82results about How to "No single point of failure" patented technology

The invention discloses a key-value data distributed caching system and a method thereof. The system comprises a key-value access proxy cluster and a key-value data storage cluster, wherein the key-value access proxy cluster is composed of a plurality of data access proxy nodes, and the key-value data storage cluster is composed of a plurality of data fragmentation storage replication clusters; the data access proxy nodes are used for sending a node business storage request, routing a data read and write request to a target data storage node, and storing or reading business data; and in the data fragmentation storage replication clusters, a master-slave data replication mechanism is adopted to determine that a single point of failure does not exist in the data, read data are read from a plurality of slave nodes, and load balance of the read data is determined. By adopting the technical scheme provided by the invention, key-value data automatic fragmentation storage is realized through a consistency hash algorithm, and horizontal expansion of data storage capacity of the system is guaranteed.

The invention discloses a cloud storage system for providing cloud storage service for users. The cloud storage system is characterized by comprising a cloud storage service provider (103), a primary agent node (102), a secondary agent node and a storage cloud (105), wherein a user sends a service request, the primary agent node selects the service scheme provided by the configuration of the optimal cloud storage service provider according to the request, and the secondary agent node selects the storage center in the data storage cloud to realize the cloud storage service. The invention also provides a data deployment method of the cloud storage system, which is used for deleting, modifying or adding user data. The invention can support services provided by multiple cloud storage service providers to accelerate the response speed of service used by the users, and can dynamically configure a bottom storage architecture according to the characteristics of the users, thereby simultaneously having the advantages of a master-slave mode architecture and a peer to peer storage architecture.

The invention provides a mirror image storage and distribution system for virtual machines. The mirror image storage and distribution system comprises cluster management servers, local object storage servers, DHT-MAP mapping servers and QEMU block drivers, wherein the local object storage servers, the DHT-MAP mapping servers and the QEMU block drivers run on test computers of a distributed cloud computing platform. By the adoption of the mirror image storage and distribution system, batch and fast deployment of distributed cross-region virtual machines can be achieved, and the defects that the period is long, dynamic extension can not be achieved and usability is low when a traditional system and method are used for deploying the virtual machines can be overcome.

The invention relates to a fireman individual-soldier system on the basis of a wireless Mesh network framework. The system consists of a wireless monitoring mainframe, a plurality of wireless individual-soldier sub-machines and a plurality of wireless repeaters, wherein the wireless monitoring mainframe consists of a monitoring module and a wireless center node module; and each wireless individual-soldier sub-machine consists of a central processing unit module, a positioning module, a vital sign detection module, a distress call module, an environment condition detection module, a carried equipment detection module and a wireless terminal node module. When the fireman individual-soldier system is applied particularly, firemen wear the wireless individual-soldier sub-machines to enter a fire site or other accident sites and a chief officer monitors information such as postures, motion tracks, vital signs, environment conditions, carried equipment conditions and the like through the wireless monitoring mainframe, so that the intelligence support of a command decision is provided for the chief officer in time and a scientific basis can also be provided for daily tactical training. By the fireman individual-soldier system, the modernization, automation and intellectualization levels of a fire command are improved and the battle damage rate of the firemen can also be effectively reduced.

The invention discloses a credible node joining method in VANET based on a block chain, belongs to the field of VANET safety, and aims to improve traffic jam and assist safe driving. If the vehicle isa malicious node, hidden dangers are buried for traffic safety, for example, the malicious node scatters false or attack data information, so that the reliability of related data information receivedby the vehicle is difficult to guarantee. And if the malicious node is rejected when the malicious node is added into the VANET, the security can be improved to a great extent. The invention providesa credible node joining method in VANET based on a block chain, and a roadside unit performs distributed verification on an applied vehicle. the method comprises: firstly, verifying whether the vehicle has a legal identity or not; then, verifying the integrity of the vehicle-mounted platform by utilizing a trusted computing technology; and finally, evaluating the reputation value of the vehicle in the last certificate validity period, namely evaluating whether the vehicle is a malicious node or not through historical performance; and allowing the verified vehicle to join, otherwise refusing.

The invention belongs to the technical field of tax system credit investigation management, and discloses a tax payment credit investigation system based on a block chain and a processing method, andthe method comprises the steps: accumulating data of each mechanism to a unified block chain platform, and integrating all information; using a specific consensus algorithm to input credit data into the block chain, and analyzing and calculating individual credit scores to form a huge information set; meanwhile, enabling the open port to continuously accept resources to form an information network, so that all participants are equivalent and mutually beneficial; constructing a tax platform introducing an interstellar file system using the interstellar file system for uploading duplicate text information, and directly storing a returned hash value and a small tax payment data file on a block chain which is used for a tax system and a credit alliance and caching the value and file in a standby traditional database at the same time. According to a data sharing method based on the block chain, member nodes in the regional autonomous alliance are mutually constrained and supervised while sharing data, and 'information island 'phenomenon is solved.

The invention discloses a distributed storage method. The method comprises the following steps of: maintaining and synchronizing a cluster node list of each node of a cluster, and maintaining cluster catalog data; allowing, by the node, a user to map a local catalog to a cluster catalog, storing a mapping relationship in a home node, and synchronizing the local catalog and the cluster catalog, between which the mapping relationship is established; maintaining a storage node list of each file by the cluster catalog, and receiving and transferring, by the node, a request for distributing the files to more nodes or deleting the files from the existing nodes, namely, a request for editing a file storage node list; and receiving a synchronous node list of a user configuration file by the cluster catalog, and executing data synchronization. The invention furthermore discloses a distributed storage system which comprises a node list unit, a cluster catalog unit, a catalog synchronization unit, a local synchronization unit, a content synchronization unit and a transfer unit. By applying embodiments of the invention, the distributed storage system which is user-friendly, strong in adaptability, transparent and reliable can be realized.

The embodiment of the invention discloses a construction method of a center node of a distributed system. The method comprises the following steps of: starting an election process by sending an election message, comparing an election round and a message request number of a node which receives the election message, and carrying out updating by adopting a large election round and a large message request number; forming a recommending message of the center node according to the updated election round and message request number, and sending the recommending message to the distributed system; and counting the recommending of the center node in the recommending message by the node, and setting the node having a recommending number reaching a preset threshold value as the center node. The embodiment of the invention further discloses a data request submitting method and a monitoring method based on the center node of the distributed system, and apparatuses of the data request submitting method and the monitoring method based on the center node of the distributed system. The center node constructed by the embodiment of the invention is stable.

The invention provides a multi-control multi-activity storage system. In the aspect of hardware, a multi-control multi-activity storage system hardware module is constructed. A module design is adopted in each main component. Controller data interconnection modules and at least two data switching exchangers are connected through redundant data links and the data switching exchangers are all in an active state. The controller data interconnection modules and each data switching exchanger all have independent data links. The communication method includes that the controller data interconnection modules enable data with to-be-sent specific forms to be distributed to the data switching exchangers which are connected and respectively send data which are provided with specific forms and are distributed to part of systems of the data switching exchangers to each data switching exchanger. An active-active mutual preparation state is between the controller data interconnection modules, an active-active mutual preparation state is also between the data switching exchangers, and the redundant design among multiple interface modules is adopted to ensure that single-point faults do not exist.

The invention discloses an optimization method of the Cassandra system in virtual environment. The optimization system of the Cassandra system in virtual environment comprises the following steps: using Point-to-Point Protocol to obtain a corresponding relation between a virtual machine and a physical machine in order to ensure the startup physical machines on each rack, and the startup virtual machines on each physical machine; organizing and classifying the virtual machines according to the startup physical machines on each rack, and the startup virtual machines on each physical machine; using dichotomy to choose the corresponding virtual machine as the present virtual machine and a main backup node and an auxiliary backup node as replica placement according to the result of organizing and classifying the virtual machines and key value requests from users; meanwhile ensuring the present virtual machine, the main backup node and the virtual machine which is closest to the virtual machine of receiving the key value requests of the users in the auxiliary backup node to communicate. The invention further discloses an optimization system of the Cassandra system in virtual environment. The optimization method and system of the Cassandra system in virtual environment is capable of resolving problems of replica placement failure and low communication efficiency in the prior art.

The invention discloses a Storm based stream computing frame text index method and system. The Storm based stream computing frame text index method includes: implementing topology of a storm, designing a Storm real-time data processing frame, and completing a webpage automatic extraction program of a web spider; automatically extracting key words; and classifying texts: classifying texts into one or more classifications according to content or attributes of the texts. The method and the system can allow backup data back in the conditions of data corruption or data loss, and recovers data; the function of centralized operation and maintenance of the system can be provided; an interface is beautiful and practical, and a convenient and visual graphical user management interface is achieved; function expansion can meet the demands of a user for later system expansion and use range expansion; and fault tolerance allows the system to have a certain fault tolerance mechanism when illegal data generates due to user input or wrong operations.

The invention discloses a cloud computing resource pool system and an implement method of the cloud computing resource pool system. The method includes the steps that multiple network notes are deployed in a cloud computing resource pool, wherein each network note is used for achieving associative network functions of one or more tenants in a cloud computing resource pool; meanwhile, corresponding network service components need to be deployed for computing nodes to serve as agents of network control functions, and the computing nodes and the network nodes achieve associative network management functions together. The network node functions in the traditional cloud computing resource pool are distributed into the multiple network nodes, and therefore if a certain network node breaks down, only the tenant using the network node is affected while other tenants in the resource pool are not affected. In this way, the problem that because only one network outlet is provided for flows of all virtual machines in a traditional cloud computing network environment, network paralysis of the whole cloud computing resource pool is caused if the network nodes break down is solved, and accordingly the robustness of the cloud computing system is improved.

The invention discloses a satellite-borne damping variable shock absorber which comprises a piston assembly, a piston cylinder, an elastic support device, a damping generation device and a sensing device; and the sensing device comprises a fiber bragg grating sensor (2), a photonic crystal fiber (3) and a fiber bragg grating demodulator (4). According to the invention, a fiber bragg grating is modulated to serve as a vibration sensor. Compared with the prior art, the shock absorber provided by the invention has the advantages that the weight is light, the structure is simple, and the flexibility is high, and the real-time feedback of vibration frequency can be realized; and simultaneously, the photonic crystal fiber is adopted as a transmission medium for sensing signals, the bending lossgenerated by optical fiber wiring can be reduced, especially the stretching loss under a vibration motion environment is reduced, and the control accuracy is improved.

The invention relates to a method for node placement of the Internet of things. The method comprises the steps of establishing a square basic monitoring area model, calculating the area of the basic monitoring area model and comparing the area of the basic monitoring area model with the area which actually needs to be monitored, establishing a novel square basic monitoring area model and calculating the area of the novel square basic monitoring area model, determining the multiple relation of the area of the novel square basic monitoring area and the area which actually needs to be monitored, determining the relations of the number of nodes of a total sensor, the distance of two adjacent sensor nodes on diagonal lines, a topological hierachy number, the number of cluster head nodes and the area which actually needs to be monitored, actually choosing and arranging a monitored area, choosing a topological structure which can represent the whole terrain configuration, configuring the dynamic-static combination routing mechanism of the whole monitoring area, and when the monitoring area needs to be expanded, the monitoring area is combined or a plurality of circles of basic monitoring areas is added on the outermost layer of the monitoring area. The method can be widely used for the node displacement of the Internet of Things.

This invention provides a double / single attribute logic grouping method based on the BGP and a device of providers including: setting different CE attribute priorities for the PE belonging to the CE in a VPN, each PE gets the CE attribute priority of the far end PE equivalent to it, when the PE decides that the CE attribute priority of the far end PE equivalent to it is not lower than that of the local, it cancels the virtual connection PW with other PE, which effectively avoids the possibility for generating loops and avoid the error study of MAC address and solves the problem of loops in double-attribute and multi-attribute network grouping.

The invention discloses a method for adaptively controlling the frequency of a self-organizing frequency hopping network. The invention aims to provide a method with good survivability. The method isrealized through the following technical scheme: in the self-organizing frequency hopping network which is composed of a plurality of network nodes with equal status, each node comprises a frequency adaptive control function entity, a frequency hopping sending unit and a frequency hopping receiving unit for transmitting wireless signals through a wireless channel, and a frequency control unit forsending a frequency hopping sequence and a data message to the frequency hopping sending unit and the frequency hopping receiving unit; in the adaptive frequency control process, each node takes a neighbor signal-to-noise ratio and an interference condition sensed by the node as control parameters; and automatically selects the most appropriate frequency set from all frequencies in the working frequency band as a sending frequency hopping frequency set of the node, calculates a new sending frequency hopping frequency set, generates a frequency set notification message, automatically adjusts the sending times of the control message, and adaptively controls the frequency of the frequency hopping network.

The invention provides a system for storing vehicle-mounted information by using a distributed data storage program and a storage method thereof, relates to enterprise applications of distributed data storage software, and in particular relates to a system for storing positional information data of a vehicle-mounted information terminal by using a distributed data storage program and a storage method thereof. The system comprises the vehicle-mounted information terminal, a server cluster, a data processor (DP), a data access processor (DAP) and storage programs, wherein the vehicle-mounted information terminal is installed on a vehicle; the server cluster comprises at least five personal computers (PCs); and the storage programs are installed on the PCs in the server cluster respectively and comprise open source distributed data storage program software Cassandra. The storage method comprises the following steps: 1) configuring the operating environment of a distributed data storage system; 2) establishing a data storage structure; 3) setting the DP and the DAP; and 4) starting the DP and the DAP.

The invention discloses a method for marking the geographical position of a host. The method comprises the following steps: configuring the corresponding relation between an IP address field and a geographical position; generating an IP address of the host according to the configured corresponding relation between the IP address field and the geographical position, and writing the mapping relation between the name of the host and the IP address into a HOSTS file; obtaining the IP address mapped by the name of the host through inquiry of the HOSTS file; and acquiring the geographical position of the host according to the inquired IP address mapped by the name of the host. The invention further discloses a system for marking the geographical position of the host. According to the technical scheme of the invention, the specific geographical position of the host can be quickly inquired without depending on the third-party system.

The invention discloses an ISCSI access load balancing method. The method comprises the following steps: each node monitors the load in each network card of the present node, and when an access request sent by a client is received, the node establishes a connection with the client by using the network card having the minimum load in the present node by using an ISCSI redirection function; when thenode monitors that the load of the network card in the present node exceeds a threshold, the node selects a connection satisfying a balance adjustment condition in the connections of the network cards and disconnects the connection, wherein the balance adjustment condition is that the sum of the load corresponding to the connection and the load of the network card having the minimum load in the present node is smaller than the threshold; if the selection is failed, a node is selected from the system to serve as a resource migration target node, and a storage device is selected form the storage devices of the present node, the connection of accessing the storage device is disconnected, and the selected storage device is migrated to the resource migration target node. By adoption of the ISCSI access load balancing method, when system load balancing is achieved, the utilization of resources can be improved, and it is easy to implement.

The invention relates to the technical field of optical paths of fiber-optic gyroscopes and particularly relates to a three-axis integrated satellite-borne fiber-optic gyroscope optical path. The three-axis integrated satellite-borne fiber-optic gyroscope optical path comprises two light sources, five 1*3 optical fiber couplers, three Y waveguides, three optical fiber rings and three detectors, the two light sources comprise a main light source and a backup light source; the two light sources are respectively welded with a single tail fiber port of one 1*3 optical fiber coupler; the couplers divide an optical signal into three paths; optical signals are respectively provided for three sensitive shafts of a gyroscope; each sensitive shaft comprises a 1*3 optical fiber coupler, a Y waveguideand an optical fiber ring; the Y waveguide and the optical fiber ring form a sensitive loop of each axis gyroscope; the 1*3 optical fiber coupler realizes the light splitting of an interference signal carrying Sagnac phase shift information, the split interference light signals are detected by the detectors, a key device-light source in the light path adopts double-light-source backup, and the light path has no single-point fault and has relatively high reliability.

The application discloses a hyper-converged normal recording system based on distributed object storage, which is used for realizing the full-distribution architecture of video collection, storage andaccess, and satisfying the high scalability, high ease of use and high reliability requirements of normal recording and broadcasting. The distributed video acquisition realized by software has good scalability and good reliability and may cope with the large-scale growth of the classroom with no single point of failure. Videos are stored in distributed object storage by levels in the form of an object, which is high in reliability, strong in scalability, good in retrieval performance, and convenient for videos shared on the Internet.

The invention relates to a distributed quick short message sending system. The system comprises a short message request receiving module, a short message sending module, a short message dispatching module and a short message data storing module; the short message request receiving module is connected with a short message sending side; the short message sending module is connected with telecommunication, Unicom and mobile communication operator systems and starts to send the short message according to the sending state of a short message array to be sent by a dispatching module; the short message dispatching module is used for inspecting the sending state of the short message array to be sent, dispatching a plurality of threads and synchronously sending the short messages, automatically transferring the fault and alarming for error information; the short message data storing module is used for storing the short message which is sent or timely sent as well as storing the information of the short message dispatching parameters. The system has the advantages that the short message can be quickly sent on real time without single point fault; lots of short messages can be sent nearly without delay, and therefore, the scene that lots of real-time short messages are to be sent applies; the fault can be automatically transferred; the computer resources are fully utilized by the thread technology.

The invention provides a distributed cluster file system. The system comprises physical servers and a storage server. Each physical server is provided with a distributed cluster management unit, a cluster logical volume management unit and a virtual case file system (VCFS) unit, wherein the distributed cluster management unit is used for providing heartbeat messages between the physical servers to ensure that all the physical servers in a cluster run normally; the cluster logical volume management unit is used for managing physical volumes, a logical volume group and logical volumes in the cluster; and the VCFS unit is used for storing the disk file data of virtual machines. The storage server is used for providing block-level storage equipment. According to the distributed cluster file system, a global namespace is adopted, file-level data is shared between system platforms, and storage equipment is added, so that the storage capacity, input / output (IO) bandwidth and load capacity of the system can be expanded on line, and the capacity can be reduced on line; and a storage system is prevented from a single-point failure, and can be ensured to run integrally continuously, and storage performance is improved.

The invention discloses a centralized control system and method for electrical appliances based on a wireless network. The centralized control system comprises a terminal control module, a routing module and a main control module, and is characterized in that the terminal control module comprises a first Thread control transponder, the routing module comprises a plurality of control routers, and each control router is internally provided with a second Thread control transponder. The centralized control system performs signal transmission by adopting a Thread network, the Thread network has a strong self-healing mesh network, and there is no single point failure, so that the system is enabled to be more stable. Meanwhile, remote control for the whole system can be easily achieved by networkequipment such as a mobile phone and a computer at any time and any place by being matched with a gateway through setting the function modules. In addition, the centralized control system continuously modifies an optimal path calculation value by using a record learning algorithm, intelligently turns on a preset electrical appliance and does not need to send an instruction manually.

A method for initiating or recovering a media-on-demand session in a distributed media-on-demand network consisting of N streaming nodes (203 ... 204), the method having: a. a first step wherein a client device (201) sends a setup request (251; 256) to a highly available redirector device (221) to request initiation or recovery of said session; b. a second step wherein the redirector device (221) selects a streaming node (203; 204) for the session; and c. a third step wherein the redirector device (221) instructs the client device (201) to use the selected media streaming node (203; 204) for that session. A further distinctive feature is that the client device (201) requests media delivery from the selected media-streaming node either from the beginning (for initiation) or from the last position.

The invention discloses a small computer room monitoring system and belongs to the technical field of monitoring. The small computer room monitoring system comprises a management center module, a power environment module and a terminal receiving module. The management center module is used for commanding and coordinating the power environment module to regulate and control a distributed computer room, the power environment module is used for executing instructions of the management center module and feeding back information, and the terminal receiving module is used for receiving power environment module video and other information transmitted by the management center module and feeding back information to the management center module. The small computer room monitoring system is stable and free of single-point failure, so that people can build a larger monitoring system conveniently; the small computer room monitoring system which is of a distributed structure is convenient to expand, and capable of providing a complete chain of evidences for analyzing accident causes through network integration.

The invention provides a distributed transaction processing method, device and system.The method comprises the steps that after a first-stage transaction in a distributed transaction is executed, the first-stage transaction is submitted; whether the first-stage transaction is successfully submitted or not is detected; if yes, a first notification message with a first detection result indicating that the first-stage transaction is successfully submitted is sent to a control thread, so that the control thread notifies second service nodes corresponding to a second server cluster and a second database cluster to start executing a second-stage transaction in the distributed transaction after receiving the first notification message.According to the distributed transaction processing method, device and system, the interaction procedure between first service nodes and the control thread is simplified, no single point fault problem exists, related requests are few, and occupation to a system and network resources is also little.

The invention relates to an information transmission system and a method in a wind power generation mode. The information transmission system in the wind power generation mode comprises an information management center, which is used for transmitting a control command to a wireless sensor node through a wireless gateway and receiving monitoring data from the wireless sensor node through the wireless gateway; the wireless sensor nodes are arranged at each monitoring part of the wind power generation mode, are used for collecting monitoring data according to the control command and then transmitting the monitoring data to the wireless gateway; and the wireless gateway is used for receiving the order command, transmitting the order command to each of the wireless sensor nodes and transmitting the monitoring data to the information management center. According to the information transmission system in the wind power generation mode, the two-way information transmission system in a wind power generation mode is achieved, so that the monitoring information can be transmitted from the wireless sensor nodes to the information management center to achieve the purpose of information collection and the control command can be transmitted from the information management center to the wireless sensor nodes to achieve the purpose of controlling the collection points.

The invention discloses a distribution type machine room monitoring system, and belongs to the technical field of monitoring. The system comprises a management center module, a dynamic environment module and a terminal receiving module. The management center module is responsible for commanding and coordinating regulating and control of the dynamic environment module over a distribution type machine room, the dynamic environment module is responsible for executing commands and feedback information of the management center module, and the terminal receiving module receives video and other information, transmitted by the management center module, of the dynamic environment module and feeds back the information to the management center module. The distribution type machine room monitoring system has the advantages that the system is stable and has no single-point failures, a client can conveniently erect the monitoring system within the larger range, due to the distribution type structure, expanding is convenient, network integration is achieved, and a complete evidence chain can be provided for analyzing the accident reason.

The invention belongs to the technical field of monitoring and discloses a monitoring system applicable to machine rooms of communities. The monitoring system comprises a management center module, a dynamic environment module and a terminal receiving module. The management center module is responsible for conducting and coordinating the dynamic environment module to control distributed machine rooms, the dynamic environment module is responsible for executing commands of the management center module and feeding information back, and the terminal receiving module receives dynamic environment module videos and other information sent by the management center module and feeds information back to the management center module. The monitoring system has the advantages of stability, no single point of failure, convenience for clients to set up larger-scale monitoring systems, distributed structure, convenience in extension, network integration and capability of providing whole evidence chains for analysis of accident causes.