Distributed caching cluster configuration

• US 9,529,772 B1
• Filed: 11/26/2012
• Issued: 12/27/2016
• Est. Priority Date: 11/26/2012
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Shankaran, Nishanth, Arya, Rajat, Sbisa, Clint Joseph, Shou, Dong, Thiruchi Loganathan, Rajaprabhu, Krishnamoorthy, Shyam
• Primary Examiner(s): Joo, Joshua
• Application Number:
  US13/685,596
• Time in Patent Office: 1,492 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: H04L 41/0803 Configuration setting
• CPC Class Codes: H04L 41/0816 the condition being an adaptation, e.g. in response to network events
• CPC Class Codes: H04L 41/0853 by actively collecting configuration information or by backing up configuration information
• 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 67/568 Storing data temporarily at an intermediate stage, e.g. caching

Abstract

A cache or other cluster is configuration-aware such that initialization and changes to the underlying structure of the cluster can be dynamically updated. For example, a client may resolve an alias from a static configuration endpoint to a node in a cluster. The client may request initial configuration from the node and then update configuration from nodes that store the current configuration of the cluster.

Claims

1. A computer-implemented method for caching, comprising:

determining, by one or more computer systems, a current configuration of a cache cluster, the cache cluster comprising a set of memory caching nodes and subject to one or more cache rules, each of the set of memory caching nodes comprising a reserve memory space;

providing, by a static configuration endpoint of the cache cluster, the current configuration to a client device, the static configuration endpoint configured to receive a request from the client device and resolve a location of the current configuration on the first reserve memory space;

providing the current configuration of the cache cluster to each memory caching node of the set of memory caching nodes, the current configuration of the cache cluster residing in the reserve memory space;

detecting a change in the cache cluster;

determining an updated configuration of the cache cluster based at least in part on the change to the cache cluster; and

providing the updated configuration to the reserve memory space of each memory caching node of the set of memory caching nodes in the cache cluster for storage, each memory caching node of the set of memory caching nodes being configured to provide the updated configuration to a client, and the reserve memory space not being subject to the one or more cache rules applicable to the cache cluster.

2. The computer-implemented method of claim 1, wherein providing the updated configuration to the reserve memory space of each memory caching node of the set of memory caching nodes further comprises:

providing the updated configuration to a first memory caching node; and

propagating, by one or more memory caching nodes from the set of memory caching nodes, the updated configuration to other memory caching nodes identified in the updated configuration, the one or more memory caching nodes comprising the first memory caching node.

3. The computer-implemented method of claim 1, wherein the static configuration endpoint provides an alias to at least one of the memory caching nodes of the set of memory caching nodes.

4. The computer-implemented method of claim 1, wherein the change is made in response to a request to change routing.

5. The computer-implemented method of claim 1, wherein the one or more cache rules includes at least one or more protocols for eviction of data or expiration of the cache memory.

6. The computer-implemented method for caching of claim 1, further comprising:

obtaining, using the current configuration, a second configuration from a particular memory caching node of the set of memory caching nodes;

verifying the current configuration by comparing times associated with the current configuration and the second configuration;

determining an additional updated configuration of the changed cache cluster based at least in part on the verifying of the current configuration against the second configuration; and

providing the additional updated configuration to the reserve memory space of each memory caching node of the set of memory caching nodes.

7. A computer-implemented method for caching, comprising:

provisioning, by one or more computer systems, a first storage node, the first storage node comprising a first reserve memory space;

provisioning a second storage node, the second storage node comprising a second reserve memory space;

determining a configuration of a cluster, the cluster comprising the first storage node and the second storage node and subject to one or more cache rules; and

sending, by a static configuration endpoint of the cluster, the configuration to a client device, the static configuration endpoint configured to receive a request from the client device and resolve a location of the configuration on the first reserve memory space;

sending the configuration to the first storage node and the second storage node for storage in the first reserve memory space and the second reserve memory space, the first reserve memory space and the second reserve memory space not being subject to the one or more cache rules applicable to the cluster.

8. The computer-implemented method of claim 7, wherein the first storage node and the second storage node are memory caching nodes.

9. The computer-implemented method of claim 7, further comprising:

provisioning a third storage node into the cluster, the third storage node comprising a third reserve memory space;

determining an updated configuration of the cluster; and

sending the updated configuration to the first storage node for storing in the first reserved memory space, the second storage node for storage in the second memory space, and the third storage node for storage in the third reserve memory space, the third reserve memory space not being subject to the one or more cache rules applicable to the cluster.

10. The computer-implemented method of claim 7, further comprising:

removing the second storage node from the cluster;

determining an updated configuration of the cluster; and

sending the updated configuration to the first storage node for storage in the first reserved memory space.

11. The computer-implemented method of claim 7, wherein the first storage node is configured to receive a request for cached data from a client and determine whether the requested data is stored within the first storage node.

12. The computer-implemented method of claim 11, wherein the first storage node determines, based at least in part on the configuration stored in the first reserve memory space, that the requested data is stored in the second storage node, and forwards the request to the second storage node.

13. The computer-implemented method of claim 7, wherein the first storage node, the second storage node, and a third storage node individually store a key space, and wherein the key space is repartitioned among the first storage node, the second storage node, and the third storage node based at least in part on the provisioning of the third storage node.

14. The computer-implemented method of claim 7, wherein, upon receiving an updated configuration, the first storage node pushes the updated configuration to the client device.

15. The computer-implemented method of claim 7, further comprising monitoring health of the first storage node and the second storage node, wherein monitoring the health comprises one or more of detecting a failure or determining that the first storage node or the second storage node are to be repaired.

16. The computer-implemented method of claim 15, wherein repairing comprises replacing the first storage node or the second storage node.

17. The computer-implemented method of claim 15 wherein repairing comprises:

determining that the first storage node is to be replaced;

provisioning a replacement storage node, the replacement storage node comprising a replacement reserved memory space;

removing the first storage node from the cluster;

adding the replacement storage node to the cluster;

determining an updated configuration of the cluster; and

sending the updated configuration for storage in the second reserved memory space and the replacement reserved memory space.

18. The computer-implemented method of claim 7, further comprising receiving a client request for an updated configuration or cached data, wherein the first storage node is configured to respond with the updated configuration.

19. One or more non-transitory computer-readable storage media having collectively stored thereon executable instructions that, when executed by one or more processors of a computer system, cause the computer system to at least:

provision a first memory caching node;

provision a second memory caching node;

determine a configuration of a cache cluster, the cache cluster comprising the first memory caching node and the second memory caching node, the cache cluster subject to one or more cache rules;

send, by a static configuration endpoint of the cache cluster, the configuration to a client device, the static configuration endpoint configured to receive a request from the client device and resolve a location of the configuration of the first memory caching node of the set of memory caching nodes;

send the configuration to the first memory caching node and the second memory caching node for storage in a reserved memory space of each of the first memory caching node and the second memory caching node, the reserve memory space not being subject to the one or more cache rules applicable to the cache cluster.

20. The one or more non-transitory computer-readable storage media of claim 19, wherein the instructions further comprise instructions that, when executed:

provide a hostname; and

resolve the hostname through a domain name server for storage to the first memory caching node.

21. The one or more non-transitory computer-readable storage media of claim 20, wherein providing the hostname further comprises providing the hostname as an alias to the first memory caching node.

22. The one or more non-transitory computer-readable storage media of claim 19, wherein the instructions further comprise instructions that, when executed provide the static configuration endpoint at a static location.

23. The one or more non-transitory computer-readable storage media of claim 22, wherein the instructions further comprise instructions that, when executed, cause the computer system to at least:

remove the first memory caching node from the cache cluster;

select the second memory caching node; and

map the static configuration endpoint to the second memory caching node.

24. A computer system, comprising:

one or more processors; and

memory including instructions that, when executed by the one or more processors, cause the computer system to;

receive, by a memory caching node in a cache cluster, a request to store a configuration of the cache cluster, the cache cluster subject to one or more cache rules;

store the configuration in a reserved memory space, the reserve memory space not being subject to the one or more cache rules applicable to the cache cluster;

send, by a static configuration endpoint of the cache cluster, the configuration to a client device in response to a request for the configuration, the static configuration endpoint configured to receive the request from the client device and resolve a location of the configuration on the reserve memory space.

25. The computer system of claim 24, wherein the computer system is a memory caching node in the cache cluster, and the instructions, when executed by the one or more processors, cause the computer system to respond to one or more requests for data stored in the cache cluster, the cache cluster storing data from a data source.

26. The computer system of claim 24, wherein the configuration includes connection information for each memory caching node in the cache cluster.

27. The computer system of claim 24, wherein the computer system is a memory caching node in a cache cluster and the instructions, when executed by the one or more processors, further cause the computer system to propagate the configuration to other memory caching nodes within the cache cluster.