Scalable leadership election in a multi-processing computing environment

Patent metadata

Field	Value
Publication number	US-11218418-B2
Application number	US-201615160347-A
Country	US
Kind code	B2
Filing date	May 20, 2016
Priority date	May 20, 2016
Publication date	Jan 4, 2022
Grant date	Jan 4, 2022

Abstract

A scalable multi-cluster resource sharing facility. A global witness process runs on a first computing platform that communicates over one or more networks to any number of nodes situated over two or more clusters. The global witness process listens on the network for occurrences of leadership and/or resource requests from nodes of different clusters. The global witness processes a request by retrieving a resource request and a respective last known state value, comparing the last known state value to a global stored state value, then storing a new state value when the respective last known state value is equal to the stored state value. Any number of contemporaneous requests can be processed by the global witness process, however only one request can be granted. The other requestors each receive a rejection of their resource request when their proffered last known state value is not equal to the stored state value.

First claim

What is claimed is:

1. A method comprising:
operating a global witness process at a third node that communicates over a network to at least a first node and a second node in different clusters, wherein the global witness process can be checked by both the first and second nodes to determine whether either the first node or the second node becomes an owner of a resource;
managing ownership of the resource shared across the different clusters using the global witness process, wherein the first node in a first cluster was previously established as the owner of the resource using the global witness process, and the second node subsequently determines that it is unable to contact the first node, managing ownership by:
receiving, at the global witness process on the third node, a request from the second node in a second cluster to become a new owner of the resource;
transmitting an authorization from the third node to the second node granting ownership of the resource to the second node; and
updating an ownership state for the resource at the third node; and

sending, in response to another request from another node to become the owner of the resource to the global witness process at the third node, a negative acknowledgement to the another node indicating that the another node is a follower.

2. The method of claim 1 , wherein the resource to be accessed corresponds to a leadership relationship.

3. The method of claim 1 , wherein multiple requestors contact the global witness process to establish a leadership relationship.

4. The method of claim 1 , wherein the another node is in a cluster different from the first node and the second node.

5. The method of claim 1 , wherein the second node is a disaster recovery change transmitter.

6. The method of claim 1 , wherein at least one follower is a disaster recovery change listener.

7. The method of claim 1 , wherein at least one of a set of followers perform a liveness measure check and the second node sends a periodic liveness signal.

8. The method of claim 1 , further comprising, comparing a last known state value to a stored state value that is stored in a global database comprising a plurality of state tables or a plurality of state values, and wherein the global database maintains ownership state values for a plurality of shared resources.

9. A non-transitory computer readable medium having stored thereon a sequence of instructions which, when executed by a processor performs a set of acts comprising:
operating a global witness process at a third node that communicates over a network to at least a first node and a second node in different clusters, wherein the global witness process can be checked by both the first and second nodes to determine whether either the first node or the second node becomes an owner of a resource;
managing ownership of the resource shared across the different clusters using the global witness process, wherein the first node in a first cluster was previously established as the owner of the resource using the global witness process, and the second node subsequently determines that it is unable to contact the first node, managing ownership by:
receiving, at the global witness process on the third node, a request from the second node in a second cluster to become a new owner of the resource;
transmitting an authorization from the third node to the second node granting ownership of the resource to the second node; and
updating an ownership state for the resource at the third node; and

sending, in response to another request from another node to become the owner of the resource to the global witness process at the third node, a negative acknowledgement to the another node indicating that the another node is a follower.

10. The non-transitory computer readable medium of claim 9 , wherein the resource to be accessed corresponds to a leadership relationship.

11. The non-transitory computer readable medium of claim 9 , wherein multiple requestors contact the global witness process to establish a leadership relationship.

12. The non-transitory computer readable medium of claim 9 , wherein the another node is in a cluster different from the first node and the second node.

13. The non-transitory computer readable medium of claim 9 , wherein the second node is a disaster recovery change transmitter.

14. The non-transitory computer readable medium of claim 9 , wherein at least one follower is a disaster recovery change listener.

15. The non-transitory computer readable medium of claim 9 , the set of acts further comprising comparing a last known state value to a stored state value that is stored in a global database comprising a plurality of state tables or a plurality of state values, and wherein the global database maintains ownership state values for a plurality of shared resources.

16. The non-transitory computer readable medium of claim 12 , wherein at least one of a set of followers perform a liveness measure check and the second node sends a periodic liveness signal.

17. A system comprising:
a non-transitory storage medium having stored thereon a sequence of instructions; and
a processor that executes the sequence of instructions to perform a set of acts comprising:
operating a global witness process at a third node that communicates over a network to at least a first node and a second node in different clusters, wherein the global witness process can be checked by both the first and second nodes to determine whether either the first node or the second node becomes an owner of a resource;
managing ownership of the resource shared across the different clusters using the global witness process, wherein the first node in a first cluster was previously established as the owner of the resource using the global witness process, and the second node subsequently determines that it is unable to contact the first node, managing ownership by:
receiving, at the global witness process on the third node, a request from the second node in a second cluster to become a new owner of the resource;
transmitting an authorization from the third node to the second node granting ownership of the resource to the second node; and
updating an ownership state for the resource at the third node; and

sending, in response to another request from another node to become the owner of the resource to the global witness process at the third node, a negative acknowledgement to the another node indicating that the another node is a follower.

18. The system of claim 17 , wherein the resource to be accessed corresponds to a leadership relationship.

19. The system of claim 17 , wherein multiple requestors contact the global witness process to establish a leadership relationship.

20. The system of claim 17 , wherein the second node is a disaster recovery change transmitter.

21. The system of claim 17 , the set of acts further comprising comparing a last known state value to a stored state value that is stored in a global database comprising a plurality of state tables or a plurality of state values, and wherein the global database maintains ownership state values for a plurality of shared resources.

22. The system of claim 17 , wherein the another node is in a cluster different from the first node and the second node.

23. The system of claim 17 , wherein at least one follower is a disaster recovery change listener.

24. The system of claim 17 , wherein at least one of a set of followers perform a

Assignees

• Nutanix Inc

Inventors

• Gupta Karan
• Konka Pavan Kumar
• Ramachandran Parthasarathy
• Olappamanna Vasudevan Arun
• Pyatkov Andrey Alexseyevich

Classifications

• H04L47/70 Primary

  Admission control; Resource allocation · CPC title

• H04L69/40 Primary

  for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection (management of faults, events, alarms or notifications in data switching networks H04L41/06) · CPC title

• H04L67/10

  in which an application is distributed across nodes in the network (software deployment G06F8/60; multiprogramming arrangements G06F9/46) · CPC title

• H04L67/42

  Electricity · mapped topic