133 results about "Data contention" patented technology

The invention provides a knowledge base and business system collaboration method under an AI platform, and belongs to the technical field of AI platform digital data processing.The method includes the steps that triple semantic analysis is conducted on query by constructing a multi-level semantic vector representation module, a parallel data retrieval engine is started, vector retrieval, graph reasoning and real-time data pulling are executed at the same time, and the query efficiency is improved; establishing a dynamic confidence evaluation mechanism to evaluate the quality of the data source, executing a weight distribution algorithm based on reinforcement learning, dynamically calculating the weight of the data source according to a query type by adopting a graph convolutional network intelligent routing decision model, and implementing multi-source data fusion and consistency verification to solve data conflicts through a weighted voting mechanism. An intelligent result sorting and filtering system is established, a multi-dimensional evaluation strategy is adopted to output high-quality answers, a continuous learning and feedback optimization loop is constructed, system performance is continuously optimized through active learning and a graph shortest path algorithm, and the technical problem that knowledge base data and a service system cannot effectively and uniformly make decisions is solved.

The invention discloses a PLC scheduling algorithm based on a multi-core soft real-time operating system, and relates to the technical field of voltage control, and the algorithm comprises the following four steps: 1, starting a target industrial control system, completing hardware self-inspection, memory allocation and communication module starting, and sending an initialization ready signal after confirming that an assembly is normal; 2, according to PLC task functions and real-time performance, four types of standardized tasks are divided, priority levels are defined, sub-priorities are configured, a dependency relationship is identified, and a task chain is established; 3, counting the number of tasks and task chains, distributing resources by using multiple binding cores, and constructing a global ready queue to adapt to dynamic scheduling; and 4, monitoring an interrupt event through an interrupt manager, realizing high-priority task preemption by semaphores, and allocating tasks to corresponding CPU cores, so that the problems of soft real-time multi-core task preemption and data conflict are solved, the scheduling real-time performance and the data security are improved, the system failure rate is reduced, and the industrial control efficiency and stability are improved.

The invention provides a data synchronization method and device based on Spring Batch and a large model. Synchronously initializing a large model micro service and a Spring Batch framework, and establishing a model service hot standby mechanism; obtaining data source features through a metadata scanning engine driven by a large model, and generating an adaptive synchronization strategy; optimizing a query plan based on the synchronization strategy, dynamically controlling data fragmentation and executing streaming preprocessing and semantic mapping conversion; a target table structure is adapted, a three-level conflict detection mechanism is adopted to eliminate data conflicts, and meanwhile transaction processing and batch writing logic are optimized; the data quality is verified through three-level verification, performance parameters are dynamically adjusted and optimized, and fault-tolerant recovery and resource elastic expansion and contraction are achieved; metadata are synchronized, an execution report is generated, knowledge base self-learning is completed based on task full-link indexes, and a data consanguinity map and a reusable processing case library are constructed. According to the scheme, diversified and dynamic enterprise data synchronization requirements can be met.

The invention discloses a service entity intelligent prediction system and method oriented to multi-dimensional decision support, and relates to the technical field of service prediction. The system comprises a service multi-source data screening module, a service multi-source data correction module and a service entity dimension index output module. According to the method, data fusion is carried out on the multi-source data of each dimension index, an adaptive data screening strategy is selected according to the data conflict detection condition in the process, and then data range verification is carried out on each to-be-input data determined by the data screening strategy; and determining whether to perform data deviation determination based on the verification result so as to determine whether to execute data correction measures, finally, after data fusion, inputting the determined to-be-input data into the prediction model, and outputting each dimension index after index data range determination and a determination result prompt. The intelligent prediction precision of the service entity is improved, and the problem that the intelligent prediction precision of the service entity is reduced in the prior art is solved.

The invention discloses a foreground and background state information intelligent synchronization method in project management. The method comprises the following steps: S1, constructing a three-layer architecture of a foreground system, a state storage center and a background system; s2, initializing a state; s3, realizing state synchronization of the foreground system and the background system through at least one synchronization mode; s4, constructing a version control and conflict resolution mechanism, when the state of each project is updated, the foreground system and the background system carry version numbers, checking the consistency between the client version number of the foreground system and the server version number of the background system when the update is submitted, if the client version number of the foreground system and the server version number of the background system are consistent, executing state updating and progressively increasing the version number, and if the version number of each project is not consistent, executing state updating and progressively increasing the version number; if yes, judging that the data conflict exists, automatically executing a preset conflict solving strategy, and broadcasting a resolution result to a foreground system and a background system. The method has the advantages that the item switching state is not lost, background state change and foreground real-time response are achieved, and conflicts are automatically eliminated without manual intervention.

The present application relates to the technical field of parallel processing, and provides a parallel processing method and system for batch write-off tasks. The method comprises: extracting a plurality of pieces of query data of a preset write-off window, comprising external write-off codes and write-off services; performing file import on the basis of the data, and generating transaction files with unique identifiers; when file name identifiers are verified, parsing the external write-off codes, and generating fund files; generating pending write-off files by means of procurement-settlement mapping authentication; identifying request time identifiers to identify write-off conflicts, and generating conflicting file clusters; updating the configuration of a parallel processing module on the basis of the conflict condition; and utilizing the updated module to process the conflicting file clusters in parallel. The present application solves the technical problems of data conflict, irrational resource allocation and low processing efficiency caused by improper parallel processing in batch write-off tasks, and realizes real-time extraction, verification, parsing and authentication of write-off data, thereby ensuring the data accuracy and consistency.

The invention relates to an inventory data synchronization and consistency verification system supporting network disconnection fault tolerance, and discloses an inventory data synchronization and consistency verification system which realizes continuous recording of inventory operation during a network disconnection period through network state detection and local cache, performs batch transmission after network recovery through a synchronous uploading and verification mechanism, and ensures data integrity. According to the inventory data synchronization and consistency verification system supporting network disconnection fault tolerance, data conflicts are automatically solved through a conflict detection and strategy processing module, a difference report is generated through consistency verification, and secondary synchronization is executed, so that safe synchronization and final consistency guarantee of inventory data are achieved, and it is ensured that the service process is continuous and stable. The system is characterized by comprising a network state detection module, a local cache module, a transaction log management module, a synchronous uploading module, a conflict detection module, a conflict resolution strategy module, a consistency verification module and an abnormity alarm and manual intervention interface module.

The invention relates to the technical field of data management, in particular to a cross-platform advertisement putting and SEO collaborative optimization system which comprises a resource time sequence scheduling module, a keyword screening module, a putting path construction module, a data conflict comparison module and a content archiving and auditing module. According to the invention, through association mining of multi-platform advertisement data and user behaviors, real-time mapping of time sequence distribution of put resources and user group states, and dynamic integration of budget weights and audience portraits in a keyword selection process, a multi-parameter interactive screening mechanism is formed for advertisement paths; text similarity judgment and conflict source tracing are synchronously introduced into channel content, an archiving standard is jointly defined by content change frequency and source credibility, data in each link automatically adjusts an attribution boundary according to a change trend, multi-source data validity is accurately distinguished in an archiving process, and structure-level linkage is realized by advertisement circulation, content integration, path adaptation and conflict detection. And the consistency and uniqueness of data updating and archiving are guaranteed.

According to the heterogeneous memory management system and method oriented to the sparse hybrid expert model, the medium, the program product and the terminal provided by the invention, the interference of cold data on hot data access delay is eliminated by constructing a double-Bank physical isolation architecture in the physical heterogeneous storage module, so that the bandwidth contention of the cold and hot data is avoided; the cold storage unit adopts a low-power-consumption memory and supports a sleep mode, so that near-zero-power-consumption storage of cold data is realized; in the semantic prefetching unit, the prefetching hit rate is increased to 89% by adopting a cosine similarity matching method, so that a dynamic data migration mechanism with a high hit rate is constructed; and the gating collaborative migration engine triggers data migration, and the calculation continuity of the calculation module is guaranteed through DMA control.

The invention relates to the technical field of remote measurement, in particular to a very simple protocol efficient and high-density meter reading method which comprises the following steps: acquiring data frame content, eliminating identity fields, generating a simplified format identifier, constructing a peak shifting scheduling list, detecting sequence consistency, generating a frame supplementing instruction, and dividing synchronous triggering groups to sort meter reading records. According to the method, identity field compression is replaced, frame body space proportion conversion is introduced, data structure content is simplified, a rhythm dislocation sequence is constructed in combination with time migration to achieve high-density node scheduling, frame loss behaviors are screened based on serial number jumping, a frame compensation response task list is automatically constructed, and a synchronous trigger sequence is constructed in combination with time difference judgment. The volume of transmission data is effectively compressed, the utilization rate of channel resources is improved, peak-shifting uploading between nodes is realized to reduce the probability of data collision, overhigh communication load caused by frequent handshake response and multi-level response is avoided, and the continuity, fault tolerance and rhythm collaboration of meter reading scheduling are enhanced.

The invention discloses a knowledge pushing method and system based on a knowledge graph and intention prediction, a terminal and a storage medium. The method comprises the steps that historical document data and new document data of multiple systems are crawled regularly and preprocessed; a triple in the preprocessed data is recognized and stored in a knowledge graph database, updating is suspended and manual auditing is triggered when new and old data conflicts are found, and updating is executed and changes are recorded after auditing is passed; collecting user behavior data, generating a behavior sequence, combining post information of the user, and utilizing the intention prediction model to output a demand probability; extracting associated knowledge according to the demand probability, and sending a knowledge packet link and a brief description to the user through an agent pop-up window; and receiving and storing feedback information of the user on the pushed knowledge packet, and performing system optimization based on the feedback information. According to the method, multi-source heterogeneous knowledge can be automatically integrated, knowledge updating can be monitored and synchronized in real time, knowledge is actively perceived and pushed based on context, and the utilization efficiency of enterprise knowledge is improved.

The invention discloses a bus communication method and system and a storage medium, and the method comprises the steps: carrying out the verification of an address according to a first address field transmitted in a current time period and a second address field received in the current time period when a bus is in an idle state in the current time period; the second address field and the first address field are obtained by the same communication device at the same time; if it is determined that no address conflict exists, data verification is carried out according to the first data field sent in the current time period and the received second data field; the second data field and the first data field are obtained at the same time; and if it is determined that no data conflict exists, obtaining the right to use the bus in the current time period, so as to send the first data field through the bus. According to the invention, the circuit is simple, the hardware overhead is low, the cost is low, and a multi-host communication mode is supported.

The invention relates to the technical field of artificial intelligence, can be applied to business scenes such as financial science and technology and medical health, and discloses an information conflict detection and data optimization method which comprises the steps of obtaining dialogue information and collecting original data resources, inputting a preset language model to generate an initial response, sequentially completing conflict detection, conflict adjustment, data screening, validity analysis and candidate data weighting by using a multi-discrimination model; and a candidate response set is generated based on the optimized data resources, and a final target response is screened out through logic consistency analysis. According to the method, the multi-layer discrimination model is constructed, data conflict detection, screening and effectiveness analysis are converted into weighted integration, multi-source data resources are systematically optimized, the logic consistency and field applicability of generated content are remarkably improved, and the method is suitable for large-scale popularization and application. The problem of reliability of generated contents caused by data conflicts and stagger quality in the prior art can be effectively solved.

The invention provides a multi-source data integration method, a storage medium, a program product and electronic equipment, and can be applied to the field of financial science and technology. The method comprises the following steps: calling a corresponding adapter to read original data from each data source according to the types of a plurality of data sources; performing quality evaluation on the original data by adopting a random forest algorithm, determining to-be-cleaned data in the original data, and performing data cleaning on the to-be-cleaned data to obtain cleaned data; standardizing the cleaned data and uncleaned data in the original data by adopting a metadata management technology; under the condition that data conflicts exist in the standardized data, conflicting data are arbitrated so as to determine real data from the conflicting data, and the data conflicts represent that logic conflicts exist in data information from different data sources. The problem that a multi-source data integration scheme in the prior art is low in integration quality is solved.

The invention requests to protect a multi-machine system synchronization method based on a CXL shared cache consistent memory. Relates to the technical field of high-performance computing and memory interconnection, at present, a plurality of synchronization mechanisms are provided in a Linux system, and the synchronization mechanisms are designed to face a multi-core and multi-process environment of a stand-alone system and are used for guaranteeing concurrent access of a plurality of processes to shared data. However, the action range of an existing synchronization mechanism is limited in a single system, and for the situation that multiple processes belonging to multiple independent host systems concurrently access data in a CXL shared memory, the cross-host concurrent access requirement exceeds the application range of the existing Linux synchronization mechanism, data competition and inconsistent access are possibly caused by direct use, and the data access efficiency is improved. Even if the system crashes, the cross-host access security on the CXL shared memory cannot be ensured. The invention aims to construct a safe and efficient synchronization mechanism oriented to multi-system parallel access CXL shared memory data so as to realize cross-host multi-process synchronous access to the data in the CXL shared memory.

The invention discloses a real-time processing method and system for distributed data streams, and particularly relates to the technical field of distributed data processing, and the method comprises the following steps: collecting multi-source data streams and marking priorities, combining data types and priorities to carry out adaptive segmentation, carrying out load balancing distribution on processing units, and resolving data conflicts; and incremental integration and caching of hotspot data are carried out, and full-link tracing and dynamic tuning are carried out. The system comprises five modules, namely an acquisition labeling module, a segmentation module, a parallel processing module, an integration optimization module and a tracing tuning module. According to the real-time processing method and system for the distributed data flow, priority processing of core data is achieved, data consistency is guaranteed, the resource utilization rate and the system maintainability are improved, and the processing requirements for high real-time performance and high reliability are met.

This invention discloses an intelligent attendance management system, characterized by comprising a data interface layer, a data processing layer, and an interaction layer. The data interface layer flexibly defines and generates structured attendance cycle rules through a dynamic cycle adaptation module, and synchronizes and standardizes raw attendance data from multiple external data sources through a raw data acquisition module. The data processing layer uses a multi-source data fusion module to resolve data conflicts according to hierarchical adjudication rules, and uses a cross-cycle calculation module to segment cross-cycle events according to cycle rules, and an anomaly judgment module to perform compliance checks and generate result records. The interaction layer provides a visual operation interface and implements role-based fine-grained access control through an identity recognition login unit. The advantages are that it solves the problems of rigid cycles, multi-source data conflicts, cross-cycle event calculation errors, and numerous invalid reminders in traditional attendance systems, thereby improving the accuracy, authenticity, and efficiency of attendance management.

The invention discloses a translation hardware acceleration platform and method based on an SGRU, and relates to the technical field of machine translation, and the platform comprises an accelerator which comprises a control module, an input and output module, an Embedding layer, a distributed pulse cache module, a model parameter cache module, a calculation module and a gating calculation core; the input and output module is respectively connected with the control module, the Embedding layer, the model parameter cache module and the PC terminal; the calculation module is respectively connected with the control module, the Embedding layer, the distributed pulse cache module and the model parameter cache module; the gating calculation core is connected with the control module and the distributed pulse cache module. A calculation heterogeneous design of a calculation module and a gating calculation core is adopted, data conflicts are avoided through distributed caching, dynamic loading and updating are carried out based on a model parameter caching module, overall control is carried out according to an SGRU-based Seq2Seq model, and high-performance and low-power-consumption translation can be realized.

The invention discloses an off-line multi-scene collaboration and data synchronization mechanism and system based on a light guide film. State data of a user in multiple application scenes such as health monitoring, safety protection, social service, virtual world and smart home are subjected to local temporary storage, priority ranking and intelligent synchronization after network disconnection recovery through a unified identity authentication center. The unique identity feature code is locally cached after the user registers for the first time, and the user does not need to depend on online authentication in an offline state. The system comprises an off-line data temporary storage module, a priority ranking module, an off-network recovery synchronization module, a conflict arbitration module, a storage space management module and a manager off-line response module. When the network is interrupted, state change is temporarily stored in a local circular queue according to the priority (the highest life safety class), sequencing is carried out in real time during writing, and low-priority data is preferentially discarded when the space is full (coverage operation is executed at the head of the queue). And after the network is recovered, reporting according to a priority sequence, and reporting overtime retry by adopting a fragment transmission and breakpoint resume mechanism. The multi-device offline conflict is automatically arbitrated according to a version number, a timestamp and a device priority rule, and a conflict log retention period can be set. And when the supervisor is offline, automatically electing a temporary supervisor, and synchronizing data after recovery. According to the method, the problems that multi-scene state data are lost when the network is interrupted, off-line operation cannot be coordinated, and data conflicts are generated after recovery are solved, and a complete closed loop of online real-time synchronization, off-line temporary storage and supplementary transmission and high availability of a manager is realized.

The application relates to a multi-serial-port concurrent data acquisition method, equipment and medium, which realizes dynamic mapping of port numbers and card IDs by querying the card ID when a PC terminal processing and display module is started and a card is plugged in or unplugged, establishes one-to-one communication connection, and reduces communication data conflicts. The card supports concurrent mode, polling mode and active trigger mode for acquisition work to adapt to different application requirements. The PC terminal processing and display module supports image and table display and can generate a report to intuitively show the data acquisition condition. The acquired data can be calibrated at a board level and a channel level, and the acquired data is corrected according to standard values. Compared with the prior art, the application has the advantages of reducing data conflicts, strong compatibility and strong intuitiveness.

This application provides a method, apparatus, device, medium, and product for multi-threaded concurrent log reading. It relates to the field of data processing technology. The method includes: controlling one thread to acquire a mutex lock through a preset mutex lock management mechanism; determining whether concurrent readability conditions and overall readability conditions are met; in response to both conditions being met, determining whether logs can be read from a preset buffer based on a preset log header threshold and the current preset buffer length; if it is determined that logs can be read from the preset buffer, verifying whether the current log file status is normal based on the determined current total log entry length, a preset file layout threshold, and the current preset buffer length; in response to successful verification, updating the global two-pointer flag value and controlling the thread to release the mutex lock through the preset mutex lock management mechanism. The method of this application achieves safe log reading for multiple concurrent threads, significantly improving concurrency performance in a multi-threaded environment while effectively avoiding data conflicts.

The invention discloses a processing method of a multi-modal operating system, belongs to the field of embedded operating systems, and mainly realizes multi-modal collaborative symbiosis of different operating systems on the same hardware platform through firmware fusion compiling, hardware resource sharing and application layer parameter unified management. The same hardware product supports the ecology of a plurality of different operating systems and supports the different operating systems to process respective skilled tasks in parallel; through firmware fusion compiling and hardware resource sharing, code redundancy and waste of hardware resources are effectively reduced, the development cost is reduced, and efficient utilization of system resources is achieved; through unified management of application layer parameters, synchronization is combined with an atomization read-write and synchronization mechanism, consistency of data states among different operating systems is guaranteed, logic errors caused by data conflicts are avoided, and data reliability is improved.

This application discloses a data conflict resolution method based on hybrid search, belonging to the field of data retrieval. The method includes: determining the domain authority evaluation value and data richness evaluation value of each data source in the dataset under the retrieval target, based on the retrieval target of the target data source in the dataset; determining the comprehensive confidence evaluation value of each data source under the retrieval target based on the domain authority evaluation value and the data richness evaluation value; and determining at least one target data source from the dataset based on the comprehensive confidence evaluation value of each data source to obtain retrieval results for the retrieval target. This application improves the reliability of data retrieval under multi-source conflict by constructing a dual evaluation mechanism of domain authority and data richness for comprehensive confidence judgment and conflict resolution.

This invention belongs to the field of RS485 wired communication technology, specifically relating to a data conflict communication method based on RS485 communication in automatic measurement. This invention aims to solve the problem that when the RRU polls data from the MCU on the PPA board, the MCU on the PA board always returns a packet within a fixed time frame, while the RS485 packet return time is unpredictable, leading to unstable data transmission. This invention provides a data conflict communication method based on RS485 communication in automatic measurement. Specifically, the MCU on the PA board periodically collects the initial data of the data packet through a main loop function and saves this data in a global variable. It periodically updates the collected initial data and saves it to the global variable, determining if there is a serial port interrupt. When the MCU on the PA board receives a data packet from the RRU via a serial port interrupt, it determines whether the data packet is a polling packet through protocol parsing. In the main loop task, the global variable flag is checked to ensure that the RS485 data bus is not disrupted by delayed data packets, thus providing a data conflict communication method based on RS485 communication in automatic measurement.

The invention relates to the technical field of land value evaluation data processing, and discloses a land value evaluation and circulation transaction system based on multi-source heterogeneous data fusion, which comprises the steps of converting policy texts, environment monitoring and market transaction data into standardized influence vectors, calculating an information unbalance degree through geographic grid projection to represent data conflict intensity, and calculating a data conflict degree; according to the method, internal contradictions of multi-source data are spatially developed, conflicts masked in traditional assessment are quantified into operable risk indexes, meanwhile, dynamic confidence degree modulation is achieved in combination with transient feature analysis, and the reliability of the system is improved. Land value judgment has trend insight and risk early warning capabilities, and a high-credibility decision-making basis is provided for circulation transaction.

This invention provides a cross-scenario data migration and adaptation method for intelligent agents in life services. The method includes extracting common dimensions across different scenarios to construct a shared feature vector space; converting unstructured labels from the source scenario into mathematical vectors; calculating the influence of source domain data on the target scenario and constructing a cross-domain attention network; based on an initial migration strategy obtained through weighted migration, adjusting the weights in reverse according to real user feedback data in the target scenario; defining a loss function and a comprehensive evaluation index, and updating parameters using gradient descent; when new and old data conflict, setting a time decay factor based on the principle of prioritizing real-time data, and if behavioral records exist in the target scenario within a specified time period, forcibly setting source domain weights and directly adopting the target domain profile. This invention achieves efficient migration and adaptation of user preference data from the source domain to the target domain through multiple stages, including constructing a common space, standardizing source data, weighted migration, closed-loop optimization, and conflict circuit breaking.

This invention discloses a fault handling method based on a dual-core lockstep processor. The dual-core lockstep processor includes a master processor and a slave processor. The dual-core lockstep processor receives data from the last four stages of the processor's pipeline via a fault detection module, outputting a fault flag signal and a fault program pointer. A front-end fault handling module receives the fault flag signal and the fault program pointer, performing a pipeline-level rollback. A back-end fault handling module controls the handshake and selection signals of the last four stages of the pipeline and executes a virtual write-back fault-tolerance mechanism, managing the operating status of the last four stages of the pipeline and the instruction fetch stage pipeline. This prevents data conflicts during superscalar processor fault handling, thereby achieving full coverage of fault handling while maintaining low time and area overhead, and simultaneously achieving high performance and high reliability. This invention can be widely applied in the field of dual-core lockstep processors.

The invention discloses a power grid heterogeneous data fusion processing method and assembly, and relates to the technical field of power informationization, and the method comprises the steps: receiving original data from a plurality of heterogeneous data sources, carrying out the automatic recognition of the heterogeneous data sources, extracting multi-dimensional features, generating feature vectors, determining the types of the data sources through template matching, and carrying out the fusion processing of the data sources. The method comprises the steps of extracting data mode information, executing standardized conversion according to a preset conversion rule, performing fusion processing on standardized data, calculating comprehensive similarity, executing time sequence alignment, detecting and resolving data conflicts, organizing fused data according to a unified data model and providing data services. Through intelligent data source identification, dynamic data conversion and multi-dimensional data fusion, the technical problem of power grid heterogeneous data integration is effectively solved, the data integration efficiency and the data quality are improved, the system maintenance cost is reduced, and good universality and expandability are achieved.

The invention relates to the technical field of knowledge base fusion, in particular to a knowledge base data fusion association method and system based on an AI large model, and the method comprises the steps: firstly collecting original data through a connector, then carrying out cleaning and duplicate removal processing, then carrying out vectorization alignment on field names, extracting entities and attributes through the AI large model, and carrying out data fusion association on the entities and attributes; according to the method, entity conflicts are solved through a density clustering algorithm and an entity similarity threshold value, a relation triple is generated, unified identifiers are distributed for entities and attributes for standardization processing, data conflicts are detected through variable coefficients, parameters are dynamically adjusted to optimize a fusion strategy, and finally a graph database is utilized to construct a vector index and provide a graph downloading interface. According to the method, on the basis of realizing data fusion, the problem that reasons are difficult to discriminate and solutions are difficult to give in a targeted manner when data conflicts occur in an existing database fusion method is ingeniously solved, the accuracy and reliability of knowledge base data fusion are remarkably improved, and the method has important innovative significance and practical value.

The application provides a database cluster management method, device, equipment, storage medium and product, which are applied to a database cluster. The database cluster includes an initial master library node and at least one standby library node. The method includes the following steps: detecting state information of the initial master library node; when detecting that the initial master library node fails, promoting one standby library node to a new master library node to take over the initial master library node; during the period when the new master library node takes over the initial master library node, detecting the state information of the initial master library node, and when detecting that the initial master library node resumes operation, identifying the initial master library node as a fault recovery node; performing role isolation control on the fault recovery node; the role isolation control includes modifying role configuration information of the fault recovery node. By detecting the initial master library node when it is unavailable to when it resumes operation, performing role isolation control on the initial master library node, and prohibiting the initial master library node from operating as a master library node, data conflicts and automatic isolation caused by double master operation are avoided.