45results about How to "Reduce business impact" patented technology

The invention discloses a method of cleaning database history data and a system of the method of cleaning the database history data. The method of cleaning the database history data includes the following steps: obtaining a data cleaning task from a data cleaning task queue established by a sub-data base of the database, wherein the data cleaning task comprises history data index information needing cleaning, the data base is divided into more than two sub-data base, and each sub-data base of all the sub-data base corresponds to the data cleaning task queue established by each sub-data base; cleaning history data needing cleaning in the database according to the obtained history data index information of the data cleaning task. Due to the technical scheme, efficiency of cleaning the data is improved.

The invention discloses a method for scheduling an application task based on a big data behavior. The method comprises the steps that current resource state information of a server is collected, if itis determined that current resources of the server are sufficient according to the resource state information, a plurality of threads are started, the threads are called to execute a plurality of tasks concurrently, and no dependency relationship exists among the tasks; Wherein the dependency relationship at least comprises one of the following items: the execution sequence of the tasks has a sequential order and a dependency relationship triggered in parallel at the same time, and the execution dependency of one task depends on a relationship that the previous task is executed; And accordingto the historical records of the execution durations of the tasks and the resource state information, the execution sequence among the tasks without the dependency relationship is scheduled in real time, and all the tasks in the plurality of tasks are traversed until all the tasks are executed. The task execution efficiency can be improved, and the service influence caused by the fact that the task is not executed completely is reduced.

The present invention discloses a data processing method, device and system, and relates to the Internet technology field. The problem can be solve d that a current limiting strategy cannot satisfy demands of business sides at present. The method comprises: obtaining the maximum capacity of user access being subjected to concurrent processing by the server of a business side; establishing a business data model comprising concurrence access volume corresponding to different business data ranges; establishing a resource restriction model comprising access restriction probability corresponding to different business data ranges according to the maximum capacity and the business data model, wherein the access restriction probability is obtained according to the maximum capacity to allow the server of the business side to perform restriction of user access in different business data ranges according to the business strategy of the business side; and sending the resource restriction model to the server of the business side to allow the server of the business side to perform current limiting operation according to the resource restriction model. The data processing method, device and system are mainly suitable in the scenes of online transaction.

The embodiment of the invention discloses a method for processing port failures based on a link aggregation control protocol LACP and an LACP terminal. The method comprises the following steps: a first terminal obtains the failure information of the first port in a first port aggregation group; the first terminal adopts the second port in the first port aggregation group as a failure information transmission port; and the first terminal adopts the second port to transit a failure transfer message to a fourth port which is corresponding to the second port in a second port aggregation group of the second terminal so that the second terminal stops to transfer the message of a third port which is corresponding to the first port in the second port aggregation group, wherein the failure transfer message is an LACP message, and the reserved field in Terminator information field of the LACP message carries the failure information. The scheme of the embodiment of the invention improves the failure transfer performance through the optimization of the LACP, does not need depending on other protocols or technologies, and cannot be limited by the network.

The invention relates to an IP DSLAM (1) traffic failure diagnosis method based on a network processor, using traffic flow mirror technique. An on-line traffic flow port to be diagnosed is confirmed, and spare traffic flow ports for diagnosis are selected and set in one group. When the network processor (13) sends the traffic flow port to be diagnosed to the on-line traffic flow port, the said traffic flow is copied in multicast mode and switched to the spare traffic flow port after script treatment. The copied traffic flow B or C are analyzed and an alarm or statistical information message is generated by treating the traffic flow D. The said invention fully uses linear speed processing ability of network processor, the CPU intervention is less and the running of device is not affected, at the same time all the diagnosis aims at the traffic flow port, and the influence on the affair of port to be diagnosed is little. The diagnostician is facilitated using near end, far end and script assistant specific diagnostic mode and the diagnose requirement is greatly satisfied.

The invention discloses a main equipment election method and device and a machine readable storage medium, which are used for reducing the influence of a VRRP protocol on service continuity in the main equipment election process. According to the invention, after the main equipment receives VRRP notification messages sent by other main equipment, the priorities, the key resource weights and the IPaddresses carried by the notification messages are compared, the main equipment election is carried out based on the priority sequence, and the new main equipment is determined. According to the method, the decision condition for judging the key resource weight on the main equipment is added in the main equipment election rule, and the main equipment state of the main equipment with the large keyresource weight is maintained preferentially, so that the influence on the current ongoing service is reduced, and the service continuity of client hosts or users is ensured as much as possible.

The invention discloses a method, device and system for generating data traffic. The method comprises the following steps: acquiring whole-network data traffic, wherein the whole-network data traffic comprises real traffic of a network and deception-induced traffic, and the deception-induced traffic is used for being confused with the real traffic so as to reduce the probability of network attack success; analyzing the whole-network data traffic; determining whether to adjust the deception-induced traffic according to the analysis result; and generating target deception-induced traffic according to an adjustment result, and mixing the target deception-induced traffic with real traffic. According to the invention, the technical problem that an existing traffic spoofing technology confusing network attacks is easy to be broken by attackers and cannot disperse too much attention of the attackers is solved.

The invention discloses an ORACLE database active-active mechanism deployment method and device, which can be used in the technical field of data processing, and the method comprises the following steps: building a second database; creating Topic which needs to be subjected to table synchronization by utilizing distributed message service; setting a database synchronization component, and introducing a transaction application in a Maven mode; according to a transaction generated by the transaction application, judging a transaction generation park, a database table and an operation type, and selecting a database for access according to a judgment result; after the transaction generated by the transaction application is executed, querying full-amount information of data to be synchronized by quoting an annotation provided by a database synchronization component, and writing the data to be synchronized into a specified Topic by the database synchronization component; creating a to-be-synchronized data consumer group, subscribing the specified Topic by the consumer group, pulling the to-be-synchronized message, and persisting the message to the current database according to the data in the current database and the data timestamp in the to-be-synchronized message.

The invention provides a network attack processing method and device The network attack processing method comprises steps of obtaining a first IP address connected to a server and a connection number corresponding to the first IP address, determining whether the connection number corresponding to the first IP address is greater than or equal to a first threshold value, if yes, shielding a source IP address connected to the first IP address. The embodiment of the network attack processing method and device not only guarantees integral stability of the server and the network, but also reduces influence on a service of an attacked IP address through shielding the source IP address, and reduces the influence on the network access.

The invention relates to a database data consistency detection method and device, a medium and electronic equipment, and the method comprises the steps: obtaining a first database connection of a source database, configuring an isolation level as a serialization isolation level, and closing automatic submission; starting a data detection transaction in the first database connection; acquiring total data of the source database at the current moment and first database log information; acquiring synchronized second database log information in a target synchronization system of the target database; based on the first database log information and the second database log information, obtaining a corresponding synchronization log from a source end synchronization system of the source database; analyzing the synchronization log into an SQL statement to be executed in a target database; acquiring total data of the target database after the execution is finished; performing data consistency detection based on the total data of the target database and the total data of the source database; and submitting a data detection transaction after the data consistency detection is completed.

The embodiment of the invention discloses a memory fault processing method and device, equipment and a storage medium, and belongs to the technical field of computers. In the embodiment of the invention, the fault analysis result is obtained by analyzing the historical fault information, and then the fault of the memory is repaired according to the fault analysis result, so that the memory fault can be analyzed more accurately. In addition, according to the scheme, memory fault repair can be started without cold reset, that is, memory faults can be repaired in time, system downtime is prevented, and service influences are reduced.

The invention discloses an information sending method, device and equipment and a readable storage medium, and relates to the field of financial institutions, the method comprises the following steps: obtaining to-be-sent theme information from a preset configuration center at intervals of a preset time period; obtaining a preset queue algorithm, and determining a target sending mode of the to-be-sent theme information according to the queue algorithm; and sending the to-be-sent theme information according to the target sending mode, so that a preset receiver receives the to-be-sent theme information. According to the invention, the number of times of information sending failure is reduced, and the service influence caused by server failure is reduced.

The invention relates to the field of Fintech. The invention further discloses a message receiving and sending method and device. The method comprises the following steps: a first proxy server determines group information to which a first client belongs, the first proxy server acquires the agency condition of a second proxy server from a registration center, wherein the current agented client of the second proxy server comprises a second client with group information, the agency condition is used for indicating the condition of the second client, and the first proxy server determines a third proxy server agenting the first client according to the agency condition of the second proxy server. According to the invention, the received and sent messages with the same group information are distributed to different proxy servers in a relatively balanced manner, and the service influence caused by the failure of the proxy servers is reduced.

The invention provides a frequency conversion-based out-of-service alarm analysis processing method and system, and the method is characterized in that the method comprises the steps: obtaining out-of-service alarm data in whole-network alarm data; Performing frequency conversion on the out-of-service alarm, and calculating a cell out-of-service rate; Grading the cell according to a preset value grading rule to obtain a value out-of-service rate index of the cell; According to a cell scene dimension division standard, carrying out statistics on a cell service rejection rate and a value servicerejection rate index of the cell; And according to a preset rectification rule, two dimensions are counted according to time and the number of discrete times to set a cell rectification library, cells in the cell rectification library are continuously judged, and early warning analysis is carried out on the cells with more warehousing times. Based on intelligent analysis and early warning of bigdata influencing service alarm, high-value cell accurate priority processing is realized; And establishing a rectification library mechanism, and carrying out long-term effective tracking and early warning on a cell processing effect.

The invention discloses a self-adaptive flow control method and device, computing equipment and a storage medium. The method comprises the following steps: acquiring the historical registration request quantity of each AMF network element in a large area and the equipment processing capability data of an IWF network element; counting the number of first registration requests initiated by each AMF network element in the current period; calculating a second registration request number corresponding to each AMF network element in the next period, and issuing the second registration request number to each corresponding AMF network element, so that each AMF network element deploys a flow control strategy according to the second registration request number. According to the method, the IWF network elements are periodically predicted and the number of the second registration requests is updated in real time without manual intervention, so that the processing capacity of each AMF network element in a signaling storm scene is maximized, the IWF network elements are effectively protected, the interruption of services in a large area is avoided, and the influence of the services is reduced to the greatest extent.