Accessible application cluster topology

• US 10,462,011 B2
• Filed: 08/27/2015
• Issued: 10/29/2019
• Est. Priority Date: 08/27/2015
• Status: Active Grant

• Current Assignee: Nicira Incorporated (Broadcom, Inc.)
• Original Assignee: Nicira Incorporated (Broadcom, Inc.)
• Inventors: Olmsted-Thompson, Jeremy
• Primary Examiner(s): Cheema, Umar
• Assistant Examiner(s): Fan, John
• Application Number:
  US14/838,310
• Publication Number:
  US 20170063630A1
• Time in Patent Office: 1,524 Days
• Field of Search: 709223
• CPC Class Codes: H04L 41/042 comprising distributed management centres cooperatively managing the network
• CPC Class Codes: H04L 41/0893 Assignment of logical groups to network elements
• CPC Class Codes: H04L 41/12 Discovery or management of network topologies
• CPC Class Codes: H04L 45/02 Topology update or discovery
• CPC Class Codes: H04L 67/1097 for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
• CPC Class Codes: H04L 67/52 specially adapted for the location of the user terminal
• CPC Class Codes: H04L 67/63 Routing a service request depending on the request content or context
• CPC Class Codes: H04L 69/329 in the application layer [OSI layer 7]

Abstract

Some embodiments provide a method for maintaining a cluster topology for a cluster of application instances operating across several datacenters. On a particular machine at which a particular one of the application instances operates, the method maintains a cluster topology that identifies, for each application instance of the cluster, the datacenter in which the application instance operates. From the particular application instance, the method receives a query request for at least a portion of the cluster topology through a programmatic interface. The method provides the requested portion of the cluster topology to the particular application instance. The particular application instance uses the cluster topology for processing application data based on the locations of a set of application instances within the several datacenters.

Claims

1. A method for maintaining a cluster topology for a cluster of application instances operating across a plurality of datacenters using a plurality of cluster managers operating on machines across the plurality of datacenters, each cluster manager for (i) managing a set of application instances of the cluster of application instances that operate on the same machine as the cluster manager and (ii) maintaining a copy of the cluster topology, the method comprising:

at a cluster manager executing on a particular machine at which a particular one of the application instances operates;

exchanging a set of control data messages that comprise the cluster topology with a set of cluster managers in the plurality of cluster managers to maintain a cluster topology that identifies, for each application instance of the cluster, the datacenter of the plurality of datacenters in which the application instance operates;

from the particular application instance, receiving a query request for at least a portion of the cluster topology; and

providing the requested portion of the cluster topology to the particular application instance, wherein the particular application instance uses the cluster topology to identify locations of application instances in the plurality of datacenters and based on the identified locations to process differently at least two different sets of application data destined for two different application instances operating in two different datacenters.

2. The method of claim 1, wherein the particular application instance is a first application instance of the cluster and the query request is to identify the datacenter in which a second application instance of the cluster operates.

3. The method of claim 1, wherein the query request is for a complete list of the application instances of the cluster and their associated datacenters.

4. The method of claim 1, wherein the query request is for a list of application instances in a particular one of the plurality of datacenters.

5. The method of claim 1, wherein each control data message of the set of control data messages comprises an invalid address that identifies the data message as a control data message.

6. The method of claim 1, wherein exchanging the set of control data messages that comprise the cluster topology comprises periodically sending control data messages that comprise the cluster topology to a set of randomly selected application instances of the cluster.

7. The method of claim 6, wherein the particular application instance is located in a particular datacenter and the set of randomly selected application instances are all selected from the particular datacenter.

8. The method of claim 6, wherein the particular application instance is located in a first datacenter and the set of randomly selected application instances comprises at least one application instance located in a different, second datacenter.

9. The method of claim 1, wherein maintaining the cluster topology comprises:

detecting that the particular application instance is no longer available; and

removing the particular application instance from the cluster topology.

10. The method of claim 1 further comprising:

maintaining an overlay network, wherein each application instance of the cluster of application instances is coupled to the other application instances of the cluster through the overlay network;

receiving the processed data from the particular application instance; and

forwarding the received processed data to a destination machine, at which a destination application instance operates, through the overlay network.

13. The method of claim 1, wherein the particular application instance is a first application instance, the query request is a first query request, the cluster of application instances is a first cluster, and the cluster topology is a first cluster topology, the method further comprising:

maintaining a second cluster topology for a second cluster of application instances;

receiving a second query request from a second application instance, of the second cluster of application instances, operating on the particular machine; and

providing the requested portion of the second cluster topology to the second application instance.

24. The method of claim 1, wherein:

the particular application instance is a first application instance that operates in a first datacenter;

the at least two different sets of application data comprise a first set of application data destined for a second application instance operating in the first datacenter and a second set of application data destined for a third application instance operating in a second datacenter;

the first application instance forwards the first set of application data to the application instance; and

the first application instance aggregates the second set of application data with other sets of application data from other application instances operating in the first datacenter before forwarding the aggregated sets of application data to the third application instance.

11. A non-transitory machine readable medium storing a control manager which when executed by at least one processing unit of a particular machine maintains a cluster topology for a cluster of application instances operating across a plurality of datacenters using a plurality of cluster managers operating on a plurality of machines across the plurality of datacenters, each cluster manager for (i) managing a set of application instances of the cluster of application instances that operate on the same machine as the cluster manager and (ii) maintaining a copy of the cluster topology, the control manager comprising sets of instructions for:

exchanging a set of control data messages that comprise the cluster topology with a set of cluster managers in the plurality of cluster managers to maintain a cluster topology that identifies, for each application instance of the cluster, the datacenter of the plurality of datacenters in which the application instance operates;

from a particular application instance executing on the particular machine, receiving a query request for at least a portion of the cluster topology; and

providing the requested portion of the cluster topology to the particular application instance, wherein the particular application instance uses the cluster topology to identify locations of application instances in the plurality of datacenters and based on the identified locations to process differently at least two different sets of application data destined for two different application instances operating in two different datacenters.

12. The non-transitory machine readable medium of claim 11, wherein the particular application instance is a first application instance and the query request is a first query request, the sets of instructions further comprising a set of instructions for receiving a second query request from a second application instance, of the cluster of application instances, operating on the particular machine.

25. The non-transitory machine readable medium of claim 11, the control manager further comprising sets of instructions for:

receiving a data message from the particular application instance, the data message comprising application data destined for a different application instance;

when the different application instance operates in a same datacenter as the particular application instance, forwarding the data message without modification; and

when the different application instance operates in a different datacenter than the particular application instance, encapsulating the data message before forwarding the data message to the different datacenter.

26. The non-transitory machine readable medium of claim 11, wherein the particular application instance is a first application instance of the cluster and the query request is to identify the datacenter in which a application instance of the cluster operates.

27. The method of claim 26, wherein the query request is to identify the datacenter in which the third application instance of the cluster operates.

14. A non-transitory machine readable medium storing a program which when executed by at least one processing unit maintains a cluster topology for a cluster of application instances operating across a plurality of datacenters, the program comprising sets of instructions for:

on a particular machine at which a first application instances operates, maintaining a cluster topology that identifies, for each application instance of the cluster, the datacenter of the plurality of datacenters in which the application instance operates;

from the first application instance, receiving a query request for at least a portion of the cluster topology through a programmatic interface; and

providing the requested portion of the cluster topology to the first application instance, wherein the first application instance uses the cluster topology to identify locations of application instances in the plurality of datacenters and (1) to process a first set of application data destined for a second application instance operating in the first datacenter by forwarding the first set of application data to the second application instance and (2) to process a second set of application data destined for a third application instance operating in a second datacenter by aggregating the second set of application data with other sets of application data from other application instances operating in the first datacenter before forwarding the aggregated sets of application data to the third application instance.

15. The non-transitory machine readable medium of claim 14, wherein the query request is to identify the datacenter in which the third application instance of the cluster operates.

16. The non-transitory machine readable medium of claim 14, wherein the query request is for a complete list of the application instances of the cluster and their associated datacenters.

19. The non-transitory machine readable medium of claim 14, wherein the set of instructions for maintaining the cluster topology comprises a set of instructions for exchanging the cluster topology among a plurality of cluster managers operating on machines across the plurality of datacenters, each cluster manager for (i) managing a set of application instances of the cluster of application instances that operate on the same machine as the cluster manager and (ii) maintaining a copy of the cluster topology.

20. The non-transitory machine readable medium of claim 19, wherein the set of instructions for exchanging the set of control data messages that comprise the cluster topology comprises a set of instructions for periodically sending the control data messages that comprise the cluster topology to a set of randomly selected application instances of the cluster.

21. The non-transitory machine readable medium of claim 14, wherein the set of instructions for maintaining the cluster topology comprises sets of instructions for:

detecting that the first application instance is no longer available; and

removing the first application instance from the cluster topology.

22. The non-transitory machine readable medium of claim 14, wherein the program further comprises sets of instructions for:

maintaining an overlay network, wherein each application instance of the cluster of application instances is coupled to the other application instances of the cluster through the overlay network;

receiving the processed data from the first application instance; and

forwarding the received processed data to a destination machine, at which a destination application instance operates, through the overlay network.

17. A method for maintaining a cluster topology for a cluster of application instances operating across a plurality of datacenters, the method comprising:

on a particular machine at which a first application instances operates, maintaining a cluster topology that identifies, for each application instance of the cluster, the datacenter of the plurality of datacenters in which the application instance operates;

from the first application instance, receiving a query request for at least a portion of the cluster topology through a programmatic interface; and

providing the requested portion of the cluster topology to the first application instance, wherein the first application instance uses the cluster topology to identify locations of application instances in the plurality of datacenters and (1) to process a first set of application data destined for a application instance operating in the first datacenter by forwarding the first set of application data to the application instance and (2) to process a second set of application data destined for a third application instance operating in a second datacenter by aggregating the second set of application data with other sets of application data from other application instances operating in the first datacenter before forwarding the aggregated sets of application data to the third application instance.

18. The method of claim 17, wherein the query request is for a list of application instances in a particular one of the plurality of datacenters.

23. The method of claim 17, further comprising:

receiving a data message from the first application instance, the data message comprising application data destined for a different application instance;

when the different application instance operates in a same datacenter as the first application instance, forwarding the data message without modification; and

when the different application instance operates in a different datacenter than the first application instance, encapsulating the data message before forwarding the data message to the different datacenter.