Systems and methods for managing sub-clusters within a multi-cluster computing system subsequent to a network-partition event
A computer-implemented method for managing sub-clusters within a multi-cluster computing system subsequent to a network-partition event may include: 1) identifying a plurality of computer clusters, 2) identifying a network-partition event that partitions each of the plurality of computer clusters into multiple sub-clusters, 3) identifying, within the multiple sub-clusters, a plurality of nodes that share a common attribute or property, and then 4) ensuring that only those nodes that share the common attribute survive subsequent to the network-partition event. Various other methods, systems, and computer-readable media are also disclosed.
1. A computer-implemented method for managing sub-clusters within a multi-cluster computing system subsequent to a network-partition event, at least a portion of the method being performed by a computing device comprising at least one processor, the method comprising:
identifying a multi-cluster computing system comprising a plurality of computer clusters, each of the plurality of computer clusters within the multi-cluster computing system comprising a plurality of nodes;
identifying a network-partition event that partitions each of the plurality of computer clusters within the multi-cluster computing system into multiple sub-clusters;
identifying, within the multiple sub-clusters of the plurality of computer clusters, a plurality of nodes that share a common attribute;
ensuring that only those nodes that share the common attribute survive subsequent to the network-partition event.
2. The method of claim 1 , wherein identifying the plurality of nodes that share the common attribute comprises:
identifying a node that is the first to contact a coordination point subsequent to the network-partition event;
identifying at least one attribute of the node;
identifying at least one additional node that possesses the attribute of the node.
3. The method of claim 2 , wherein the node that is the first to contact the coordination point subsequent to the network-partition event comprises a proxy node that is configured to contact the coordination point on behalf of all other nodes that share the common attribute.
4. The method of claim 2 , further comprising at least one of:
preventing at least one node that lacks the common attribute from contacting the coordination point;
preventing at least one node that lacks the common attribute from contacting the coordination point until completion of a pre-defined delay period.
5. The method of claim 1 , wherein ensuring that only those nodes that share the common attribute survive comprises:
allowing those nodes that share the common attribute to perform tasks on behalf of the plurality of computer clusters;
preventing nodes that lack the common attribute from performing tasks on behalf of the plurality of computer clusters.
6. The method of claim 1 , wherein the network-partition event comprises at least one communication failure that prevents sub-clusters within each of the plurality or computer clusters from communicating with one another.
7. The method of claim 6 , wherein the communication failure is caused by at least one of:
a node failure;
a communication-infrastructure failure.
8. The method of claim 1 , wherein each of the plurality of computer clusters comprises at least one of:
a plurality of virtual machines that are programmed to execute on at least one physical machine;
a plurality of physical machines.
9. The method of claim 1 , wherein the common attribute comprises at least one of:
a physical location of the nodes;
a virtual location of the nodes;
at least one application hosted by the nodes.
10. A system for managing sub-clusters within a multi-cluster computing system subsequent to a network-partition event, the system comprising:
an identification module programmed to:
identify a multi-cluster computing system comprising a plurality of computer clusters, each of the plurality of computer clusters within the multi-cluster computing system comprising a plurality of nodes;
identify a network-partition event that partitions each of the plurality of computer clusters within the multi-cluster computing system into multiple sub-clusters;
identify, within the multiple sub-clusters of the plurality of computer clusters, a plurality of nodes that share a common attribute;
a node-survival module programmed to ensure that only those nodes that share the common attribute survive subsequent to the network-partition event;
at least one processor configured to execute the identification module and the node-survival module.
11. The system of claim 10 , wherein the identification module identifies the plurality of nodes that share the common attribute by:
identifying a node that is the first to contact a coordination point subsequent to the network-partition event;
identifying at least one attribute of the node;
identifying at least one additional node that possesses the attribute of the node.
12. The system of claim 11 , wherein the node that is the first to contact the coordination point subsequent to the network-partition event comprises a proxy node that is configured to contact the coordination point on behalf of all other nodes that share the common attribute.
13. The system of claim 11 , wherein the node-survival module is further programmed to:
prevent at least one node that lacks the common attribute from contacting the coordination point;
prevent at least one node that lacks the common attribute from contacting the coordination point until completion of a pre-defined delay period.
14. The system of claim 10 , wherein the node-survival module ensures that only those nodes that share the common attribute survive subsequent to the network-partition event by:
allowing those nodes that share the common attribute to perform tasks on behalf of the plurality of computer clusters;
preventing nodes that lack the common attribute from performing tasks on behalf of the plurality of computer clusters.
15. The system of claim 10 , wherein the network-partition event comprises at least one communication failure that prevents sub-clusters within each of the plurality of computer clusters from communicating with one another.
16. The system of claim 15 , wherein the communication failure is caused by at least one of:
a node failure;
a communication-infrastructure failure.
17. The system of claim 10 , wherein each of the plurality of computer clusters comprises at least one of:
a plurality of virtual machines that are programmed to execute on at least one physical machine;
a plurality of physical machines.
18. The system of claim 10 , wherein the common attribute comprises at least one of:
a physical location of the nodes;
a virtual location of the nodes;
at least one application hosted by the nodes.
19. A computer-readable storage medium comprising one or more computer-executable instructions that, when executed by at least one processor of a computing device, cause the computing device to:
identify a multi-cluster computing system comprising a plurality of computer clusters, each of the plurality of computer clusters within the multi-cluster computing system comprising a plurality of nodes;
identify a network-partition event that partitions each of the plurality of computer clusters within the multi-cluster computing system into multiple sub-clusters;
identify, within the multiple sub-clusters of the plurality of computer clusters, a plurality of nodes that share a common attribute;
ensure that only those nodes that share the common attribute survive subsequent to the network-partition event.
20. The computer-readable storage medium of claim 19 , wherein the one or more computer-executable instructions, when executed by the processor of the computing device, further cause the computing device to:
allow those nodes that share the common attribute to perform tasks on behalf of the plurality of computer clusters;
prevent nodes that lack the common attribute from performing tasks on behalf of the plurality of computer clusters.