4947results about How to "Load balancing" patented technology

An approach to abstracting the circuit switched nature of the public switched telephone network (PSTN) by using VoIP to provide voice actuated services is disclosed. By carrying a telephone call using VoIP technology for a short distance (frequently within a server room) significant benefits to call handling and capacity management can be obtained. Specifically, a PSTN-to-IP gateway is used to receive (and place) calls over the PSTN and route those calls internally to servers over an IP network in a packet switched format. A number of computer systems can receive and handle the calls in the IP format, including: translating the packets into an audio format suitable for speech recognition and creating suitable packets from computer sound files for transmission back over the PSTN.

A system and method for scheduling jobs in a multiprocessor machine is disclosed. The status of CPUs on node boards in the multiprocessor machine is periodically determined. The status can indicate the number of CPUs available, and the maximum radius of free CPUs available to execute jobs. Memory allocation is also monitored. This information is provided to a scheduler that compares the status of the resources available against the resource requirements of jobs. The node boards and CPUS, as well as other resources such as memory, are arranged in hosts. The scheduler then schedules jobs to hosts that indicate they have resources available to execute the jobs. If none of the hosts indicate they have resources available to execute the jobs, the scheduler will wait until the resources become available. A best fit of job to resources is attained by scheduling jobs to hosts that have the maximum number of free CPUs for a radius corresponding to the CPU radius requirement of a job. Once the job is scheduled to a host, it is dispatched to a host and resources required to execute the job are allocated to the job at the host.

The invention relates to a cloud computing resource scheduling method based on dynamic reconfiguration virtual resources. The method comprises the steps of: using cloud application load information collected by a cloud application monitor as a basis; making a dynamic decision based on the load capacity of the virtual resources for running cloud application and the current load of the cloud application; and dynamically reconfiguring virtual resources for cloud application based on the decision result. Dynamic adjustment of resources is realized by a method for reconfiguring virtual resources for cloud application, without needing dynamic redistribution of physical resources or stopping executing cloud application. The method can dynamically reconfigure the virtual resources according to the load variation of the cloud application, optimize allocation of the cloud computing resources, realize effective use of the cloud computing resources, and meet the requirements on dynamic scalability of cloud application. In addition, the method can avoid waste of the cloud computing resources, and save the cost for using resources for cloud application users.

The embodiment of the invention provides a method for implementing a virtual network and the virtual network. The method comprises the following g steps: step (1), obtaining a transfer type, a source and a destination of a data packet sent by a host computer by a controller of the virtual network; step (2), formulating a forwarding strategy according to the transfer type, a port identifier of the source and a target identifier of the destination; and step (3), sending the forwarding strategy to each virtual access exchanger and each middle exchanger through which the data packet is to be passed, and executing the forwarding strategy to the received data packet received by the virtual access exchanger and the middle exchanger. In the forwarding process of the data packet, the controller can obtain various attributes of the data packet as target identifiers, formulate the forwarding strategy according to the target identifiers and forward the data packet more intensively and flexibly; and multi-user and multi-path forwarding is supported, flow load is balanced, and network resources are utilized more efficiently.

A method and apparatus for performing polling is presented. Polling servers include the intelligence and as such may control polling clients. What is polled, when it is polled and how it is polled is controlled by the polling server and may be dynamically changed and any time. As such, the polling server may dynamically reconfigure the polling process for a polling client. In one embodiment, the polling server directs and controls the polling client(s) in a manner that effects load-balancing in a network.

The invention discloses a distributed file system metadata management method facing to high-performance calculation. The method comprises the following steps of: 1) establishing a catalogue metadata server cluster, a file metadata server cluster and an object storage server cluster; 2) establishing a global counting-based bloom filter in the catalogue metadata server cluster; 3) when the operation request of a client side arrives, skipping to execute step 4) or 5); 4) enabling the catalogue metadata server cluster to respond to the catalogue operation request of the client side to manage the catalogue metadata; and 5) enabling the file metadata server cluster to respond to the file operation request of the client side to manage the file metadata data. According to the distributed file system metadata management method disclosed by the invention, the metadata transferring problem brought by catalogue renaming can be effectively solved, and the distributed file system metadata management method has the advantages of high storage performance, small maintenance expenditure, high load, no bottleneck, good expansibility and balanced load.

A dynamic session redirector provides the ability to receive requests for network accessible storage resources and redirect them to the appropriate resources as needed while providing a single system image to the system requesting access to the resources. The redirector provides an ability to reconfigure the network storage resources in order to balance the load among the servers. By altering the association between individual storage resources and the servers responsible for them on the storage network, a more efficient and reliable system for access to the data may be provided.

The invention provides a dynamic flow dispatch method and system based on a software definition network. The method comprises a step of carrying out deep analysis of flow through flow information uploaded by a switch, and classifying flow types by a flow identification and grading module according to priorities, a step of calling a statistic management module to obtain information of switch ports, calculating joint links of the switch, and sending information of the joint links to a topology management module, a step of calling the topology management module to obtain a feasible path for flow direction and calculating real-time sate statistic information of the feasible path, and a step of analyzing features of the joint links to calculate weight of the joint links in the overall selection, working out the optimal dispatch strategy based on the weight and giving out the optimal dispatch strategy. The invention can achieve the optimal real-time dispatch of network flow, so that flow is enabled to reach a destination in a fastest manner, and purposes of improving the bandwidth utilization rate and optimizing network performances are finally realized.

The invention relates to a multitask concurrent executive system and a method for hybrid network service. The system comprises a client side, a load-balancing server, a signalling / media control server and an application server cluster. The running method thereof is as follows: after the client side receives a service request, the load-balancing server selects an optimum application server and transmits the service request to a service bus thereof. After an admission control module of the application server carries out admission control on the request, the service bus transmits the request to a dynamic flow execution engine, the dynamic flow execution engine starts a service flow corresponding to the request to analyze and process, and the service bus calls a corresponding service capability component in a low deferred event service engine and interacts with the signalling / media control server through the resource adapter so as to call the service resource of the latter; and a processing result is returned to the client side according to a reverse process. The multitask concurrent executive system supports the supply of hybrid network service over a telecommunication network and the internet and the multitask concurrent on a heterogeneous network and terminals.

The invention discloses a method for distributing metadata of a distributed type file system. The method uses a catalogue of the file system as a basic unit of Hash, adopts an extensible Hash method to distribute the metadata to each metadata server to ensure that clients can position the position of the metadata; the metadata has high accessing efficiency, even distribution and balanced load; meanwhile, the method maintains storage locality of the catalogue, realizes conveniently prefetching, and improves the accessing efficiency. A unique and unchanged identifier is given to the catalogue to avoid metadata migration caused by the path name of Hash and improve the system performance; participation or exit of the metadata server cannot cause redistribution of mass metadata so as to furthest reduce the metadata migration and ensure high expandability.

Methods, apparatus, and programming recorded in machine readable memory are provided for the index, search and retrieval of objects on a global network. This inventive system embeds a distributed index in a routing layer to enable fast search. The method provides dynamic insertion, lookup, retrieval, and deletion of participating nodes, objects and associated metadata in a completely decentralized fashion. Nodes can dynamically join and leave the network. This infrastructure can be applied to content networks for publishing, searching, downloading, and streaming.

The invention relates to a smart video monitoring system based on a cloud platform, and belongs to the technical field of remote smart monitoring. The system is characterized in that a video monitoring cloud platform is developed on a server cluster established on the basis of a Hadoop cloud computing platform, and video data acquired by front-end equipment are transmitted to a server in a stream way; through adoption of a massive distributed cloud computing technology, real-time tracking, processing, analysis and outputting of video streams are realized; and a service is provided externally, so that a terminal user can perform checking and calling conveniently. In the system, the characteristics of distributed high-concurrence access storage, high availability, rapid response, easiness in management and maintenance and the like are integrated. Video acquisition, video encoding and transcoding, video data storage, a streaming media distributed server and a client are all distributed on a physical host node in an access network. Compared with a conventional video monitoring system, the smart video monitoring system has the advantages that video real-time processing and transcoding efficiency and video data storage capacity are increased greatly; the requirement of people on the video monitoring system in a big data era is met; and a relatively large commercial value is achieved.

The invention provides a task scheduling device in a multimode distributive system. The task scheduling device comprises a new task pool, a task filter, a task filtering pool and a task dispatcher, wherein the new task pool is used for storing new tasks, and the new tasks at least comprise new tasks in which task marks are added; the task filter is used for filtering out tasks with empty marks from the new tasks stored in the new task pool; the task filtering pool is used for storing the tasks with empty marks, filtered by the task filter; and the task dispatcher is used for predicting a model according to task completion time by aiming at each task stored in the task filtering pool so as to predict the completion time for executing the task on each processing node in the node processing set, and allocating the task to the processing node with shortest predicted completion time.

An improved Internet delivery system for digitized photographs is described. The system embodies methodologies that speed the delivery of digitized images from photographic film scan centers to consumers across the Internet with high reliability. By delivering all the digital images to a more powerful centralized server infrastructure prior to the consumer's earliest opportunity to view them on-line, these images or pictures can be downloaded to a viewer of interest (e.g., user or family member) quickly enough to provide an appropriate Web response. Optimizations are also described that streamline the throughput with load balancing and provide virtually guaranteed uniform service with fail-over. In this manner, the system provides improved access, storage, and on-line availability of digital images.

One embodiment disclosed relates to a method of load balancing traffic in a switching mesh. A packet is received at a switch of the mesh. A MAC destination address is gathered from the packet. An entry in a MAC address table is found with the MAC destination address, and a switch identifier associated with the MAC destination address is obtained. An entry in a switch table is found with the switch identifier, and data associated with the switch identifier is obtained. A hash function is used on field data from the packet to generate a hash value, and the hash value and the associated data are used to generate an index value. An entry in the tag table is found with the index value, and a path tag associated with the index value is obtained. The path tag is inserted into the packet. The path tags may be weighted.

Various exemplary embodiments are a system and related method for application sensitive forwarding of a request for a web service including a broker that manages the forwarding of the request, a metric table that stores statistics for the web service, and a database storing the metric table and least one load balancing policy. Various exemplary embodiments include an optimization policy manager that manages forwarding of the request, a class of services table that stores information, and a database storing the class of services table and at least one optimization policy. Various exemplary embodiments include a database storing a security-based policy and a location-based policy defining trust environments, a request processor that receives the request from the client and determines a current environment of the client based on the at least one location-based policy, and a quarantining subsystem that drops the request when the request violates the security-based policy.

The invention discloses a method for realizing flow distribution based on front-end scheduling, a device thereof, an accessing method based on front-end scheduling, and a system for realizing flow distribution based on front-end scheduling. The method for realizing flow distribution based on front-end scheduling comprises the steps of receiving a request which comes from a client for acquiring target server connecting information; selecting a target server for the client according to a customized strategy from a server to be selected; and returning the target server connecting information to the client. The method of the invention can realize flexible configuration for a client accessing flow, load equalization of a server cluster, fault tolerance and disaster recovery.

In order to achieve the load balance of network access traffic, the invention discloses a method for providing domain name resolution services, belonging to the field of communication. The method comprises the following steps: receiving a DNS (domain name system) request message with a specified domain name, which is sent by a client, and forwarding the DNS request message to at least two DNS servers for domain name resolution, wherein the DNS servers belong to different ISPs (Internet service provider); receiving the response message of the IP (Internet protocol) addresses of each web server, which is acquired after the operation of the domain name resolution is conducted and returned by each DNS server; selecting on the principle that the load of each outlet is balanced from the acquired IP addresses of each web server according to the current free bandwidth of each outlet corresponding to each web server; and sending the selection results to the client. Accordingly, the invention can prevent the traffic from leaned to one side and achieve the load balance in the network system. The invention discloses an outlet gateway device, a DNS relay device and a communication system.

The invention discloses a method for scheduling virtual machines, which belongs to the field of computer networks. The method comprises the following steps of: 1) running a physical machine monitor on each physical server for regularly collecting loads of all virtual machines and sending the loads to a virtual machine scheduler, and receiving and executing instructions sent by the virtual machine scheduler; 2) regularly judging the virtual machines with load data changes and the physical server where the virtual machines are positioned by the virtual machine scheduler; 3) adjusting the virtual machines with load data changes by the virtual machine scheduler by using bin packing algorithms to obtain the target corresponding relationship of the virtual machines and the physical servers; 4) comparing the current corresponding relationship with the target corresponding relationship of the virtual machines and the physical servers by the virtual machine scheduler to generate a virtual machine scheduling plan; and 5) scheduling the virtual machines by the physical machine monitor according to the virtual machine scheduling plan. Compared with the prior art, the invention has the effect of load balance and can also make the physical servers in an idle state dormant and further reduce the energy consumption.

A data center management unit (100) for managing and controlling power distribution to computers in a data center, comprising:a first power inlet (101) for connectivity to a first power feed;a second power inlet (102) for connectivity to a second power feed;a plurality of power outlets (111, 112, 113, 114, 115, 116, 117, 118) for providing power to the computers;a plurality of power switches (131, 132, 133, 134, 135, 136, 137, 138) each having a first input coupled to the first power inlet, a second input coupled to the second power inlet, and an output coupled to a respective power outlet; anda processor (161) adapted to control the power switches for dynamically switching individual power outlets between the first power inlet and the second power inlet or vice versa, and for dynamically switching off individual power outlets.

The invention discloses a method of compressing a neural network. The connection relation of the neural elements of the neural network is expressed by a plurality of matrixes. The compression method comprises steps that a dividing step: a plurality of matrixes are divided into a plurality of matrixes, and every matrix is divided into a plurality of sub-matrixes; a compressing step: every sub-matrix is compressed, and every sub-matrix is compressed into a sparse matrix; and an encoding step: every compressed sparse sub-matrix is encoded. The invention provides a device used for compressing the neural network.

The invention relates to the technical field of cloud computing, in particular to a dynamic virtual resource management strategy method for performance weighting under a cloud platform. According to the method, at first, contrast matrixes of various request resources and performance weighting values are configured, according to the needs of users and real-time monitoring data of the cloud platform, a load performance metric model based on load performance weighting is provided, and then the dynamic virtual resource management strategy method under the cloud platform is provided according to the model. The method comprises a virtual machine deployment strategy and a virtual machine transfer strategy, on the condition of meeting the existing load request, the cloud platform ensures that the load degree of a physical server in each scheduling area to be maximized, and on the basis, the number of the physical servers, needing to be started, of the cloud platform can be minimized. According to the method, the needs of the users are taken into consideration, data of individuation weight setting and cloud platform system real-time performance monitoring are given, the needs of QoS can be met, resources of the cloud platform system are utilized reasonably, and the load performance of the cloud platform is ensured.

The invention relates to the field of mobile communication, in particular to a distributed type base station technology for realizing load balancing. The invention provides load balancing equipment, a distributed type base station system and a communication method thereof. The method comprises the following steps: when the load balancing equipment receives data sent by a RRU, determining whether a BBU corresponding to the RRU exists according to the relation between the RRU and the BBU; if yes, forwarding the data to the BBU corresponding to the RRU; and if not, according to the currently available resources of all BBUs, choosing one of BBUs to which data is forwarded to. Because of the load balancing equipment arranged between the BBUs and the RRU, the load distribution can be determined according to the currently available resources of the BBUs, namely one of BBUs is chosen to establish a data link with the newly accessed RRU according to the currently available resources of the BBUs, so that the aims of balancing the load of the BBUs and fully utilizing the BBUs to process resources are realized.

The system comprises a delivery server and at least two robot servers. Wherein, the delivery server is used to receive the user's message and to deliver the user's message to the relevant robot server; the robot server is used to respond to the user's message and to automatically give a feedback reply. The invention also reveals a method thereof.

Load balancing of activities on physical disk storage devices is accomplished by monitoring reading and writing operations to blocks of contiguous storage locations on the physical disk storage devices. A list of exchangeable pairs of blocks is developed based on size and function. Statistics accumulated over an interval are then used to obtain access activity values for each block and each physical disk drive. A statistical analysis leads to a selection of one block pair. After testing to determine any adverse effect of making that change, the exchange is made to more evenly distribute the loading on individual physical disk storage devices.

Methods or systems for management and / or optimization of at least a portion of a data network by generating a set of paths between each origin and destination pair, pruning the set of paths to generate a pruned set of paths; and computing an optimum path between each origin and destination pair. Methods and systems for generating a diverse set of path options for the routing of traffic within at least a portion of a network comprising: generating a set of paths between each origin and destination pair; and pruning the set of paths to generate a pruned set of diverse path options within at least a portion of a network.

A peer-to-peer storage system includes a storage coordinator that centrally manages distributed storage resources in accordance with system policies administered through a central administrative console and a lexical recovery sub-system that automatically creates versions of files that are thereafter maintained by the system. The storage resources, or “nodes,” are otherwise unused portions of storage media, e.g., hard disks, that are included in the devices such as personal computers, workstations, laptops, file servers, and so forth, that are connected to a corporate computer network, and are thus otherwise available only individually to the respective devices. The storage coordinator assigns the nodes to various “replication groups” and allocates the storage resources on each of the nodes in a given group to maintaining dynamically replicated current and previous versions of the group files. The storage nodes in a given group perform dynamic file replication and synchronization operations by communicating directly, that is, peer-to-peer, using a message-based protocol. The storage coordinator also manages distributed searches of file content on the network by selecting one node from each group to search through the associated group files. The selected nodes report the search results back to the storage coordinator, which organizes the results and provides them to the user. The user may then restore or recover a previous version of a file or review a current version of the file by selecting the desired file version from the search results.

Techniques are disclosed for communicating to remote devices information about change events related to changes in access to an enterprise system. A device access management system may facilitate communication about a change event to the remote devices. Information about a change event may be stored in a change event object based on the type of change event (e.g., a policy change, an application change, and a settings change). A change event queue may persistently store information corresponding to change events. One or more computing nodes may be scheduled to execute an action process for each change event based on the type of the change event. A computing node may communicate information (e.g., an instruction to implement adjust access) about a change event to remote devices. A change event may persist on the queue until all remote devices are notified about the change event.

The invention relates to a network device, and a report transfer method based on multi-core processor. Wherein, said network device comprises: classifying unit and transfer processor with multi-core unit; the classifying unit receives report at any port of network device, selects core unit from the transfer processor, and sends received report to selected core unit; one core unit of transfer processor transfers report when receives report. The invention can use classifying unit to select core unit to balance the load between core units, to make them parallel transfer reports, to improve the report transfer ability of network device.