Abstract: A request to create a set of security policies for an application is received at a graphical user interface. Information identifying a set of source VMs, a set of destination VMs, and a set of target VMs also are received, wherein the target VMs are executing the application and are supported by (a) node(s) in a clustered virtualization environment. A set of inbound rules identifying (a) category(ies) of source VMs permitted to initiate connections with a subset of target VMs and a set of outbound rules identifying (a) category(ies) of destination VMs to which the subset of target VMs are permitted to initiate connections are received at the graphical user interface. Upon receiving a request to apply the security policies, the policies are configured based at least in part on the inbound and outbound rules and a visual representation of the security policies is presented in the graphical user interface.

Abstract: Disclosed herein are related to a method, a system, and a non-transitory computer readable medium for backing up a virtual machine. In one aspect, an instruction to store a requested state of a virtual machine is received. Configurations of the virtual machine in the requested state are determined, and a configuration data indicative of the determined configurations of the virtual machine is generated. The configuration data is stored. Another instruction to load the virtual machine in the requested state from a plurality of states of the virtual machine may be received. The stored configuration data indicative of the configurations of the virtual machine in the requested state may be retrieved. The configurations of the virtual machine in the requested state may be identified based on the retrieved configuration data, and the virtual machine in the requested state may be loaded according to the identified configurations.

Abstract: Disclosed herein are related to a method, a system, and a non-transitory computer readable medium for providing a consolidated cloud system. The consolidated cloud system includes a plurality of cloud systems. A user request to access content in the consolidated cloud system is received, where the content is stored by a target cloud system of the plurality of cloud systems. The content from the target cloud system is retrieved, and the content is provided as part of the consolidated cloud system. The device generating the user request may be unaware of which cloud system stores the content.

Abstract: A method of eliminating false positives when testing a device management platform for controlling endpoints. The method includes sending, by a computing device, a request to a cloud controller to create an entity. The request causes operations including i) creating, by an endpoint responsive to a receipt of an interface request from the cloud controller, the entity, ii) receiving, by the cloud controller from the endpoint, response data generated by the entity, the response data indicating the creation of the entity, and iii) generating, by the cloud controller, control data based on the response data. The method includes receiving the control data from the cloud controller; sending, responsive to the receipt of the control data, a call to the endpoint for entity data; receiving entity data from the endpoint; comparing, the response data to the entity data to validate the response data; and generating a test status.

Abstract: A system and method include providing backup and restore capability to a cluster node. The cluster node includes a controller virtual machine (CVM) that is communicably coupled to a plurality of cloud platforms and provides the capability of carrying out backup and restore at the cloud platforms without having to run a CVM at the cloud platform. The CVM can backup objects to the cloud platform and store metadata information related to the object in storage. The CVM also takes snapshots of the object and stores the snapshots. The CVM also determines a changed data chunks that include the changes made to the object from the last snapshot. To restore the object, the CVM restores the object based, in part, on the snapshots, the changed data chunks, and the metadata stored at the node.

Abstract: A request to create a set of security policies for an application is received at a graphical user interface. Information identifying a set of source VMs, a set of destination VMs, and a set of target VMs also are received, wherein the target VMs are executing the application and are supported by (a) node(s) in a clustered virtualization environment. A set of inbound rules identifying (a) category(ies) of source VMs permitted to initiate connections with a subset of target VMs and a set of outbound rules identifying (a) category(ies) of destination VMs to which the subset of target VMs are permitted to initiate connections are received at the graphical user interface. Upon receiving a request to apply the security policies, the policies are configured based at least in part on the inbound and outbound rules and a visual representation of the security policies is presented in the graphical user interface.

Abstract: Disclosed herein are related to a method, a system, and a non-transitory computer readable medium for backing up a virtual machine. In one aspect, an instruction to store a requested state of a virtual machine is received. Configurations of the virtual machine in the requested state are determined, and a configuration data indicative of the determined configurations of the virtual machine is generated. The configuration data is stored. Another instruction to load the virtual machine in the requested state from a plurality of states of the virtual machine may be received. The stored configuration data indicative of the configurations of the virtual machine in the requested state may be retrieved. The configurations of the virtual machine in the requested state may be identified based on the retrieved configuration data, and the virtual machine in the requested state may be loaded according to the identified configurations.

Abstract: A system and method include receiving a request to call a requested application programming interface (API) at a target cloud computing platform. The system and method further include determining that the requested API is unsupported by the target cloud platform. The system and method further include generating a set of workflows associated with the requested API. The system and method also include generating a set of tasks common to the plurality of cloud platforms including the target cloud platform based on the set of workflows. The system and method also include generating a set of target APIs associated with the target cloud platform based on the set of tasks and an identity of the target cloud platform. The system and method also include communicating the set of target APIs t to the target cloud platform.

Abstract: A system and method includes an orchestration engine to determine workloads associated with operations on cloud platforms. The orchestration engine can receive a request to a universal application programming interface (API) or a request to an API associated with a cloud platform. The orchestration engine can determine the workload associated with the operation, and select one or more cloud platforms to distribute the workload based at least on a policy. The system and method also includes a lifecycle management engine that can receive requests to implement lifecycle rules on an object stored at a cloud platform. The lifecycle management engine can translate received requests to one API into requests to an API associated with a target cloud platform. The lifecycle management engine can receive status of the object from the target cloud platform, and provide the status to the requesting client.

Abstract: Disclosed herein are related to a method, a system, and a non-transitory computer readable medium for tracking data objects associated with a virtual machine. In one approach, an object container of the virtual machine is generated. The object container includes data objects associated with the virtual machine. For each of the data objects, a corresponding tag is generated. Each tag is indicative of a corresponding data object. Each tag includes a global identification of the corresponding data object. The global identification is unique across a distributed database. The tags are stored at the distributed database.

Abstract: Disclosed herein are related to a method, a system, and a non-transitory computer readable medium for providing a consolidated cloud system. The consolidated cloud system includes a plurality of cloud systems. A user request to access content in the consolidated cloud system is received, where the content is stored by a target cloud system of the plurality of cloud systems. The content from the target cloud system is retrieved, and the content is provided as part of the consolidated cloud system. The device generating the user request may be unaware of which cloud system stores the content.

Abstract: A method of eliminating false positives when testing a device management platform for controlling endpoints. The method includes sending, by a computing device, a request to a cloud controller to create an entity. The request causes operations including i) creating, by an endpoint responsive to a receipt of an interface request from the cloud controller, the entity, ii) receiving, by the cloud controller from the endpoint, response data generated by the entity, the response data indicating the creation of the entity, and iii) generating, by the cloud controller, control data based on the response data. The method includes receiving the control data from the cloud controller; sending, responsive to the receipt of the control data, a call to the endpoint for entity data; receiving entity data from the endpoint; comparing, the response data to the entity data to validate the response data; and generating a test status.

Abstract: A system for tracking resource usage in network. In a specific embodiment, the system includes a memory in a distributed-forwarding switch that maintains resource evolution information. A module, which communicates with the memory, is adapted to update the resource evolution information when the resource changes or information associated with the resource changes. In a more specific embodiment, the memory is implemented via plural resource-tracking hash tables associated with forwarding engines in switches connected to a network. Each hash table communicates with a corresponding resource-tracking module that monitors changes in accompanying forwarding engines and updates the hash tables in response thereto. The resource-tracking modules may also selectively provide information in the resource-tracking modules to a user interface in response to queries or control signals originating from a user interface.

Abstract: In one embodiment, entries in an address table of the forwarding engine in a linecard or supervisor are updated to a current time value when the MAC address in the entry matches the destination MAC address of a received packet to prevent flooding if the linecard or supervisor is a non-ingress linecard or supervisor of an asymmetrically routed flow.

Abstract: A method and apparatus for tracking at least one Layer-2 (L2) resource of a network device is provided. The method includes monitoring software operations on the L2 resource. The method further includes monitoring the state of the L2 resource in a corresponding hardware device (e.g., the network device). The L2 resource may be at least one of a Media Access Control (MAC) address, a Virtual Local Area Network (VLAN) identifier, a MAC address, and a VLAN identifier pair.

Abstract: In one embodiment, a load-balancing algorithm ensures that both network nodes at endpoints of a group of physical links aggregated into a logical channel are using the same load-sharing algorithm and also ensures that the load-sharing algorithm is normalized so that the same flow traverses the same physical link in both directions.

Abstract: In one embodiment, a load-balancing algorithm ensures that both network nodes at endpoints of a group of physical links aggregated into a logical channel are using the same load-sharing algorithm and also ensures that the load-sharing algorithm is normalized so that the same flow traverses the same physical link in both directions.

Abstract: In one embodiment, entries in an address table of the forwarding engine in a linecard or supervisor are updated to a current time value when the MAC address in the entry matches the destination MAC address of a received packet to prevent flooding if the linecard or supervisor is a non-ingress linecard or supervisor of an asymmetrically routed flow.

Abstract: Potential forwarding loops in bridged networks are detected. A switch monitors the movement of packet addresses in the forwarding table to different ports. If the port associated with an address changes multiple times in a short time duration, then a potential forwarding loop is detected, and a loop corrector is notified to take appropriate action to correct the problem, such as, but not limited to changing the state of a port to a blocking state or disabling an interface, etc. If there are multiple movements outside the short time duration, they may be correct movements adjusting to the actual reconfiguration of a network, and thus, these movements are effectively filtered out of consideration.

Abstract: Disclosed are, inter alia, methods, apparatus, data structures, computer-readable media, mechanisms, and means for detecting potential forwarding loops in bridged networks. A switch monitors the movement of packet addresses in the forwarding table to different ports. If the port associated with an address changes multiple times in a short time duration, then a potential forwarding loop is detected, and a loop corrector is notified to take appropriate action to correct the problem, such as, but not limited to changing the state of a port to a blocking state or disabling an interface, etc. If there are multiple movements outside the short time duration, they may be correct movements adjusting to the actual reconfiguration of a network, and thus, these movements are effectively filtered out of consideration.