master server group among the master server groups, based on source Internet protocol address of the read request.

9. The method as claimed in claim 8, further comprising

second assigning the read request to a first master server that is available and belongs to the first master server group, if the first configured timeout has reached.

10. The method as claimed in claim 9, wherein based on a output of a second weighted non-uniform random number generator is used, if one or more of the master servers belonging to the first master server group are available for processing the read request.

11. The method as claimed in claim 10, further comprising

repeating the second selecting and the second assigning steps, until the read request is accepted for processing by one of the master servers or a second configured timeout has reached;

and queuing the read request in a common queue, if the second configured timeout has reached.

12. The method as claimed in claim 10, wherein in case of receiving the write request:

third selecting a second master server group among the master server groups, based on source Internet protocol address of the write request; and

third assigning the write request to a second master server that is available and belongs to the second master server group.

13. The method as claimed in claim 12, wherein based on a output of a third weighted non-uniform random generator is used, if one or more of the master servers belonging to the second master server group are available for processing the write request.

14. The method as claimed in claim 13, further comprising

repeating the third selecting and third assigning steps, until the write request is accepted for processing by one of the master servers or a third configured timeout has reached;

and

queuing the write request in a master queue, if the third configured timeout has reached.

15. A system having at least a processor configured for receiving a request at a first directory server among a plurality of directory servers, wherein each of the plurality of directory servers being configured to operate as one of a master directory server or a replication directory server;

and identifying whether the received request is one of a read request or a write request.

16. The system as claimed in claim 15, further comprising detecting availability status of each of the plurality of directory servers, grouping the replication directory servers to replication server groups;

and grouping the master directory servers to master server groups.

17. The system as claimed in claim 15, wherein in case of receiving the read request, further comprising first selecting a first replication server group among the replication server groups based on source Internet protocol address of the read request.

18. The system as claimed in claim 17, further comprising first assigning the read request to a first replication server that is available and belongs to the first replication server group, wherein the first assigning is based on output of a first weighted non-uniform random number generator, if one or more of the replication servers belong to the first replication server group are available for processing the read request, and repeating the first selecting and the first assigning steps until the read request is accepted for processing by one of the replication servers or a first configured timeout has reached.

19. The system as claimed in claim 17, further comprising second selecting a first master server group among the master server groups, based on source Internet protocol address of the read request;

second assigning the read request to a first master server that is available and belongs to the first master server group, if the first configured timeout has reached, and wherein based on a output of a second weighted non-uniform random number generator is used, if one or more of the master servers belonging to the first master server group are available for processing the read request.

20. The system as claimed in claim 19, further comprising

repeating the second selecting and the second assigning steps, until the read request is accepted for processing by one of the master servers or a second configured timeout has reached;

and queuing the read request in a common queue, if the second configured timeout has reached.

21. The system as claimed in claim 17, wherein in case of receiving the write request, third selecting a second master server group among the master server groups, based on source Internet protocol address of the write request;

and third assigning the write request to a second master server that is available and belongs to the second master server group, based on a output of a third weighted non-uniform random generator is used, if one or more of the master servers belonging to the second master server group are available for processing the write request.

22. The system as claimed in claim 21, further comprising repeating the third selecting and third assigning steps, until the write request is accepted for processing by one of the master servers or a third configured timeout has reached;

and queuing the write request in a master queue, if the third configured timeout has reached.