34results about How to "Implement failure recovery" patented technology

The invention provides a layered exchange and control method for real-time monitoring system data by power dispatching. A CIM/XML (common information module)/(extensible markup language) interface, a CIM/E interface and other models of the electric power subsystems are processed and combined to establish a device model of the large power grid, therefore, perfect interaction of various application data including SCADA (supervisory control and data acquisition) real-time data, WAMS (wide area measurement system) data, EMS (enhanced message service) advanced application data and dynamic security analysis warning information can be realized, and power system control commands among the systems can be smoothly transmitted and executed. A data acquisition and exchange function can support multiple communication protocols, multi-application, multi-type and large amounts of data, satisfy the real-time requirements of all transmission environments, and provide comprehensive multi-scale real-time operation information of the power grid for other application analysis systems. A close integration mode is used to realize application integration and each subsystem and application can be monitored and managed, so a uniform access interface and a man-machine interface can be ensured to provide for each application system.

The invention discloses a method for fast constructing Hadoop cluster based on container cloud technology. The method comprises the following steps: (1) determining selected physical server number according to actual demand, regarding the entire selected physical server as a cluster, and assigning a master node of the cluster, and distributing actual deployment nodes, namely as the slave node; (2) installing a CentOS7 operation system for each physical server in the cluster; (3) installing and starting a container orchestration service Kubernetes and the network service Flannel on each master node of the cluster by use of a yuminstall command; (4) installing MySQL database, a mirror image warehouse service and management portal; (5) logging a management portal interface by use of an administrator account, establishing a user; and logging the management user by use of the established user, creating the cluster environment of the user, and constructing a Hadoop cluster.

The embodiment of the invention provides a method and equipment for many-one reducibility scene multicast data transmission. Data transmission stability can be well improved. The method comprises the steps that provider periphery PE equipment receives first multicast routing information sent by first provider equipment and second multicast routing information sent by second P equipment; the PE equipment carries out route optimization on the first multicast routing information and the second multicast routing information, and a route corresponding to the first multicast routing information is determined as mainly-used multicast routing; the PE equipment determines first identification according to the first multicast routing information corresponding to the mainly-used multicast routing, and the first identification is used for identifying the first P equipment as a main root node; and the PE equipment sends the multicast routing information corresponding to the PE equipment to the first P equipment and the second P equipment respectively.

The invention discloses an FPGA (field programmable gate array)-based failure recovery method. The FPGA-based failure recovery method can be used for monitoring the working condition of an FPGA after power on; if the phenomenon that logic loading fails or breaks down and the like occurs, the logic is reset by a hardware watchdog; if software embedded into the FPGA is loaded unsuccessfully or breaks down, the software is reset through logic software so as to realize failure recovery of the module, and moreover, operations, such as burning and configuration, on the FPGA are not influenced. The method is suitable for the conditions that failure recovery is needed, so that the work reliability is improved; moreover, the method has a general function, and can be suitable for various Xilinx FPGA-based failure recovery conditions.

The invention provides a TWDM-PON system with multi-point fault protection capability, fault isolation is effectively realized; the condition that the whole body is started is avoided; More importantly, more importantly, the structure is simple, The method comprises the following steps: designing an internal structure of a ring network and a remote node RN so that the network can realize fault recovery even under the condition that only one remote node RN can work normally; N: realizing feeder optical fiber Comprehensive protection, the reliability of the network is greatly improved; Besides,directly transmitting the signal to the target optical network unit ONU along the star network. Compared with an annular structure, no other nodes are needed, the transmission delay and the link lossare reduced, the network unit ONU adopts a bus type topological structure, the working optical fiber and the protection optical fiber transmit signals at the same time, the coupler instead of the optical switch is used along the way, the ONU is automatically protected without switching, and meanwhile the network cost is reduced.

The invention provides a multi-layer network fault recovery method, which obtains multi-layer network fault information in real time; determines the fault recovery initiation layer according to the fault information; performs intra-layer fault recovery; performs inter-layer fault recovery; thereby realizing multi-layer network fault recovery. Layer network failure recovery. The present invention is used to realize the coordination between integrated control of multi-layer transmission and multi-layer network fault recovery strategies, which can eliminate the incoordination of recovery actions caused by lack of information interaction mechanism between multi-layer fault recovery strategies and Many disadvantages caused by this improve the multi-layer network fault recovery capability.

The invention discloses a switch RSTP convergence enhancement method, and the method comprises the steps that a link fault detection module of the switch configures local link fault detection configuration information, is linked with an RSTP module of the switch, and carries out link fault detection negotiation with an opposite-end switch according to the local link fault detection configuration information to detect communication link fault information; the communication link fault information is quickly notified to other switches in the network through a RUN message, the RUN message packaged with the communication link fault information and sent by other switches is received at the same time, and the RUN message is analyzed to quickly complete the corresponding processing of the RSTP protocol so as to achieve the purpose of quickly switching the link and further realize fault recovery. According to the method, the link state problem and the possible link single-pass problem can be sensed in time, the whole network and a processing mechanism are quickly notified through the RUN message, the convergence of the whole network and the link switching time are accelerated, and the RSTP protocol convergence time of the switch is improved from the second level to the millisecond level.

The invention discloses a data processing method and related equipment, which can determine a fault recovery allowable duration when a target node executes a target job, determine a fault recovery parallelism degree matched with the fault recovery allowable duration, divide check point data into M partition data when the target node outputs the check point data, wherein m is the same as the value of the fault recovery parallelism degree, and when the target node has a fault in the process of executing the target job, read the M partition data in parallel according to the fault recovery parallelism degree so as to perform fault recovery. The tolerable time of check point write-out and recovery can be estimated according to the dependency relationship and the execution duration of the historical jobs, and then the minimum parallelism degree capable of meeting the fault recovery requirement and the job response duration is estimated, so that the resource consumption is reduced to the maximum extent, and fault recovery based on the check point data can be effectively realized.

A redundant electronic control unit of an aviation piston engine realizes self-rescue of the RAM by optimizing the redundancy of the RAM data storage structure in the electronic control unit. When the original data is destroyed, the RAM data can be restored and corrected in time. Dividing the program into multiple modules and setting no-operation instructions and breakpoints between different modules can ensure that the program can be executed smoothly, and can realize rollback and fault recovery. The program execution is monitored through the counter, and when the program is in disorder, it can be restored in time to ensure the normal operation of the program. The electronic control unit can ensure the accuracy of sensor data collection by taking the average value of multiple samples of the sensor. The aviation piston engine redundant electronic control unit of the present invention can improve the safety and stability of engine work in application.