Managing a network system

1. A method for managing a network system, the method comprising:

collecting, by a processor, metrics for a plurality of nodes in the network system;

determining, by the processor, a plurality of zones including the plurality of nodes based on the metrics for the network system; and

for each zone of the plurality of zones, determining a computational architecture to be implemented for the zone based on the metrics for each node of the plurality of nodes in the zone,

wherein the computational architecture to be implemented for each zone comprises one of a single-hierarchy architecture, a multi-hierarchy architecture, and a peer-to-peer architecture, and

wherein at least some of the zones of the plurality of zones implement different computational architectures simultaneously.

2. The method of claim 1 , further comprising:

continuing to collect the metrics after implementing the computational architecture for each of the plurality of zones; and

modifying the computational architecture for at least one of the plurality of zones based on the continued collected metrics for the at least one of the plurality of zones.

3. The method of claim 2 , further comprising:

modifying a boundary of at least one of the plurality of zones based on the continued collected metrics.

4. The method of claim 2 , further comprising:

applying a management policy based on the continued collected metrics after modifying the computational architecture.

5. The method of claim 1 , further comprising:

constructing a computation-communication graph including the plurality of zones; and

managing the network system based on the computation-communication graph.

6. The method of claim 1 , wherein the nodes in each one of the plurality of zones include a plurality of monitoring brokers and a plurality of data collectors, and wherein each of the plurality of monitoring brokers receives the metrics from each of the plurality of data collectors.

7. The method of claim 6 , wherein each of the plurality of monitoring brokers communicates with other nodes in the zone via an identifier associated with each of the plurality of nodes, and wherein the plurality of monitoring brokers are included in the computational architecture for each zone of the plurality of zones.

8. The method of claim 7 , further comprising:

determining a zone leader of each zone of the plurality of zones based on a predetermined zone leader election process, wherein the zone leader of each zone of the plurality of zones communicates with other zone leaders.

9. The method of claim 8 , further comprising:

determining a leader of zone leaders of the plurality of zones, wherein the leader of zone leaders communicates with other zone leaders of the plurality of zones.

10. The method of claim 9 , further comprising:

for the plurality of zones, determining the computational architecture to be implemented for the zone leaders of the plurality of zones based on the metrics for the zone leaders of the plurality of zones.

11. The method of claim 9 , further comprising:

sending the metrics to the leader of the zone leaders of the plurality of zones; and

modifying the computational architecture for at least one of the plurality of zones based on the metrics.

12. The method of claim 6 , further comprising:

aggregating the metrics at each of the plurality of monitoring brokers in response to a request of the network system;

analyzing the metrics at each of the plurality of monitoring brokers, wherein each of the plurality of monitoring brokers communicates each other; and

determining an anomalous behavior of the network system based on the metrics.

13. The method of claim 12 , further comprising:

sending the metrics from the plurality of monitoring brokers to the zone leader.

14. A computer system to manage a network system, the computer system comprising:

a processor in communication with a memory storing instructions that when executed cause the processor to:

collect metrics for a plurality of nodes in the network system;

determine a plurality of zones including the plurality of nodes based on the metrics for the network system; and

for each zone of the plurality of zones, determine a computational architecture to be implemented for the zone based on the metrics for each node of the plurality of nodes in the zone,

wherein the computational architecture to be implemented for each zone comprises one of a single-hierarchy architecture, a multi-hierarchy architecture, and a peer-to-peer architecture, and

wherein at least some of the zones of the plurality of zones implement different computational architectures simultaneously; and

a data storage device storing the metrics and information for the plurality of zones.

15. The computer system of claim 14 , wherein the processor is to continue to collect the metrics after implementing the computational architecture for each of the plurality of zones, to modify the computational architecture for at least one of the plurality of zones based on the continued collected metrics for the at least one of the plurality of zones, and to modify a boundary of at least one of the plurality of zones based on the continued collected metrics.

16. The computer system of claim 15 , wherein the nodes in each one of the plurality of zones include a plurality of monitoring brokers and a plurality of data collectors, and

wherein each of the plurality of monitoring brokers receives the metrics from each of the plurality of data collectors, and

wherein each of the plurality of monitoring brokers communicates with other nodes in the zone via an identifier associated with each of the plurality of nodes, and

wherein the plurality of monitoring brokers are included in the computational architecture for each zone of the plurality of zones, and

wherein the processor is configured to determine a zone leader of each zone of the plurality of zones based on a predetermined zone leader election process, to analyze the metrics at each of the plurality of monitoring brokers, to determine an anomalous behavior of the network system based on the metrics, and to send the metrics from the plurality of monitoring brokers to the zone leader, wherein the zone leader of each zone of the plurality of zones communicates with other zone leaders.

17. A non-transitory computer readable storage medium storing software that when executed by a processor performs a method for managing a network system, the method comprising:

collecting metrics for a plurality of nodes in the network system;

determining a plurality of zones including the plurality of nodes based on the metrics for the network system; and

for each zone of the plurality of zones, determining a computational architecture to be implemented for the zone based on the metrics for each node of the plurality of nodes in the zone,

wherein the computational architecture to be implemented for each zone comprises one of a single-hierarchy architecture, a multi-hierarchy architecture, and a peer-to-peer architecture, and

wherein at least some of the zones of the plurality of zones implement different computational architectures simultaneously.