A remote replication link detection method and system

Abstract

The present invention provides a remote replication link detection method and system thereof, relating to the field of computer technology, comprising: obtaining first information; performing path planning based on the first information, calculating the current number of communication requests of each backup storage node in turn, and allocating heartbeat detection paths by analyzing the communication load status, thereby obtaining a heartbeat detection path table between the primary storage node and the backup storage node; according to the heartbeat detection path table, triggering heartbeat requests based on the path table records and receiving feedback, updating the communication status field in the path table, and generating an updated heartbeat detection path table; constructing a link status view based on the updated heartbeat detection path table; and according to the link status view, analyzing the path communication status and judging the link health status based on preset rules, thereby generating a link health status assessment result. The present invention reduces frequent communication requests, improves query efficiency, and reduces the response time of the system under high load conditions.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a remote replication link detection method and system thereof. Background Art

[0002] With the continuous improvement of the level of informatization, data backup and disaster recovery technologies are playing an increasingly important role in ensuring data security. As a key disaster recovery method, remote replication technology realizes data backup and recovery by regularly synchronizing data in the primary storage system to the backup storage system. In a distributed storage system, the replication link between the primary storage system and the backup storage system is the channel for data transmission, and the health status of the replication link directly affects the success of each backup task. However, in the existing technology, the detection of the replication link between the primary storage system and the backup storage system mostly adopts the traditional heartbeat detection mechanism. This method often has a huge performance bottleneck. Especially when the number of nodes is large, the number of heartbeat detection requests increases exponentially, which brings a huge burden to the system. The application effect in large-scale systems is poor, and it cannot effectively solve the performance problem of replication link detection.

[0003] Based on the above shortcomings of the prior art, there is an urgent need for a remote replication link detection method and system. Summary of the Invention

[0004] The purpose of the present invention is to provide a remote replication link detection method and system to improve the above-mentioned problem. To achieve the above-mentioned purpose, the technical solution adopted by the present invention is as follows:

[0005] In a first aspect, the present application provides a remote replication link detection method, comprising:

[0006] Acquire first information, where the first information includes unique identification information of all task nodes in the primary storage system and the backup storage system and their current communication load status;

[0007] Performing path planning based on the first information, calculating the current number of communication requests of each backup storage node in turn, and allocating heartbeat detection paths by analyzing the communication load status, to obtain a heartbeat detection path table between the primary storage node and the backup storage node;

[0008] According to the heartbeat detection path table, triggering a heartbeat request based on the path table record and receiving feedback, updating the communication status field in the path table, and generating an updated heartbeat detection path table;

[0009] Extracting the communication status and timestamp information of each path according to the updated heartbeat detection path table to construct a link state view;

[0010] According to the link status view, a link health status evaluation result is generated by analyzing the path communication status and judging the link health status based on preset rules.

[0011] In a second aspect, the present application also provides a remote replication link detection system, comprising:

[0012] An acquisition module, configured to acquire first information, wherein the first information includes unique identification information of all task nodes in the primary storage system and the backup storage system and their current communication load status;

[0013] A planning module performs path planning based on the first information, calculates the current number of communication requests of each backup storage node in turn, and allocates a heartbeat detection path by analyzing the communication load status, thereby obtaining a heartbeat detection path table between the primary storage node and the backup storage node;

[0014] An updating module is configured to trigger a heartbeat request based on the path table record and receive feedback, update the communication status field in the path table, and generate an updated heartbeat detection path table according to the heartbeat detection path table;

[0015] A construction module is used to extract the communication status and timestamp information of each path according to the updated heartbeat detection path table, and construct a link state view;

[0016] The evaluation module is configured to generate a link health status evaluation result by analyzing the path communication status and judging the link health status based on preset rules according to the link status view.

[0017] The beneficial effects of the present invention are:

[0018] The present invention performs link status queries based on the route status view cached locally by the nodes. Compared with the traditional method that requires real-time interaction, it reduces frequent communication requests, improves query efficiency, and reduces the response time of the system under high load conditions. The present invention avoids the redundancy and unbalanced load problems caused by all nodes sending heartbeat requests to the remote end by planning a specific heartbeat detection path for each node and ensuring a balanced number of paths for each node. This not only improves the efficiency of link detection, but also ensures the load balance of data synchronization tasks and optimizes the utilization of overall network resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0020] Figure 1 A flowchart of a link detection method for remote replication according to an embodiment of the present invention is shown;

[0021] Figure 2 A schematic diagram of the structure of a link detection system for remote replication according to an embodiment of the present invention;

[0022] Figure 3 A diagram of the virtual components of the storage system.

[0023] Markings in the figure: 901, acquisition module; 902, planning module; 903, update module; 904, construction module; 905, evaluation module. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0025] It should be noted that similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are used only to distinguish the description and should not be understood as indicating or implying relative importance.

[0026] Example 1:

[0027] This embodiment provides a remote replication link detection method.

[0028] See also Figure 1 , the figure shows that the method includes steps S100 to S500.

[0029] Step S100: Acquire first information, where the first information includes unique identification information of all task nodes in the primary storage system and the backup storage system and their current communication load status;

[0030] It is understandable that in actual applications, the primary storage system and the backup storage system usually adopt a distributed topology, and each node is responsible for handling different data synchronization tasks. Therefore, accurately obtaining the identification information and real-time communication load status of each node is crucial for optimizing system performance and achieving load balancing. To achieve this goal, this step requires real-time monitoring of each node in the system to collect their current load status, such as key performance indicators such as network latency, bandwidth usage, and request processing capacity. Load status data not only reflects the real-time health status of the node, but also helps the path control component to reasonably plan the heartbeat path in subsequent steps to avoid performance bottlenecks caused by overloading of a single node.

[0031] Step S200: performing path planning based on the first information, calculating the current number of communication requests for each backup storage node in turn, and allocating heartbeat detection paths by analyzing the communication load status, thereby obtaining a heartbeat detection path table between the primary storage node and the backup storage node;

[0032] Furthermore, step S200 includes steps S210 to S240.

[0033] Step S210: Collect and process the load information of the standby storage node according to the first information, and obtain a load status table of each standby storage node by counting the current number of communication requests of each standby storage node;

[0034] Step S220: Initialize the heartbeat route reference count according to the load status table, and obtain an initialized heartbeat route reference count table by setting the initial reference count value of each backup storage node to 0;

[0035] Step S230: Allocate a heartbeat route to the first node of the primary storage system according to the heartbeat route reference count table. By traversing all nodes of the backup storage system, select the two nodes with the smallest reference counts as heartbeat targets, and update the reference counts to obtain the heartbeat route allocation status of the first node of the primary storage system.

[0036] Step S240: Allocate heartbeat routes to the remaining nodes in the primary storage system based on the heartbeat route allocation of the first node in the primary storage system. Loop through each primary storage node and select the two backup storage nodes with the smallest current reference counts for heartbeat route allocation. Update the reference counts to obtain a heartbeat detection path table between the primary storage system nodes and the backup storage nodes.

[0037] During this process, the path control component plays a key role in the primary storage system. Located on the same node as the link arbitration component, it is responsible for selecting a suitable backup storage system node for each primary storage system node to allocate a heartbeat detection path. Specifically, the path control component dynamically adjusts path allocation based on the communication load status of each backup storage node and the current number of communication requests. In the system, the load status of each backup storage node changes in real time, so the path control component needs to rely on this real-time data for calculations to ensure load balancing and allocate at least two suitable backup storage nodes for heartbeat detection to each task node of the primary storage system based on the load conditions. This allocation not only prevents a single node from being overloaded with requests, but also ensures a balanced load on each path, thereby optimizing the performance of the entire system.

[0038] Step S300: According to the heartbeat detection path table, triggering a heartbeat request based on the path table record and receiving feedback, updating the communication status field in the path table, and generating an updated heartbeat detection path table;

[0039] Furthermore, step S300 includes steps S310 to S340.

[0040] Step S310: triggering a heartbeat request based on the heartbeat detection path table, sending a heartbeat signal to each backup storage node recorded in the path table to obtain a heartbeat request sending record;

[0041] Step S320: Send a record according to the heartbeat request, perform heartbeat feedback reception processing, and obtain a heartbeat response record by collecting heartbeat response data from each backup storage node;

[0042] Step S330: Analyze and process the communication status based on the heartbeat response record, update the communication status of each heartbeat route by evaluating the response time and success rate, and obtain the latest communication status of each route;

[0043] Step S340: update the heartbeat detection path table according to the latest communication status, and generate an updated heartbeat detection path table by integrating all updated communication status information.

[0044] It should be noted that during this process, the heartbeat sending component and the heartbeat receiving component are deployed in the primary storage system and the backup storage system, respectively. The heartbeat sending component, located in each task node of the primary storage system, is responsible for triggering and sending heartbeat requests to the remote nodes of the backup storage system. The heartbeat receiving component, located in each task node of the backup storage system, receives heartbeat requests from the primary storage node and returns a response. Specifically, the heartbeat sending component sends heartbeat requests to the designated remote backup storage node according to the heartbeat detection path table. Each request contains information such as a timestamp and communication status. The feedback after transmission is received by the heartbeat receiving component and transmitted back to the primary storage system. Based on the feedback, the primary storage system updates the communication status field in the path table through the heartbeat sending component, marking the latest communication status of each path, such as "online" or "offline." This step, by triggering and receiving heartbeat requests in real time, ensures real-time feedback and updates on the status of each path in the path table. This dynamic update mechanism can promptly reflect any network failures or load changes in the system, ensuring accurate detection of link health status. By using a one-way heartbeat and feedback mechanism, the overall system performance and responsiveness are optimized, while unnecessary communication requests are significantly reduced, reducing the system burden. The updated path table provides accurate, real-time basic data for subsequent link status assessment and health monitoring, ensuring the reliability and efficiency of data synchronization tasks.

[0045] Step S400: extract the communication status and timestamp information of each path according to the updated heartbeat detection path table, and construct a link status view;

[0046] Furthermore, step S400 includes steps S410 to S430.

[0047] Step S410: Extract the communication status and timestamp based on the updated heartbeat detection path table. By analyzing the communication success or failure, response time, and last communication timestamp of each entry in the path table, a data list containing the local node ID, remote node ID, route status, and last communication timestamp is obtained.

[0048] Step S420: constructing a data structure according to the data list by integrating the local node ID, remote node ID, route status, and last update time of each path into a 4-tuple format to obtain a 4-tuple list;

[0049] Step S430 : Generate a link state view according to the 4-tuple list, and generate a link state view by integrating all 4-tuples into a unified view framework.

[0050] It can be understood that in this step, the heartbeat view component is responsible for extracting the communication status (such as "online" or "offline") and timestamp (recording the last update time) of each path from the path table, and integrating this data into a structured data format, such as a 4-tuple form <local node ID, remote node ID, route status, last update time>. The updated heartbeat detection path table contains the communication status and timestamp data of each path. The heartbeat view component extracts the information of each path based on this data and generates a 4-tuple according to the status of each path. Each 4-tuple represents the complete status of a path, including the identifier of the local node, the identifier of the remote node, the status of the current path, and the update time of the last heartbeat request. In this way, the heartbeat view component not only establishes a detailed status view for each path, but also enables the link health of the entire system to be represented in a structured manner, which is convenient for subsequent analysis and decision-making.

[0051] Step S500: According to the link status view, a link health status evaluation result is generated by analyzing the path communication status and judging the link health status based on preset rules.

[0052] Furthermore, step S500 includes steps S510 to S530.

[0053] Step S510: Analyze the path communication status according to the link status view, and obtain the current communication status statistics of each path by counting the communication status of each path in the statistical view;

[0054] Step S520: performing health status determination processing on the communication status statistics based on the preset health status rules to obtain a preliminary health status assessment of the entire link;

[0055] Step S530: Perform an assessment based on the preliminary health status assessment, and generate a link health status assessment result by integrating the status assessment of each path and the overall health of the system.

[0056] In this step, the link arbitration component in the primary storage system collects link status views from all nodes and evaluates the communication status of each path according to pre-set rules. Each path's communication status information (such as "online" or "offline") and timestamp data serve as the basis for evaluating link health. Specifically, the link arbitration component analyzes the status of each path in the link status view and generates a link health assessment result based on pre-set health status judgment rules. Pre-set rules include:

[0057] Health status: If the status of all paths is "online", the link health status is "healthy";

[0058] Subhealthy status: If a path is "offline", but the number of offline paths does not exceed half, the link health status is "subhealthy";

[0059] Fault status: If more than half of the paths are "offline", the link health status is "faulty";

[0060] Offline Status: If all paths are "Offline", the link health status is "Offline".

[0061] By applying these rules, the link arbitration component can determine the health of the entire replication link based on the communication status of each path and generate a final assessment result. This assessment provides a basis for subsequent system optimization, fault response, and resource allocation. This step, through a rule-based health status assessment mechanism, provides a real-time and accurate assessment of the health of the entire link. This automated health status assessment can quickly identify potential link issues and respond to network failures in a timely manner, improving system reliability and fault tolerance.

[0062] Example 2:

[0063] The difference between this embodiment and the above-mentioned embodiment 1 is that step S330 further includes steps S331 to S334.

[0064] Step S331: performing data aggregation processing according to the updated heartbeat detection path table, and obtaining a grouped heartbeat response data set by evaluating and classifying the heartbeat response patterns using cluster analysis;

[0065] Specifically, cluster analysis algorithms, such as K-means or DBSCAN, are used in this step to evaluate and classify heartbeat response patterns. Cluster analysis can be used to categorize heartbeat response data into several distinct groups, each representing a specific response pattern. For example, some nodes may exhibit stable response patterns (low latency, high success rate), while others may exhibit unstable patterns (high latency, low success rate). By automatically identifying these patterns, cluster analysis helps the system better understand the health of the network.

[0066] The cluster analysis process includes the following steps:

[0067] Step S3311, data preparation: First, extract the communication status data of each path from the updated heartbeat detection path table and convert it into a data format suitable for cluster analysis. The data of each path may include indicators such as communication success rate and response time.

[0068] Step S3312: Applying a clustering model: A clustering algorithm is applied to group the data and automatically identify different heartbeat response patterns. Common clustering algorithms such as K-means divide the heartbeat response data into multiple categories based on a set number of clusters (K value), while DBSCAN automatically clusters based on the density of data points.

[0069] Step S3313, grouping result output: Finally, a grouped heartbeat response data set is obtained, each data set contains heartbeat paths with similar response patterns for subsequent analysis.

[0070] This step uses cluster analysis to deeply identify and classify heartbeat response data, enabling the system to understand the diversity of network conditions at a higher level. Through automated classification, the system can identify potential network issues and bottlenecks, enabling more precise fault location and optimization. Furthermore, this data-driven analysis approach dynamically adapts to changes in the network environment, demonstrating high scalability and application value in large-scale distributed systems.

[0071] Step S332: Perform success rate prediction processing based on the heartbeat response data set, and predict future success rates by applying a support vector machine to obtain a prediction success rate model adjusted based on historical data and real-time feedback;

[0072] During this process, the link arbitration component and the heartbeat view component work closely together to generate a success rate prediction model that can be adjusted in real time by inputting historical data extracted from the heartbeat response data set into the support vector machine model for training. In this step, the support vector machine is used to build a regression model to predict future changes in success rates by learning the success rate patterns of historical heartbeat response data. This step uses the support vector machine to predict the success rate of the heartbeat response data, effectively combining historical data with real-time feedback, thereby achieving a dynamic prediction of the future link health status. Compared with traditional static evaluation methods, the support vector machine model can capture more complex patterns and provide more accurate success rate predictions. This prediction model not only improves the accuracy of link health status assessment, but also improves the system's early warning capabilities for possible future failures, and enhances the system's adaptability and fault tolerance.

[0073] Step S333: Optimize and analyze the response time based on the prediction success rate model, and obtain an optimized response time strategy by searching for the best response time configuration to minimize delay and maximize communication efficiency.

[0074] It's understood that in this step, the link arbitration component uses the output of the predicted success rate model as input, combined with optimization algorithms such as genetic algorithms, simulated annealing, or particle swarm optimization, to optimize the system's response time. This optimization not only effectively reduces response time and improves the system's real-time responsiveness, but also reduces unnecessary delays while ensuring the stability and success rate of data synchronization, thereby optimizing overall system performance.

[0075] Step S334: Dynamic communication status update processing is performed according to the response time strategy. The success rate and response time of the communication data are analyzed in real time and compared with the preset performance threshold. The status judgment standard is adjusted based on the comparison result to obtain the latest communication status of each route.

[0076] It is understandable that the link arbitration component evaluates the status of the current network link based on the communication data collected in real time, including the success rate and response time of heartbeat requests, and compares it with the pre-set performance thresholds. These performance thresholds may include key indicators such as the upper limit of communication delay, response time and success rate, which are used to determine whether the path is in an "online", "sub-healthy" or "offline" state. Compared with the static status judgment in traditional methods, the dynamic adjustment of thresholds can more accurately reflect the health status of the current network, identify problems in the link in a timely manner and make adjustments. This method not only enhances the real-time and flexibility of the system, but also improves the fault tolerance of the system, ensuring the efficient implementation of data synchronization tasks in complex network environments.

[0077] Example 3:

[0078] like Figure 3 As shown, this embodiment specifically involves configuring task nodes and planning and updating heartbeat detection paths in the primary and backup storage systems. This embodiment details how to optimize the link detection process and improve system performance and stability by deploying arbitration components, control components, view components, and send / receive components in the primary and backup storage systems. Here, node1, node2, node3, node_n, and node_m represent node names, respectively.

[0079] Suppose the primary storage system has m task nodes; the backup storage system has n task nodes.

[0080] In the primary storage system, the arbitration component and the control component are deployed on nodes 1 and 2. Arbitration component 1 is in operation, while arbitration component 2 is in standby. Each task node is deployed with the view component and the send component. In the backup storage system, the arbitration component and the control component are deployed on nodes 1 and 2. Arbitration component 1 is in operation, while arbitration component 2 is in standby. Each task node is deployed with the view component and the receive component.

[0081] The control component in the primary storage system selects two heartbeat routes for nodes node_1, node_2, and node_3. In the primary storage system, arbitration component 1 periodically broadcasts query requests to view components 1 through m, querying the remote node status views stored in these view components. Each sending component periodically triggers heartbeat checks, sends heartbeat requests to the remote nodes along the pre-planned heartbeat routes, and feeds the results back to the local view component to construct the remote node status view.

[0082] In the backup storage system, arbitration component 1 periodically broadcasts query requests to view components 1 through m, querying the remote node status views stored in these view components. When the receiving component receives the request from the primary storage system node, it notifies the view component to build a view and update the timestamp. While broadcasting the query view, if arbitration component 1 detects that the timestamp of a route in the remote node status view has not been updated for more than 10 seconds, it marks the route as "offline."

[0083] Example 4:

[0084] like Figure 2 As shown, this embodiment provides a remote replication link detection system, the system including:

[0085] An acquisition module 901 is configured to acquire first information, where the first information includes unique identification information of all task nodes in the primary storage system and the backup storage system and their current communication load status;

[0086] Planning module 902 performs path planning based on the first information, calculates the current number of communication requests for each backup storage node in turn, and allocates heartbeat detection paths by analyzing the communication load status, thereby obtaining a heartbeat detection path table between the primary storage node and the backup storage node;

[0087] An updating module 903 is configured to trigger a heartbeat request based on the path table record and receive feedback, update the communication status field in the path table, and generate an updated heartbeat detection path table according to the heartbeat detection path table;

[0088] A construction module 904 is configured to extract the communication status and timestamp information of each path according to the updated heartbeat detection path table, and construct a link state view;

[0089] The evaluation module 905 is configured to generate a link health status evaluation result by analyzing the path communication status and judging the link health status based on preset rules according to the link status view.

[0090] In a specific embodiment of the present invention, the planning module 902 includes:

[0091] a first planning unit, configured to collect and process load information of the standby storage nodes according to the first information, and obtain a load status table of each standby storage node by counting a current number of communication requests of each standby storage node;

[0092] The second planning unit is configured to initialize the heartbeat route reference count according to the load status table, and obtain an initialized heartbeat route reference count table by setting an initial reference count value of 0 for each backup storage node;

[0093] The third planning unit is configured to allocate a heartbeat route to the first node of the primary storage system based on the heartbeat route reference count table, traverse all nodes of the backup storage system, select the two nodes with the smallest reference counts as heartbeat targets, and update the reference counts to obtain the heartbeat route allocation status of the first node of the primary storage system;

[0094] The fourth planning unit is used to allocate heartbeat routes to the remaining nodes in the main storage system based on the heartbeat route allocation of the first node of the main storage system. By looping each main storage node and selecting the two backup storage nodes with the smallest current reference counts for heartbeat route allocation, the reference count is updated to obtain a heartbeat detection path table between the main storage system node and the backup storage node.

[0095] In a specific embodiment of the present invention, the update module 903 includes:

[0096] A first updating unit is configured to trigger a heartbeat request according to the heartbeat detection path table, and obtain a heartbeat request sending record by sending a heartbeat signal to each backup storage node recorded in the path table;

[0097] A second updating unit is configured to send a record according to a heartbeat request, receive and process heartbeat feedback, and obtain a heartbeat response record by collecting heartbeat response data from each backup storage node;

[0098] The third updating unit is used to analyze and process the communication status according to the heartbeat response record, update the communication status of each heartbeat route by evaluating the response time and success rate, and obtain the latest communication status of each route;

[0099] The fourth updating unit is configured to update the heartbeat detection path table according to the latest communication status, and generate an updated heartbeat detection path table by integrating all updated communication status information.

[0100] In a specific embodiment of the present invention, the construction module 904 includes:

[0101] The first construction unit is configured to extract the communication status and timestamp based on the updated heartbeat detection path table, and obtain a data list including the local node ID, the remote node ID, the route status, and the last communication timestamp by analyzing the communication success or failure, the response time, and the last communication timestamp of each entry in the path table;

[0102] The second construction unit is used to construct a data structure according to the data list, by integrating the local node ID, remote node ID, route status and last update time of each path into a 4-tuple format to obtain a 4-tuple list;

[0103] The third construction unit is configured to generate a link state view according to the 4-tuple list, and generate the link state view by integrating all 4-tuples into a unified view framework.

[0104] In a specific embodiment of the present invention, the evaluation module 905 includes:

[0105] a first evaluation unit, configured to analyze the path communication state according to the link state view, and obtain current communication state statistics of each path by counting the communication state of each path in the statistical view;

[0106] A second evaluation unit performs health status determination processing on communication status statistics based on preset health status rules to obtain a preliminary health status evaluation of the entire link;

[0107] The third evaluation unit is configured to perform evaluation processing based on the preliminary health status evaluation, and generate a link health status evaluation result by integrating the status evaluation of each path and the overall health status of the system.

[0108] In a specific embodiment of the present invention, the third updating unit includes:

[0109] A first aggregation unit is configured to perform data aggregation processing according to the updated heartbeat detection path table, and obtain a grouped heartbeat response data set by evaluating and classifying the heartbeat response patterns using cluster analysis;

[0110] A first prediction unit is configured to perform success rate prediction processing based on the heartbeat response data set, predict future success rates by applying a support vector machine, and obtain a prediction success rate model adjusted based on historical data and real-time feedback;

[0111] A first optimization unit is configured to perform response time optimization analysis based on the prediction success rate model, and obtain an optimized response time strategy by searching for an optimal response time configuration to minimize latency and maximize communication efficiency;

[0112] The first adjustment unit is used to perform dynamic communication status update processing according to the response time strategy, analyze the success rate and response time of communication data in real time and compare them with the preset performance threshold, adjust the status judgment standard based on the comparison result, and obtain the latest communication status of each route.

[0113] The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed by the present invention, which should be covered by the scope of protection of the present invention.

Claims

1. A remote replication link detection method, characterized in that: include: Acquire first information, where the first information includes unique identification information of all task nodes in the primary storage system and the backup storage system and their current communication load status; Performing path planning based on the first information, calculating the current number of communication requests of each backup storage node in turn, and allocating heartbeat detection paths by analyzing the communication load status, to obtain a heartbeat detection path table between the primary storage node and the backup storage node; According to the heartbeat detection path table, triggering a heartbeat request based on the path table record and receiving feedback, updating the communication status field in the path table, and generating an updated heartbeat detection path table; Extracting the communication status and timestamp information of each path according to the updated heartbeat detection path table to construct a link state view; According to the link status view, by analyzing the path communication status and judging the link health status based on preset rules, a link health status evaluation result is generated; Among them, path planning processing is performed based on the first information, by calculating the current number of communication requests of each backup storage node in turn, and allocating the heartbeat detection path by analyzing the communication load state, to obtain a heartbeat detection path table between the primary storage node and the backup storage node, including: The path control component is responsible for selecting a suitable backup storage node for each primary storage node to perform heartbeat detection, thereby performing heartbeat detection path allocation and obtaining a heartbeat detection path table between the primary storage node and the backup storage node. The path control component dynamically adjusts the path allocation based on the communication load status of each backup storage node and the current number of communication requests to prevent a single node from bearing too many requests and ensure balanced load on each path. According to the heartbeat detection path table, triggering a heartbeat request based on the path table record and receiving feedback, updating the communication status field in the path table, and generating an updated heartbeat detection path table, including: According to the heartbeat detection path table, trigger processing of the heartbeat request is performed, and a heartbeat request sending record is obtained by sending a heartbeat signal to each backup storage node recorded in the path table; Sending a record according to the heartbeat request, performing reception processing of the heartbeat feedback, and obtaining a heartbeat response record by collecting heartbeat response data from each backup storage node; Analyzing and processing the communication status according to the heartbeat response record, updating the communication status of each heartbeat route by evaluating the response time and success rate, and obtaining the latest communication status of each route; The heartbeat detection path table is updated according to the latest communication status, and an updated heartbeat detection path table is generated by integrating all updated communication status information.

2. A remote replication link detection method according to claim 1, characterized in that: According to the updated heartbeat detection path table, the communication status and timestamp information of each path are extracted to construct a link state view, including: Extracting communication status and timestamps based on the updated heartbeat detection path table, and obtaining a data list including the local node ID, remote node ID, route status, and last communication timestamp by analyzing the communication success, response time, and last communication timestamp of each entry in the path table; Performing a data structure construction process according to the data list, by integrating the local node ID, the remote node ID, the route status and the last update time of each path into a 4-tuple format to obtain a 4-tuple list; A link state view is generated based on the 4-tuple list, and a link state view is generated by integrating all 4-tuples into a unified view framework.

3. The remote replication link detection method according to claim 1, wherein: According to the link status view, by analyzing the path communication status and judging the link health status based on preset rules, a link health status assessment result is generated, including: Analyzing the path communication status according to the link status view, and obtaining the current communication status statistics of each path by counting the communication status of each path in the statistical view; Performing health status determination processing on the communication status statistics based on preset health status rules to obtain a preliminary health status assessment of the entire link; An assessment process is performed based on the preliminary health status assessment, and a link health status assessment result is generated by integrating the status assessment of each path and the overall health status of the system.

4. A remote replication link detection system, characterized in that: include: An acquisition module, configured to acquire first information, wherein the first information includes unique identification information of all task nodes in the primary storage system and the backup storage system and their current communication load status; A planning module performs path planning based on the first information, calculates the current number of communication requests of each backup storage node in turn, and allocates a heartbeat detection path by analyzing the communication load status, thereby obtaining a heartbeat detection path table between the primary storage node and the backup storage node; An updating module is configured to trigger a heartbeat request based on the path table record and receive feedback, update the communication status field in the path table, and generate an updated heartbeat detection path table according to the heartbeat detection path table; A construction module is used to extract the communication status and timestamp information of each path according to the updated heartbeat detection path table, and construct a link state view; An evaluation module, configured to generate a link health status evaluation result by analyzing the path communication status and judging the link health status based on preset rules according to the link status view; Among them, path planning processing is performed based on the first information, by calculating the current number of communication requests of each backup storage node in turn, and allocating the heartbeat detection path by analyzing the communication load state, to obtain a heartbeat detection path table between the primary storage node and the backup storage node, including: The path control component is responsible for selecting a suitable backup storage node for each primary storage node to perform heartbeat detection, thereby performing heartbeat detection path allocation and obtaining a heartbeat detection path table between the primary storage node and the backup storage node. The path control component dynamically adjusts the path allocation based on the communication load status of each backup storage node and the current number of communication requests to prevent a single node from bearing too many requests and ensure balanced load on each path. The update module includes: A first updating unit is configured to trigger a heartbeat request according to the heartbeat detection path table, and obtain a heartbeat request sending record by sending a heartbeat signal to each backup storage node recorded in the path table; A second updating unit is configured to send a record according to the heartbeat request, perform reception processing of the heartbeat feedback, and obtain a heartbeat response record by collecting heartbeat response data from each backup storage node; a third updating unit, configured to analyze and process the communication status according to the heartbeat response record, update the communication status of each heartbeat route by evaluating the response time and success rate, and obtain the latest communication status of each route; The fourth updating unit is configured to update the heartbeat detection path table according to the latest communication status, and generate an updated heartbeat detection path table by integrating all updated communication status information.

5. A remote replication link detection system according to claim 4, characterized in that: The building blocks include: A first construction unit is configured to extract communication status and timestamps based on the updated heartbeat detection path table, and obtain a data list including a local node ID, a remote node ID, a route status, and a last communication timestamp by analyzing the communication success or failure, the response time, and the last communication timestamp of each entry in the path table; a second construction unit, configured to construct a data structure according to the data list, by integrating the local node ID, remote node ID, route status, and last update time of each path into a 4-tuple format to obtain a 4-tuple list; The third construction unit is configured to generate a link state view according to the 4-tuple list, and generate a link state view by integrating all 4-tuples into a unified view framework.

6. A remote replication link detection system according to claim 4, characterized in that: The evaluation module includes: a first evaluation unit, configured to analyze the path communication state according to the link state view, and obtain current communication state statistics of each path by counting the communication state of each path in the statistical view; A second evaluation unit, configured to perform health status determination processing on the communication status statistics based on a preset health status rule to obtain a preliminary health status evaluation of the entire link; The third evaluation unit is configured to perform evaluation processing based on the preliminary health status evaluation, and generate a link health status evaluation result by integrating the status evaluation of each path and the overall health status of the system.

Images