Dynamically instantiating dual-queue systems

• US 9,838,467 B2
• Filed: 04/29/2015
• Issued: 12/05/2017
• Est. Priority Date: 03/17/2014
• Status: Active Grant

• Current Assignee: Splunk Inc. (Cisco Systems, Inc.)
• Original Assignee: Splunk Inc. (Cisco Systems, Inc.)
• Inventors: Vlachogiannis, Ioannis, Papadomitsos, Panagiotis
• Primary Examiner(s): Hussain, Imad
• Application Number:
  US14/699,984
• Publication Number:
  US 20150264152A1
• Time in Patent Office: 951 Days
• Field of Search: 709206, 709213
• CPC Class Codes: G06F 11/00 Error detection; Error correction; Monitoring error detection, correction or monitoring in information storage based on relative movement between record carrier and transducer G11B20/18; monitoring, i.e. supervising the progress of recording or reproducing G11B27/36; in static stores G11C29/00
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3442 for planning or managing the needed capacity
• CPC Class Codes: G06F 11/3452 Performance evaluation by statistical analysis
• CPC Class Codes: G06F 16/00 Information retrieval; Database structures therefor; File system structures therefor
• CPC Class Codes: G06F 2201/81 Threshold
• CPC Class Codes: G06F 2201/86 Event-based monitoring
• CPC Class Codes: G06F 2201/87 Monitoring of transactions
• CPC Class Codes: G06F 9/46 Multiprogramming arrangements
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46

Abstract

A computer-implemented method, system, and computer-readable media are disclosed herein. In embodiments, the computer-implemented method may entail receiving, by a data service, live data associated with an entity. The entity may be, for example, a customer of the data service. The method may further include determining that a dual-queue node assigned to the entity is uninstantiated on the data service. As a result, a dual-queue node associated with the entity may be instantiated on the data service. The dual-queue node may be instantiated by initializing a live data queue, of the dual-queue node, in which to place the live data for processing and a stale data queue, of the dual-queue node, in which to store a persistent backup of the live data. The method may then route the live data to the dual-queue node. The dual-queue node may then process the live data. Additional embodiments are described and/or claimed.

Claims

1. A computer-implemented method of implementing a multi-tenant dual-queue system comprising:

receiving, by a data service, live data associated with an entity;

based on determining that a dual-queue node assigned to the entity is uninstantiated on the data service, dynamically instantiating the dual-queue node assigned to the entity, by initializing a live data queue and a stale data queue for the dual-queue node, wherein the initialized live data queue is enabled to receive the live data for processing and the initialized stale data queue is enabled to store a persistent backup of the live data; and

routing the live data to the dual-queue node.

2. The computer-implemented method of claim 1,

wherein the entity is a customer of the data service, and

wherein determining that the dual-queue node assigned to the entity is uninstantiated on the data service includes determining whether an existing dual-queue node of the data service is assigned to the customer.

3. The computer-implemented method of claim 1,

wherein the entity is a customer of the data service, and

wherein the live data is received from an application associated with the customer of the data service.

4. The computer-implemented method of claim 1,

wherein the entity is a customer of the data service,

wherein the live data is received from an application associated with the customer, and

wherein the live data is related to usage or performance of the application.

5. The computer-implemented method of claim 1, wherein the entity is a source of the live data.

6. The computer-implemented method of claim 1,

wherein the entity is an Internet of things device;

and

wherein the live data is generated by the Internet of things device.

7. The computer-implemented method of claim 1,

wherein the entity is an Internet of things device;

and

wherein the live data is received from the Internet of things device.

8. The computer-implemented method of claim 1,

wherein the live data is associated with a customer of the data service and a type of data,

and

wherein the entity is based on a combination of the customer and the type of data.

9. The computer-implemented method of claim 1,

wherein determining that the dual-queue node assigned to the entity is uninstantiated on the data service includes determining whether an existing dual-queue node of the data service is assigned to the entity, and

wherein determining whether an existing dual-queue node of the data service is assigned to the entity includes analyzing a queue-entity mapping that correlates a plurality of existing dual-queue nodes of the data service with a corresponding plurality of entities.

10. The computer-implemented method of claim 1, wherein initializing the live data queue further comprises:

allocating a portion of memory to serve as the live data queue of the dual-queue node.

11. The computer-implemented method of claim 1, wherein initializing the stale data queue comprises:

identifying a partition of a persistent data store associated with the entity; and

assigning the partition to the dual-queue node to serve as the stale data queue.

12. The computer-implemented method of claim 1, wherein initializing the stale data queue comprises:

identifying a partition of a persistent data store associated with the entity, wherein the partition of the persistent data store contains previously received live data associated with the entity; and

assigning the partition to the dual-queue node to serve as the stale data queue.

13. The computer-implemented method of claim 1, further comprising:

processing the live data by the dual-queue node, wherein processing the live data by the dual-queue node comprises;

providing the live data to the live data queue for analysis of the live data, wherein the analysis includes performing one or more data operations on the live data from the live data queue.

14. The computer-implemented method of claim 1, further comprising:

processing the live data by the dual-queue node wherein processing the live data by the dual-queue node comprises;

providing the live data directly to the live data queue for analysis of the live data;

and

saving, by the dual-queue manager, the persistent backup of the live data to the stale data queue.

15. The computer-implemented method of claim 1, wherein dynamically instantiating a dual-queue node associated with the entity further comprises:

initializing a queue manager for the dual-queue node;

allocating a portion of memory to serve as the live data queue;

identifying a partition of a persistent data store associated with the entity;

associating the partition with the dual-queue node to serve as the stale data queue; and

processing the live data by the dual-queue node, wherein processing the live data by the dual-queue node comprises;

providing, by the dual-queue manager, the live data directly to the live data queue

for processing of the live data; and

saving, by the dual-queue manager, the persistent backup of the live data to the stale data queue.

16. The computer-implemented method of claim 1, wherein the entity is a first entity, the dual-queue node is a first dual-queue node, and the method further comprising:

receiving, by the data service, a request for data associated with a second entity;

determining that a second dual-queue node assigned to the second entity is uninstantiated on the data service, and that a partition of a previously instantiated dual-queue node includes data associated with the second entity;

based on a result of the determining, dynamically instantiating the second dual-queue node by;

allocating a portion of memory as a live data queue of the second dual-queue node, and

assigning the partition as a stale data queue of the second dual-queue node; and

loading a portion of the data associated with the second entity into the live data queue of the second dual-queue node from the stale data queue of the second dual-queue node for transmission of the portion of the data associated with the second entity to satisfy the request.

17. The computer-implemented method of claim 1, wherein the entity is a first entity, the dual-queue node is a first dual-queue node, and the method further comprising:

receiving, by the data service, a request for data associated with a second entity;

determining that a second dual-queue node assigned to the second entity is uninstantiated on the data service, and that a partition of a previously instantiated dual-queue node includes data associated with the second entity;

based on a result of the determining, dynamically instantiating the second dual-queue node by;

allocating a portion of memory as a live data queue of the second dual-queue node, and

assigning the partition as a stale data queue of the second dual-queue node;

loading a portion of the data associated with the second entity into the live data queue of the second dual-queue node from the stale data queue of the second dual-queue node for transmission of the portion of the data associated with the second entity to satisfy the request;

and

deleting the portion of the data from the stale data queue of the second dual-queue node in response to receiving acknowledgement from the entity that the portion of the data was received.

18. The computer-implemented method of claim 1, wherein the live data is first live data, the entity is a first entity, the dual-queue node is a first dual-queue node, and the method further comprising:

receiving, by the data service, a request for data associated with a second entity;

determining that a second dual-queue node assigned to the second entity is instantiated on the data service for processing of second live data; and

transmitting at least a portion of the second live data to the second entity to satisfy the request.

19. The computer-implemented method of claim 1, further comprising:

determining that the dual-queue node is inactive; and

in response to determining that the dual-queue node is inactive, terminating the dual-queue node.

20. The computer-implemented method of claim 1, further comprising:

determining that the dual-queue node is inactive based on a length of time that has passed since receiving live data associated with the entity; and

in response to determining that the dual-queue node is inactive, terminating the dual-queue node.

21. The computer-implemented method of claim 1, further comprising:

determining that the dual-queue node is inactive based on a length of time that has passed since receiving and satisfying a request for data from the dual-queue node; and

in response to determining that the dual-queue node is inactive terminating the dual-queue node.

22. The computer-implemented method of claim 1, further comprising:

determining that the dual-queue node is inactive based on a length of time that has passed since receiving the live data or any additional live data associated with the entity; and

in response to determining that the dual-queue node is inactive, terminating, the dual-queue node, wherein termination of the dual-queue node maintains a current state of data stored within the stale data queue.

23. The computer-implemented method of claim 1, wherein the live data is first live data, the entity is a first entity, the dual-queue node is a first dual-queue node, and the method further comprising:

receiving, by the data service, second live data associated with a second entity;

determining that a second dual-queue node assigned to the second entity is instantiated on the data service;

based on a result of the determining, routing the second live data to the second dual-queue node; and

processing the second live data by the second dual-queue node.

24. The computer-implemented method of claim 1, wherein dynamically instantiating the dual-queue node is accomplished on a multi-tenant server that supports other dual-queue nodes dynamically instantiated for a plurality of other entities.

25. The computer-implemented method of claim 1, wherein dynamically instantiating the dual-queue node is accomplished on a multi-tenant server that supports other dual-queue nodes dynamically instantiated for a plurality of other entities, the method further comprising:

receiving a request to migrate the dual-queue node from the multi-tenant server to a single tenant server; and

in response to receiving the request, forwarding the live data to the single tenant server along with any stale data that is disjoint from the live data.

26. The computer-implemented method of claim 1, wherein dynamically instantiating the dual-queue node is accomplished on a multi-tenant server that supports other dual-queue nodes dynamically instantiated for a plurality of other entities, the method further comprising:

in response to subscription of the entity to a single tenant server offered by the data service, receiving a request to migrate the dual-queue node from the multi-tenant server to a single tenant server;

in response to receiving the request, forwarding the live data to the single tenant server along with any stale data that is disjoint from the live data; and

terminating the dual-queue node on the multi-tenant server.

27. The computer-implemented method of claim 1, wherein a portion of the received live data is dropped prior to being processed due to instability of the system on which the dual-queue node is instantiated, the method further comprising:

based on the portion of received live data being dropped, recovering the portion of the received live data from the stale data queue; and

providing the recovered portion of the received live data to the live data queue for processing of the recovered portion of the received live data.

28. The computer-implemented method of claim 1, wherein a portion of the received live data is dropped prior to being processed due to instability of the system on which the dual-queue node is instantiated, the method further comprising:

based on the portion of received live data being dropped, recovering the portion of the received live data from the stale data queue; and

providing the recovered portion of the received live data to the live data queue for processing of the recovered portion of the received live data, wherein system instability includes crashes of the system on which the dual-queue node is instantiated.

29. A system for implementing a multi-tenant dual-queue system comprising:

one or more processors; and

one or more computer-readable storage media containing instructions which, in response to execution by the one or more processors, cause the one or more processors to;

receive live data associated with an entity;

based on determining that a dual-queue node assigned to the entity is uninstantiated on the data service, dynamically instantiate the dual-queue node assigned to the entity, by initializing a live data queue and a stale data queue of the dual-queue node, wherein the initialized live data queue is enabled to receive the live data for processing and the initialized stale data queue is enabled to store a persistent backup of the live data; and

send the live data to the dual-queue node to process the live data.

30. One or more non-transitory computer-storage media having executable instructions to implement a multi-tenant dual-queue system stored thereon, which, when executed by a computing device, cause the computing device to:

receive live data associated with an entity;

based on determining that a dual-queue node assigned to the entity is uninstantiated on the data service, dynamically instantiate the dual-queue node, assigned to the entity, by initializing a live data queue and a stale data queue of the dual-queue node, wherein the initialized live data queue is enabled to receive the live data for processing and the initialized stale data queue is enabled to store a persistent backup of the live data; and

send the live data to the dual-queue node to process the live data.