Placement of control and data plane resources

• US 10,404,613 B1
• Filed: 03/31/2014
• Issued: 09/03/2019
• Est. Priority Date: 03/31/2014
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Brooker, Marc John, Levy, Marc, Kusters, Norbert Paul, Richardson, David R., Olson, Marc Stephen
• Primary Examiner(s): Bates, Kevin T
• Assistant Examiner(s): Mahmud, Golam
• Application Number:
  US14/230,539
• Time in Patent Office: 1,982 Days
• Field of Search: 709226
• CPC Class Codes: H04L 47/70 Admission control; Resource allocation

Abstract

Computing resource services have computing resources allocated among regions and data zones, where regions can be completely separate deployments of an infrastructure to provide the highest degree of independence. The more regions that are utilized by a customer, the higher the level of fault tolerance may exist. Current networks consist of heavily distributed infrastructures or mainly centralized infrastructures, both of which can cause complete control plane failures when even a subset of a data zone or region loses network connectivity. The placement of control plane and data plane resources provides for optimal availability and durability during infrastructure failures.

Claims

1. A computer-implemented method, comprising:

in response to receiving a volume creation request;

applying predetermined logic to determine a set of fault zones in which to construct a control plane for a volume and in which to instantiate the volume;

applying the predetermined logic to determine a number of dynamic control agents to form the control plane;

selecting, from a set of dynamic control agents in the set of fault zone a subset of dynamic control agents complying with the number;

instructing individual dynamic control agents of the subset to execute a consensus protocol to form consensus on one or more settings for the volume;

providing data corresponding to the volume into at least one data store operably interconnected to the control plane; and

causing the volume to be instantiated in accordance with the predetermined logic.

2. The computer-implemented method of claim 1, wherein the consensus protocol is configured to enforce consistent operation between the subset of dynamic control agents.

3. The computer-implemented method of claim 1, wherein the data includes at least metadata of the volume, a location of the volume, and/or a service hosting the volume.

4. The computer-implemented method of claim 1, wherein the subset of dynamic control agents are members of a larger number of different control planes, such that a control plane for each volume includes a set of dynamic control agents and each of the subset of dynamic control agents is a member of a control plane of other volume.

5. A system, comprising:

at least one computing device that implements one or more services, wherein the one or more services;

in response to receipt of a request to create a volume, apply selection logic to select a subset of a set of computer system processes, the set of computer system processes being;

able to participate in operation as a control plane for the volume; and

distributed among a plurality of computing devices; and

transmit one or more instructions that cause the subset to operate as the control plane for the volume, the instructions including at least instructions for executing a consensus protocol to form consensus on one or more settings related to the volume.

6. The system of claim 5, wherein the selection logic is based at least in part on a type of volume, one or more customer heuristics, one or more volume heuristics and/or a placement of the volume in a computing environment.

7. The system of claim 5, wherein the set of computer system processes execute the consensus protocol for replica control in a replicated database, wherein the consensus protocol is a quorum-based voting system scheme.

8. The system of claim 5, wherein the control plane and a data store operably interconnected with the control plane are located in a same fault zone as the volume.

9. The system of claim 8, wherein the system is further configured to assign at least one new computer process to the control plane when the subset of computer processes included in the control plane no longer maintains a quorum.

10. The system of claim 5, wherein each of the set of computer system processes is a member of more than one control plane at a same or different time.

11. The system of claim 5, wherein the one or more services are further configured to maintain an authoritative volume and a replicate volume in separate data zones in a same region.

12. The system of claim 5, the control plane is further configured to:

receive a notification that the volume was removed from a server;

instruct a computer processes of the control plane to dissolve the control plane;

instruct each of the computer processes to de-provision resources associated with the volume; and

instruct each of the computer processes to purge data related to the volume.

13. A non-transitory computer-readable storage medium having collectively stored thereon executable instructions that, if executed by one or more processors of a computer system, cause the computer system to:

receive a request to create a volume for a control plane;

receive a set of parameters for the control plane to be created for the volume;

determine, based at least in part on the set of parameters, a zone in which to construct the control plane for the volume;

select, based at least in part on the set of parameters, a subset of a set of control agents, the set of control agents able to participate in operation of the control plane for the volume including at least executing a consensus protocol to form consensus on one or more settings related to the volume; and

provide information indicating the subset thereby enabling the subset to be used to create the control plane for the volume.

14. The non-transitory computer-readable storage medium of claim 13, wherein the executable instructions further comprise instructions that cause the computer system to maintain the subset of the set of control agents on a same server as a data plane machine.

15. The non-transitory computer-readable storage medium of claim 13, wherein the executable instructions further comprise instructions that cause the computer system to:

dynamically configure a system type on a sliding scale between a distributed system with a quorum and a centralized system located in proximity relative to the volume based on business logic; and

update the system type at any point during existence of the volume.

16. The non-transitory computer-readable storage medium of claim 13, wherein the executable instructions further comprise instructions that cause the computer system to:

maintain multiple members in the subset of the set of control agents; and

instruct the subset of the set of control agents to execute the consensus protocol.

17. The non-transitory computer-readable storage medium of claim 13, wherein the executable instructions further comprise instructions that cause the computer system to determine the subset of the set of control agents based at least in part on a zone in which to construct the volume.

18. The non-transitory computer-readable storage medium of claim 17, wherein the executable instructions that cause the computer system to determine the subset of the set of control agents include instructions that cause the computer system to determine the subset of the set of control agents based at least in part on a power reliability factor in the zone, a concentration of data centers in the zone or a distance between data centers in the zone in which to construct the volume.

19. The non-transitory computer-readable storage medium of claim 13, wherein the executable instructions further comprise instructions that cause the computer system to maintain a control management component as an executive function control plane, wherein the control management component is configured to manage control planes across a network.

20. The non-transitory computer-readable storage medium of claim 19, wherein the executable instructions that cause the computer system to maintain the control management component further include instructions that cause the computer system to cause the control management component to:

receive an application programming interface (API) request specifying new parameters for the control planes; and

reselect a subset of control agents from the set of control agents to satisfy the new parameters.