Replication group pools for fast provisioning

• US 10,452,681 B1
• Filed: 11/30/2015
• Issued: 10/22/2019
• Est. Priority Date: 11/30/2015
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Chen, Tao, Parthasarathy, Madhuvanesh, Agrawal, Shobha, Ping, Fan, Jain, Divya Ashok Kumar, Brooker, Marc John, Zulak, Matthew Joseph, Pallikila, Lakshmi N.
• Primary Examiner(s): Trujillo, James
• Assistant Examiner(s): Shechtman, Cheryl M
• Application Number:
  US14/954,928
• Time in Patent Office: 1,422 Days
• Field of Search: None
• CPC Class Codes: G06F 16/23 Updating
• CPC Class Codes: G06F 16/27 Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
• CPC Class Codes: G06F 16/955 using information identifiers, e.g. uniform resource locators [URL]

Abstract

Data replication groups may be used to store data in a distributed computing environment. The data replication groups may include a set of nodes executing a consensus protocol to maintain data durably. In order to increase the speed of provisioning, a pool of data replication groups may be maintained. When a request is received a data replication group may be selected from the pool, removed from the pool, and associated with the request. Furthermore, a replacement data replication group may be provisioned and added to the pool.

Claims

1. A computer-implemented method, comprising:

generating a pool of data replication groups, where the data replication groups of the pool comprise a plurality of nodes suitable to execute a consensus protocol, by at least;

identifying a group of computer systems for placement of the plurality of nodes of the data replication groups;

initializing the plurality of nodes on the identified group of computer systems;

causing the plurality of nodes to execute the consensus protocol; and

storing pool information indicating that at least a portion of the data replication groups of the pool are ready to receive traffic, as a result of validating at least a portion of the plurality of nodes, where the pool information enables one or more computer systems to determine that the portion of the data replication groups can be used to fulfill requests;

receiving a request to utilize computing resources of the pool of data replication groups to store data;

selecting, based at least in part on the pool information, a particular data replication group of the pool to fulfill the request to utilize computing resources of the pool of data replication groups to store data;

associating the selected particular data replication group with a requestor responsible for submitting the request;

removing the selected particular data replication group from the pool; and

executing a process for replacing, in the pool, at least the selected particular data replication group that was removed from the pool.

2. The computer-implemented method of claim 1, wherein identifying the group of computer systems for placement of the plurality of nodes of the data replication groups of the pool further comprises transmitting a request to a metrics service for information identifying the group of computer systems.

3. The computer-implemented method of claim 2, wherein the request to the metrics service further comprises one or more constraints on the placement of the plurality of nodes of the data replication groups of the pool.

4. The computer-implemented method of claim 3, wherein the one or more constraints further comprise at least one constraint that causes the metrics service to identify the group of computer systems for placement of at least a portion of the plurality of nodes of the data replication groups of the pool such that the portion of the nodes are hosted by at least two distinct server computer systems.

5. A system, comprising:

one or more processors; and

memory that includes instructions that in response to execution by the one or more processors, cause the system to at least;

initialize a pool of data replication groups, where the pool of data replication groups contains at least a first number of data replication-groups groups, by at least;

determining a group of computer systems to host a plurality of nodes for the first number of data replication groups;

provisioning the plurality of nodes on the determined group of computer systems; and

executing a consensus protocol between at least a portion of the plurality of nodes;

receive a request for a data replication group;

fulfill the request, by at least, selecting a particular data replication group of the pool of data replication groups;

remove the selected particular data replication group from the pool;

determine that a number of data replication groups in the pool is less than the first number of data replication groups; and

add a replacement data replication group to the pool of data replication groups, the replacement data replication group is suitable to execute the consensus protocol.

6. The system of claim 5, wherein the memory further includes instructions that, in response to execution by the one or more processors, cause the system to determine a second number of data replication groups to include in the pool of data replication groups based at least in part on a rate at which requests for data replication groups are received by the system.

7. The system of claim 5, wherein the instructions that, in response to execution by the one or more processors, cause the system to add the replacement data replication group to the pool of data replication groups further comprise instructions that cause the system to add the replacement data replication group based at least in part on a rate at which requests for data replication groups are received by the system.

8. The system of claim 5, wherein selecting the particular data replication group of the pool of data replication groups further comprises selecting the particular data replication group based at least in part on one or more attributes of the particular data replication group.

9. The system of claim 8, wherein the one or more attributes of the particular data replication group include a number of nodes of the data replication group.

10. The system of claim 8, wherein the one or more attributes of the particular data replication group include an indication that the particular data replication group has been validated.

11. The system of claim 5, wherein the consensus protocol further comprises a Paxos consensus protocol.

12. The system of claim 5, wherein the instructions that, in response to execution by the one or more processors, cause the system to determine the group of computer systems to host the plurality of nodes further include instructions that cause the system to determine the group of computer systems based at least in part on one or more constraints on placement of the plurality of nodes on the group of computer systems.

13. A non-transitory computer-readable storage medium comprising executable instructions that, in response to execution by one or more processors of a computer system, cause the computer system to at least:

provision a first number of data replication groups to be included in a pool of data replication groups, the first number of data replication groups comprising a plurality of nodes suitable to implement a consensus protocol, by at least;

identifying a group of computer systems to host the plurality of nodes;

initializing the plurality of nodes on the identified group of computer systems; and

causing the plurality of nodes to execute the consensus protocol;

receive a request for a data replication group;

select a particular data replication group of the pool of data replication groups to fulfill the request;

remove the selected particular data replication group from the pool of data replication groups; and

add new data replication groups to the pool of data replication groups to at least replace the selected particular data replication group, such that a number of data replication groups in the pool of data replication groups is equal to or greater than the first number of data replication groups to be included in the pool of data replication groups.

14. The non-transitory computer-readable storage medium of claim 13, wherein the instructions that cause the computer system to fulfill the request further include instructions that cause the computer system to validate health information of the selected particular data replication group.

15. The non-transitory computer-readable storage medium of claim 13, wherein the instructions that cause the computer system to select the particular data replication group further include instructions that cause the computer system to select the particular data replication group based at least in part on a number of nodes of the particular data replication group.

16. The non-transitory computer-readable storage medium of claim 13, wherein the instructions that cause the computer system to associate the selected particular data replication group with the request further include instructions that cause the computer system to enable the selected particular data replication group to receive traffic on behalf of a requestor associated with the request.

17. The non-transitory computer-readable storage medium of claim 13, wherein the instructions that cause the computer system to associate the selected particular data replication group with the request further include instructions that cause the computer system to enable the selected particular data replication group to store state information of a computing system operated by a requestor associated with the request.

18. The non-transitory computer-readable storage medium of claim 13, wherein the instructions that cause the computer system to identify the group of computer systems further include instructions that cause the computer system to identify the group of computer systems based at least in part on one or more constraints.

19. The non-transitory computer-readable storage medium of claim 13, wherein the instructions that cause the computer system to provision the first number of data replication groups further include instructions that cause the computer system to determine a second number of data replication groups to include in the pool of data replication groups based at least in part on a rate at which requests for data replication groups are received by the computer system.

20. The non-transitory computer-readable storage medium of claim 13, wherein the consensus protocol is a Paxos consensus protocol.