摘要:
Goal-based availability and change management are handled over groups of heterogeneous storage controllers. Probabilistic and deterministic methods are employed to determine the allocation and placement of storage volumes to storage controllers, as well as the degree of data redundancy necessary to achieve data availability goals The probabilistic methods can take into account past observations of controller availability, and operator beliefs, as well as the state of storage controller configuration, in coming up with a probabilistic estimate of future availability.
摘要:
Goal-based availability and change management are handled over groups of heterogeneous storage controllers. Probabilistic and deterministic methods are employed to determine the allocation and placement of storage volumes to storage controllers, as well as the degree of data redundancy necessary to achieve data availability goals. The probabilistic methods can take into account past observations of controller availability, and operator beliefs, as well as the state of storage controller configuration, in coming up with a probabilistic estimate of future availability.
摘要:
A multi-plan generation system generates a multi-plan for a multi-layer storage area network. The system generates a predetermined number of plans while generating and retaining globally optimum plans. The system dynamically calculates and orders these plans according to optimization criteria and user requirements. The system comprises a planner for each of the layers in a storage area network such as a host planner, a storage planner, a network planner, a zone planner, etc. Each of the planners applies a variance optimization, a preview optimization, and a detail optimization and interacts with a policy repository to obtain policy guidance, enabling the system to prune the set of plans. A control module invokes the planners in a predetermined order.
摘要:
Techniques for automatically designing an application-requirement based configuration for a distributed computing system are provided. One or more application-level templates are obtained, wherein the one or more templates are representative of one or more requirements associated with one or more applications. Using the one or more application-level templates, one or more logical flows are created. The one or more logical flows are then used to generate one or more system-level configuration plans of the distributed computing system.
摘要:
Techniques for automatically designing an application-requirement based configuration for a distributed computing system are provided. One or more application-level templates are obtained, wherein the one or more templates are representative of one or more requirements associated with one or more applications. Using the one or more application-level templates, one or more logical flows are created. The one or more logical flows are then used to generate one or more system-level configuration plans of the distributed computing system.
摘要:
Goal-based availability and change management are handled over groups of heterogeneous storage controllers. Probabilistic and deterministic methods are employed to determine the allocation and placement of storage volumes to storage controllers, as well as the degree of data redundancy necessary to achieve data availability goals The probabilistic methods can take into account past observations of controller availability, and operator beliefs, as well as the state of storage controller configuration, in coming up with a probabilistic estimate of future availability.
摘要:
A method for lifecycle management for file systems has several steps. At least one storage container adapted to support a file system is allocated. The at least one storage container and the file system are associated with a service class. The file system is created in the storage container. The service class is used during lifecycle actions performed on the file system. Lifecycle actions may comprise, e.g., creating, extending, replicating, moving, or migrating a file system. The service class for the file system may also be maintained for a lifecycle of the file system.
摘要:
In an exemplary aspect of the invention, a method is disclosed for lifecycle management for file systems. The method comprises the following steps. At least one storage container adapted to support a file system is allocated. The at least one storage container and the file system are associated with a service class. The file system is created in the storage container. The service class is used during lifecycle actions performed on the file system. Lifecycle actions may comprise, e.g., creating, extending, replicating, moving, or migrating a file system. The service class for the file system may also be maintained for a lifecycle of the file system.
摘要:
The present invention is a method of doing business over a network that: receives a request for transmitting digital information after a start time and before an end time, determines the time required to transmit the digital information based on the number of packets in the information and the network speed, schedules a transmit time for the digital information, and accepts the digital information for transmission only if the time required to transmit is less than or equal to the difference between the transmit time and the end time. Pricing of the transmission can be determined by the priority of transmission, whether the information is transmitted the first time or rescheduled, and whether the user receives an acknowledgment.
摘要:
A filter adaptor connects a computer to a network through a network connection. The network has a large plurality of broadcast type messages having a given protocol. Each of the broadcast messages has a destination field and a source field. The adaptor has a preference list of one or more fixed channels that the computer is capable of receiving. Each fixed channel has a channel identifier that is also listed in the preference list. Further, the filter adapter has a filtering process that monitors the source field and discards any message that does not have a source identifier in the source field that matches one of the channel identifiers on the preference list.