System and method of a managing multiple data centers

Abstract

A method and apparatus of a device that automates a task over a network that includes a plurality of data centers is described. In an exemplary embodiment, the device receives state change information from a plurality of devices in the plurality of data centers. In addition, the device generates an overall state for the network. The device further receives a task to be performed over the plurality of data centers. The device additionally schedules the task to be performed using the overall network state.

What is claimed is: 
     
         1 . A non-transitory machine-readable medium having executable instructions to cause one or more processing units to perform a method to automate a task over a network that includes a plurality of data centers, the method comprising:
 receiving state change information from a plurality of devices in the plurality of data centers;   generating an overall state for the network;   receiving a task to be performed over the plurality of data centers; and   scheduling the task to be performed using the overall network state.   
     
     
         2 . The non-transitory machine-readable medium of  claim 1 , wherein the overall network state includes characteristics selected from the group consisting of resource usage of each individual device in the network, resource usage of each of the plurality of data centers, configuration of each of the devices in the network, topology of each data center, topology of the network that includes the data centers, error counters, and fault indications. 
     
     
         3 . The non-transitory machine-readable medium of  claim 2 , wherein device is selected from the group consisting of a computer system, a network element, and a storage system. 
     
     
         4 . The non-transitory machine-readable medium of  claim 2 , wherein the scheduling comprises:
 determining the resource usage of each of the plurality of data centers; and   for each of the plurality of data centers,
 determining if the resource usage for that data center is sufficient to complete the task, 
 configuring other data centers of the plurality of data centers to handle operations of that data center, and 
 performing the task for that device of that data center. 
   
     
     
         5 . The non-transitory machine-readable medium of  claim 4 , wherein the configuring comprises:
 reconfiguring one or more devices in the other data centers to handle the operations of that data center.   
     
     
         6 . The non-transitory machine-readable medium of  claim 1 , wherein the task is selected from the group consisting of software upgrade/downgrade, bug vulnerability detection, and network performance monitoring. 
     
     
         7 . The non-transitory machine-readable medium of  claim 1 , wherein a data center is a site that include up to thousands of device managed by a synchronization controller, each of the devices in the data center reports state update changes to the synchronization controller, and the received state change information is received from the synchronization controller. 
     
     
         8 . The non-transitory machine-readable medium of  claim 7 , wherein the received state change information includes that state change updates except for forwarding state changes. 
     
     
         9 . The non-transitory machine-readable medium of  claim 1 , wherein the network further includes a branch office is a site including up to a hundred devices managed by an automation controller. 
     
     
         10 . A method to automate a task over a network that includes a plurality of data centers, the method comprising:
 receiving state change information from a plurality of devices in the plurality of data centers;   generating an overall state for the network;   receiving a task to be performed over the plurality of data centers; and   scheduling the task to be performed using the overall network state.