193 results about "Network partition" patented technology

Directing a video player to a video source is disclosed. An indication that a client requires video streaming instructions is received. A past download performance for a plurality of clients that or correlated to the client requiring video streaming instructions is determined. At least one of: a content source, and a bitrate, is selected, based at least in part on the determined past download performance of the plurality of correlated clients. The client is provided with instructions based at least in part on the selection.

Methods and apparatus provide formation and management of intelligent application clusters in a storage area network. Disk arbitration mechanisms ensure that a cluster is owned by a single member. In the event of a network partition, each cluster member involved arbitrates to gain ownership of a cluster. High availability mechanisms allow monitoring of system resources and effective failover capabilities.

A system and method for versioning of atomic units of data (such as XML fragments) in a distributed data store are disclosed. The distributed data store may maintain a linear version history for each fragment, such that each has no more than one successor and one predecessor, with no implicit branching. An application programming interface may provide an explicit branching request (indicating an intent to merge modified data in a child branch into its parent branch), along with read, create, create-merge, and/or integrate requests. A reconciliation policy may be specified on a per-application basis, and may include automatic reconciliation using a callback function or interactive reconciliation, with inputs returned to the application. Writes to a minority side of a network partition may be refused unless an explicit branch is requested. Transactions may not be supported in client applications or may be supported through explicit branching. Transactions may be used internally.

A means for guaranteeing the proper behavior as specified by the JMS semantics of clustered message server when the individual computer that comprise the cluster are separated by a network partition. A clustered message server is responsible for the reliable transportation of messages between different distributed computer applications. It employs multiple computers to perform a function that otherwise appears to be performed by a monolithic server running on one computer, but with more capacity and reliability than can be provided by one computer. If a computer in the cluster fails, another computer should automatically assume the role of the failed computer. However, it is not possible for the other machines in the cluster to detect the difference between the failure of one or more computers in the cluster, and the failure of data network connecting those computers. In ordinary clusters, different actions would be required in these two cases, but since they are impossible to distinguish, computer failure is always assumed and network failure is ignored and the consequence non-deterministic. The invention described here provides a means of responding to failures that yields correct behavior as specified by the JMS semantics whether the failure is due to computer failure or network failure.

A cluster system is treated as a set of resource groups, each resource group including an highly available application and the resources upon which it depends. A resource group may have between 2 and M data processing systems, where M is small relative to the cluster size N of the total cluster. Configuration and status information for the resource group is fully replicated only on those data processing systems which are members of the resource group. A configuration object/database record for the resource group has an associated owner list identifying the data processing systems which are members of the resource group and which may therefore manage the application. A data processing system may belong to more than one resource group, however, and configuration and status information for the data processing system is replicated to each data processing system which could be affected by failure of the subject data processing system-that is, any data processing system which belongs to at least one resource group also containing the subject data processing system. The partial replication scheme of the present invention allows resource groups to run in parallel, reduces the cost of data replication and access, is highly scalable and applicable to very large clusters, and provides better performance after a catastrophe such as a network partition.

A hybrid quorum/consensus and primary-backup fault-tolerance model in an object-based distributed data storage system. When a primary manager fails, a hierarchy of network entities is established in which a group of realm managers first authorizes a failure-handling event through quorum/consensus and a backup manager (for the failing primary manager) then executes the decision of the quorum of realm managers. The realm managers, operating by consensus, determine whether (a) the primary manager can indeed be asserted to be down, and (b) whether there is a quorum of realm managers in agreement on this decision. If both are true, a master realm manager instructs the backup manager to proceed to execute the necessary steps to become the primary manager and function as the primary until the original primary manager is brought back into service. The hybrid fault-tolerance approach handles both single unit failures and network partitions in a unified way, without creating a single cluster out of the fault domain.

The invention discloses a method for non-interrupted operation of a high availability system through a technology of interacting information by a magnetic disk. For a multi-node high availability cluster system with shared storage equipment, a small space of a shared storage is used as a quorum disk. Various high availability nodes can simultaneously read and write the quorum disk; a quorum disk daemon process of each node writes state information of the node into the quorum disk, and detects stored information of all nodes in the quorum disk, and when nodes fail and network partition occurs, a main control node of the whole cluster is intelligently selected from normally working nodes, so that the cluster system restores a normal working state; therefore, split-brain of the high availability system is avoided and the non-interrupted normal operation of the high availability system is ensured.

A mediator for efficiently exchanging management data in a hierarchical management network for managing network elements of a data transport network partitions resources of the network elements among a plurality of management representations of the network elements and selectively forwards management data associated with respective management representations to a higher level of the hierarchy. The mediator modifies the management data so that the higher level (i.e. a northbound element management system or a network management system) receives the management data in a view consistent with the management representation of the network element. Client-based management representations may be created to permit clients to manage selected resources of a network element, and testing-based management representations may be created to facilitate concurrent testing of a network element.

The invention discloses a parallel computation method for Newton power flow of a large-scale electric power system, which belongs to the technical field of safe analysis and computation of electric power systems. A modified equation set of the Newton power flow is resolved by using a computer according to a parallel cluster through a program in combination with parallelization technology for updating a Jacobian matrix by a generalized minimal residual method combined with a quasi-diagonal parallel preconditioner so as to realize parallelization of a power flow resolving process of the large-scale electric power system. The method does not need network partitioning in any form and has simple computation and high generality; a forward/backward sweeping dependency relationship does not exist inside an algorithm, and parallel efficiency and computation speed are high; and computation can be performed by the common parallel computer and the parallel cluster and the method is convenient to popularize and apply, and the like. The method can be applied to the parallel power flow computation of the electric power system and is particularly suitable for the parallel computation of the Newton power flow of a large-scale complex electric power system.

This invention discloses an automatic dividing method for networks to a multi-antenna-scattered structure with a moment method, which carries out three times of network partition to the dispersion surface on the basis of setting up geometrical models, firstly, it neglects antennas connected by the scatterer surfaces to apply a recursive dichotomy to divide the surface of the scatterer into triangles and quadrangles surface elements, secondly, it considers the antenna of the scatterer surface and divides the divided triangle and the quadrangle surface elements again into a square element with the line-surface link points at the center of the quadrangle, several triangles without line-surface link points and quadrangles in terms of different positions of the line-surface link points, thirdly, carries out network dividing to the narrow and long elements to get triangles and quadrangles even smaller than those of last two times.

Techniques are disclosed for maintaining high availability (HA) for virtual machines (VMs) running on host systems of a host cluster, where each host system executes a HA module in a plurality of HA modules and a storage module in a plurality of storage modules, where the host cluster aggregates, via the plurality of storage modules, locally-attached storage resources of the host systems to provide an object store, where persistent data for the VMs is stored as per-VM storage objects across the locally-attached storage resources comprising the object store, and where a failure causes the plurality of storage modules to observe a network partition in the host cluster that the plurality of HA modules do not. In one embodiment, a host system in the host cluster executing a first HA module invokes an API exposed by the plurality of storage modules for persisting metadata for a VM to the object store. If the API is not processed successfully, the host system: (1) identifies a subset of second HA modules in the plurality of HA modules; (2) issues an accessibility query for the VM to the subset of second HA modules in parallel, the accessibility query being configured to determine whether the VM is accessible to the respective host systems of the subset of second HA modules; and (3) if at least one second HA module in the subset indicates that the VM is accessible to its respective host system, transmits a command to the at least one second HA module to invoke the API on its respective host system.

The invention discloses a block chain fragmentation method and device, and the method comprises the steps: determining a target network partition to which a new node belongs through employing a K neighbor classification algorithm according to K partition notification responses received by the new node, so that the new node can be added into the target network partition. The K neighbor classification algorithm can screen out the network partition including the most neighbor nodes of the new node according to the partition notification response; therefore, the new nodes can be divided into the network partitions including the most neighbor nodes by using the K neighbor classification algorithm, so that the nodes with relatively short distances can be divided into the same network partition,the communication efficiency between different nodes in the network partition is improved, and the transaction processing efficiency in the network partition is further improved. Besides, each networkpartition can independently process the transaction, so that each fragment can process different transactions in parallel, the transaction processing and confirmation concurrency is improved, and thethroughput of the whole block chain network is improved.

The invention relates to a multi-sink deployment and fault tolerance method for a wireless sensor network. A multi-sink optimal deployment method based on improved particle swarm optimization clustering is firstly brought forward, the improved particle swarm optimization is combined with K-means clustering, clustering and partitioning are carried out on all sensor nodes in a network area, the multiple sink nodes are deployed in the center of each cluster respectively, and thus, effects of reasonable partitioning and optimal multi-sink node deployment of the network are achieved. In addition, in view of a condition in which a certain sink node fails in the multi-sink network, a routing algorithm with a fault tolerance mechanism is brought forward, a fault tolerance mechanism with multi-path routing is adopted, two paths from a source node to different sink nodes are built, and when a certain sink node fails, alternate routing is activated. Through deploying multiple sink nodes in the wireless sensor network, the network reliability can be improved, when one sink node fails, other several sink nodes can replace the node for work, reliable operation of the network is ensured, and the service life of the network is prolonged.

The invention is a large-scale complex network dividing method, including three parts of network preprocessing method, sub-network judging method and network dividing method, firstly adopting the network preprocessing method to improve large-scale network topology into a diagram with side conductivity not less than 2, then further combining the sub-network judging method and the network dividing method to divide the large-scale network into several ring sub-networks and fan-shaped sub-network, where it uses the network dividing method to divide the whole network into sub- networks, and divides the sub-networks into a ring sub-network with closed outer ring link and a fan-shaped sub-network with unclosed outer ring link. The large-scale network uses each sub-network as unit and adopts a distributed intelligent recovery policy, able to flexibly and conveniently solve the existence problem that the large- scale complex network confronts and each sub-network can simultaneously and parallel process many failures, thus wining time for failure recovery.

The invention relates to an active power distribution network power supply area division method and an apparatus thereof. The method comprises the following steps of using a fuzzy clustering method to acquire an initial power supply subarea of a power distribution network; establishing a power supply area target function and a constraint condition; merging the initial power supply subarea and an initial power supply subarea which has a switch connection relationship with the initial power supply subarea into a combination subarea, and using the power supply area target function and the constraint condition to determine an area optimization scheme of the combination subarea; and according to the area optimization scheme of the combination subarea, determining an area optimization scheme of a power supply area. By using the technical scheme of the invention, a basic mode of subarea optimization and total assessment is adopted so that a reasonable network subarea structure is acquired; and switch operation of a network is simplified and operation efficiency of the power distribution network is increased.

The invention provides a neural network wind power short-term forecasting method based on a fuzzy partition theory. The neural network wind power short-term forecasting method based on the fuzzy partition theory adopts the mode of combining a fuzzy theory, artificial intelligence and a statistical theory through analyzing the important features of wind velocity variation and the relationship between wind velocity and power. When wind power forecasting is conducted, wind scale fuzzy partition processing is conducted on wind velocity data obtained from weather forecasting according to periods of time, BP neural network partition forecasting is conducted, a forecast power value is obtained through multiplying a partition forecast value by a membership degree value of the partition forecast value and adding all partition values, a probability statistics modified algorithm is conducted, and the forecast power is obtained. The neural network wind power short-term forecasting method based on the fuzzy partition theory improves the accuracy of a forecasting model effectively.

The invention belongs to the technical field of block chains, and provides a Byzantine fault-tolerant consensus method and system. The Byzantine fault-tolerant consensus method comprises the steps of:generating a block through a packaging transaction of a main node, and broadcasting the block to a consensus node in a blockchain; verifying the received newly generated block through using the consensus node, voting a current proposal according to a verification result, and broadcasting the voted current proposal to other consensus nodes; counting votes of the newly generated block through usingconsensus node, and judging whether the proportion of votes voting through in total votes exceeds a preset threshold value or not; if so, submitting the newly generated block, generating submission information and broadcasting the submission information to other consensus nodes; if not, generating non-assigned submission information and broadcasting the non-assigned submission information to other consensus nodes; and counting the submission information of the newly generated block through using the consensus node. According to the Byzantine fault-tolerant consensus method and the system of the invention, the fault-tolerant rate of the Byzantine fault-tolerant consensus method is improved, the tolerableness of network partition is realized, and the problem that states of the consensus nodes in the blockchain cannot be consistent due to the network partition is solved.

The application provides a method and device for confirming a network topology and a storage medium. The method for confirming the network topology comprises the following steps: confirming at least one network partition in a network area; obtaining the device information of each device in the network partition, wherein the device information comprises device identifiers, device roles and link information, and the link information is used to represent the connection among devices; and confirming the network topology of each network partition according to the device information of each device.The application can recognize interconnection devices among network partitions.