Abstract: The current document is directed to a modular reinforcement-learning-based application manager that can be deployed in various different computational environments without extensive modification and interface development. The currently disclosed modular reinforcement-learning-based application manager interfaces to observation and action adapters and metadata that provide a uniform and, in certain implementations, self-describing external interface to the various different computational environments which the modular reinforcement-learning-based application manager may be operated to control. In addition, certain implementations of the currently disclosed modular reinforcement-learning-based application manager interface to a user-specifiable reward-generation interface to allow the rewards that provide feedback from the computational environment to the modular reinforcement-learning-based application manager to be tailored to meet a variety of different user expectations and desired control policies.

Abstract: The current document is directed to a modular reinforcement-learning-based application manager that can be deployed in various different computational environments without extensive modification and interface development. The currently disclosed modular reinforcement-learning-based application manager interfaces to observation and action adapters and metadata that provide a uniform and, in certain implementations, self-describing external interface to the various different computational environments which the modular reinforcement-learning-based application manager may be operated to control. In addition, certain implementations of the currently disclosed modular reinforcement-learning-based application manager interface to a user-specifiable reward-generation interface to allow the rewards that provide feedback from the computational environment to the modular reinforcement-learning-based application manager to be tailored to meet a variety of different user expectations and desired control policies.

Abstract: Methods and apparatuses are presented for managing remote computers that are separated from their manager by a firewall. In one or more embodiments, the apparatus includes one or more computer sub-systems managed by the remote manager. Each sub-system includes a processor and a communication interface coupled to the processor, where the communication interface is further coupled to the remote manager via a network. A firewall is implemented between each sub-system and the remote manager such that the firewall blocks accesses to the sub-system from the remote manager. Notwithstanding the presence of the firewall, and without an specific routing through the firewall, each sub-system receives configuration commands from the remote manager through as a text feed.

Abstract: One particular implementation of the present invention may take the form of a method for correlating partial views of network resources into normalized or aggregated views for use by an administrator. The implementation may compare a newly detected resource against the known set of normalized resources of the network to correlate the new resource with an existing network resource. More particularly, one method for correlating network resources may compare key/value pairs associated with a new resource with key/value pairs associated with a first normalized resource. If the new resource does not correlate to the first normalized resource, the method may then compare the key/value pairs of the new resource to the key/value pairs of the second normalized resource. A second implementation may compare a unique key/value pair to a map containing the unique key/value pairs of each normalized resource, until the unique key/value pair is found.

Abstract: Methods and apparatuses are presented for managing remote computers. In one or more embodiments, the apparatus includes one or more management servers comprising a registry of manageable objects. The one or more management servers are capable of being executed on a first machine. One or more management applications are coupled to the one or more management servers. The one or more management applications are capable of being executed on a second machine, where the first and second machines are separate. A first cache connector is coupled to the one or more management servers, where the first cache connector retrieves and stores at least one of the manageable objects listed in the registry. A second cache connector is coupled to the one or more management applications, where the first and second cache connectors form a cache management connection. The processing required to implement the first cache connector is distributed across the first machine.

Abstract: One particular implementation of the present invention may take the form of a method for correlating partial views of network resources into normalized or aggregated views for use by an administrator. The implementation may compare a newly detected resource against the known set of normalized resources of the network to correlate the new resource with an existing network resource. More particularly, one method for correlating network resources may compare key/value pairs associated with a new resource with key/value pairs associated with a first normalized resource. If the new resource does not correlate to the first normalized resource, the method may then compare the key/value pairs of the new resource to the key/value pairs of the second normalized resource. A second implementation may compare a unique key/value pair to a map containing the unique key/value pairs of each normalized resource, until the unique key/value pair is found.