Backup and recovery management method based on multiple data

Abstract

The invention relates to the technical field of data management, in particular to a backup and recovery management method based on multiple data, which comprises the following steps of: monitoring the updating frequency and the association dependency relationship of multiple data streams in real time, and generating an activity map comprising a time sensitivity weight and a cross-data link influence factor; dynamically dividing a time window according to the activity atlas, executing atomized snapshot backup on the associated data chain when the time window is closed, and generating a time-space consistency identifier; and reversely analyzing the associated data chain topological structure based on the time-space consistency identifier, preferentially recovering the high impact factor node and rebuilding the dependency path in parallel. According to the method, blocking and failure caused by dependency conflicts in the traditional recovery process are effectively avoided, the method is particularly suitable for a service system with complex data chain crossing, high reliability and recovery robustness are achieved, and the system recovery time and the service interruption duration are remarkably shortened.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and in particular to a backup and recovery management method based on multiple data. Background Art

[0002] With the widespread deployment of distributed systems and multi-source heterogeneous data architectures in scenarios such as cloud computing, the Internet of Things, and intelligent manufacturing, data backup and recovery have become core means to ensure system reliability and business continuity. Existing backup technologies are mostly based on fixed-cycle triggering, single-point snapshots, or simple file-level replication. They lack awareness of data flow status and structural dependencies, and are difficult to adapt to data scenarios with high-frequency updates, strong coupling, and multi-chain parallelism.

[0003] On the one hand, current systems often fail to identify differences in data activity, resulting in the same backup strategy for high-frequency critical data and low-frequency edge data, leading to wasted resources and backup delays. On the other hand, traditional backup mechanisms fail to identify dependency topologies when faced with data chain structures involving cross-service calls, nested dependencies, or event chain coupling, easily leading to state misalignment, data loss, or circular dependency blockage during recovery.

[0004] Furthermore, the recovery process often relies on linear scanning or reverse timestamp restoration, which doesn't fully consider data dependencies and priorities. It also lacks efficient concurrent scheduling and exception handling mechanisms. This is especially true in the event of system-level failures or link interruptions. Traditional methods are inefficient and risky in rebuilding complex business data chains, easily leading to prolonged service interruptions. Summary of the Invention

[0005] The present invention provides a multiple data-based backup and recovery management method with data activity recognition, dependency structure perception and efficient recovery scheduling capabilities, so as to realize a more real-time and consistent full-process data protection system.

[0006] The backup and recovery management method based on multiple data includes the following steps:

[0007] S1. Dynamic Data Activity Perception: Real-time monitoring of the update frequency and associated dependencies of multiple data streams, generating an activity map that includes time-sensitivity weights and cross-data chain impact factors;

[0008] S2. Spatiotemporal consistency backup: Dynamically divide time windows based on the active graph, perform atomic snapshot backups of the associated data chain when the time window closes, and generate spatiotemporal consistency identifiers.

[0009] S3. Topology-driven recovery: Reversely parse the topology of the linked data chain based on spatiotemporal consistency identifiers, prioritize the recovery of high-impact factor nodes, and reconstruct dependent paths in parallel.

[0010] Optionally, S1 includes quantification of the update frequency, specifically including creating a sliding time window for each data stream, counting the number of data updates in the sliding time window and superimposing a time decay factor to generate an update frequency value.

[0011] Optionally, S1 also includes parsing of associated dependencies, specifically extracting call chain identifiers embedded in the data flow, constructing a dependency graph across data flows, and calculating the betweenness centrality of each node in the dependency graph as an initial influencing factor.

[0012] Optionally, the update frequency value is normalized to generate a time sensitivity weight, and the cross-data chain impact factor is iteratively corrected based on the betweenness centrality and dependency depth in the dependency graph to finally generate an activity graph.

[0013] Optionally, the correction of the cross-data link impact factor is expressed as: Among them, β v is the corrected cross-data link impact factor, C B (v) is the betweenness centrality value of node v in the dependency graph, Denotes the dependency depth penalty factor term, d v represents the dependency depth, i.e. the shortest path length from a node to a source node, and γ represents the dependency depth penalty factor.

[0014] Optionally, the generation of the activity graph specifically includes normalizing the update frequency value of the data corresponding to each node to obtain a time sensitivity weight under a unified scale, characterizing the activity level of the data in the current time period, combining the betweenness centrality of each node in the dependency graph and its dependency depth in the topological structure, and applying a dependency depth correction factor to the centrality index to obtain the initial cross-data chain influence factor of each node, suppressing the influence weight of the deep indirect dependent node in the global topology, and when the time sensitivity weight of the node exceeds a preset threshold, the dynamic frequency coupling mechanism is activated to adjust the gain of the cross-data chain influence factor of its directly downstream dependent node, strengthen the driving effect of the highly active node on the downstream node in its dependent path, and combine the time sensitivity weight of each node with the corrected cross-data chain influence factor to form an activity graph that supports spatiotemporal consistency backup and topology-driven recovery strategy.

[0015] Optionally, the S2 specifically includes:

[0016] S21, dynamic window calculation: Determine the backup time window length of each data link based on the time sensitivity weight value in the activity graph;

[0017] S22, atomic snapshot triggering: When an update event is detected for any node in the associated data chain, the backup time window countdown for the chain is reset. When the backup time window closes, the states of all associated nodes are frozen and a consistent snapshot is generated. The associated data chain represents a set of nodes with direct or indirect dependencies constructed through a dependency graph.

[0018] S23, identifier encoding: concatenate the snapshot timestamp, data link topology hash value and check code in a preset format to generate a spatiotemporal consistency identifier, where the topology hash value comes from the cross-data link influence factor distribution in the activity graph.

[0019] Optionally, the length of each data link backup time window is calculated as: Among them, W i represents the actual backup time window length of data link i, T base is the set benchmark time window length, α i is the time sensitivity weight of data link i, is the dynamic adjustment coefficient.

[0020] Optionally, the S3 specifically includes:

[0021] S31, topology reconstruction: parsing the topological hash value in the spatiotemporal consistency identifier, restoring the node dependency and cross-data chain impact factors recorded in the activity graph;

[0022] S32, recovery priority sorting: Arrange nodes in descending order of cross-data chain impact factors to generate a main recovery queue, and simultaneously identify nodes in the topology that have no predecessor dependencies to form parallel recovery sub-queues;

[0023] S33, perform recovery in stages.

[0024] Optionally, the phased recovery includes three phases, wherein:

[0025] Phase 1: Extract high-impact factor nodes above the preset threshold from the head of the main recovery queue and synchronously start preloading all their dependencies;

[0026] Phase 2: Start multi-threaded concurrent recovery for nodes in the parallel recovery subqueue, while monitoring the dependency satisfaction status of nodes in the main queue.

[0027] Phase 3: When a circular dependency is detected, the loop is forcibly resolved based on the impact factor weight and conflict markers are recorded.

[0028] Beneficial effects of the present invention:

[0029] The present invention introduces a dynamic update frequency perception mechanism, combines call chain parsing and graph structure analysis, constructs an activity graph that reflects the data update activity and dependency relationship, and controls the length of the backup time window based on the time sensitivity weight, so that high-frequency updated data chains can obtain a shorter backup cycle, and low-frequency data can be adaptively extended, thereby improving resource utilization and the accuracy of the backup strategy. Compared with the traditional fixed time period backup method, it has stronger real-time and differentiated scheduling capabilities.

[0030] The present invention realizes a backup consistency solution with structure recognition capability through cross-node state freezing and atomic snapshot mechanism, combined with topological hashing and unified version number binding strategy. Compared with the traditional time point-centric snapshot method, the dependency structure and influencing factor distribution in the activity graph are integrated into the snapshot generation, ensuring not only time consistency during data recovery, but also reconstruction of the data chain structure according to the dependency relationship, thereby improving the data consistency and integrity protection capabilities in multiple data chains and strong coupling scenarios.

[0031] The present invention constructs a main recovery queue based on the priority of influencing factors, and is supplemented by a parallel sub-queue scheduling strategy, thereby realizing concurrent recovery of non-dependent nodes and priority reconstruction of core nodes. Through ring dependency detection and intelligent de-looping mechanism, it effectively avoids blockages and failures caused by dependency conflicts in the traditional recovery process. It is particularly suitable for business systems with complex cross-data chains, has high reliability and recovery robustness, and significantly shortens system recovery time and business interruption duration. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0033] Figure 1 Schematic diagram of a method flow in an embodiment of the present invention;

[0034] Figure 2 A schematic diagram of a spatiotemporal consistency backup embodiment of the present invention. DETAILED DESCRIPTION

[0035] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments. It is also noted that, to provide a more detailed description, the following embodiments are best and preferred embodiments, and those skilled in the art may employ alternative methods for implementing certain known technologies. Furthermore, the accompanying drawings are intended only to provide a more detailed description of the embodiments and are not intended to limit the present invention.

[0036] It should be noted that references in the specification to "one embodiment," "an embodiment," "an exemplary embodiment," "some embodiments," etc. indicate that the described embodiments may include specific features, structures, or characteristics, but not every embodiment necessarily includes such specific features, structures, or characteristics. In addition, when specific features, structures, or characteristics are described in conjunction with an embodiment, it is within the knowledge of persons skilled in the relevant art to implement such features, structures, or characteristics in conjunction with other embodiments (whether or not explicitly described).

[0037] In general, terms can be understood, at least in part, from their use in context. For example, depending at least in part on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending at least in part on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0038] like Figure 1-Figure 2 As shown, the backup and recovery management method based on multiple data includes the following steps:

[0039] S1. Dynamic Data Activity Perception: Real-time monitoring of the update frequency and associated dependencies of multiple data streams, generating an activity map that includes time-sensitivity weights and cross-data chain impact factors;

[0040] S2. Spatiotemporal consistency backup: Dynamically divide time windows based on the active graph, perform atomic snapshot backups of the associated data chain when the time window closes, and generate spatiotemporal consistency identifiers.

[0041] S3. Topology-driven recovery: Reversely parse the topology of the linked data chain based on spatiotemporal consistency identifiers, prioritize the recovery of high-impact factor nodes, and reconstruct dependent paths in parallel.

[0042] The following are some examples:

[0043] Scenario: Multi-data source backup for financial trading systems.

[0044] S1 executes:

[0045] Monitor the call chain of "transaction records (high frequency update) → risk control logs (medium frequency) → audit trails (low frequency)"

[0046] Generate activity map: time sensitivity weight = [0.9, 0.6, 0.3], impact factor = [0.8, 0.5, 0.2];

[0047] S2 execution:

[0048] Divide the difference time window (transaction record window = 5ms, other = 20ms);

[0049] When the 5ms window closes, an atomic snapshot is performed on the transaction record and the strongly associated risk control log, generating the identifier "FIN_TX_5ms_A1_B2";

[0050] S3 Execution:

[0051] The identifier is parsed to determine that the transaction record (A1) is a high-impact factor node and is restored first;

[0052] Parallel recovery of risk control logs (B2) and audit trails (via other identifiers).

[0053] S1 specifically includes:

[0054] S11, update frequency quantization: for each data stream D i Create a sliding time window of length T and count the timestamp t of each data update in the window j , and superimpose the time attenuation factor to calculate the dynamic update frequency value:

[0055] Among them, F i Indicates data flow D i Dynamic update frequency value, n i is the data flow D in the time window i The number of update events, t j is the time from the jth update event to the start of the time window, λ is the time decay factor (controls the degree of decay of historical updates) with a value range of 0.01 to 0.1, adjusted according to the length of the time window, and the default value is 0.05. T represents the total length of the sliding time window.

[0056] S12, association dependency analysis: extract the call chain identifier embedded in each data flow, build the data dependency graph G = (V, E), and calculate the betweenness centrality C for each node v∈V B (v), as the initial impact factor:

[0057] Among them, σ st is the number of shortest paths from node s to node t, σ st (v) is the number of shortest paths passing through node v, C B (v) is the betweenness centrality value of node v in the data dependency graph.

[0058] In the data dependency graph, G represents the data dependency graph, that is, the call or dependency structure between multiple data flows, V represents the set of nodes in the graph, each node represents a data flow or data unit, and E represents the set of edges in the graph, each edge represents the dependency relationship of one data flow on another data flow (direction is "dependent → dependent");

[0059] S13, activity graph synthesis: First, the dynamic update frequency value F i Normalized to time sensitivity weight:

[0060] α i is the time sensitivity weight; then combined with the betweenness centrality C of each node in the dependency graph B (v) and dependency depth d v (i.e. the shortest path length from the node to the source node), and correct its cross-data chain impact factor: β v is the cross-data chain impact factor of node v, d v is the dependency depth, γ∈(0,1) represents the dependency depth penalty factor (indicates the proportion of cross-data chain impact factors retained for each layer of dependency chain, and the recommended value is γ=0.7);

[0061] When the update frequency of a node exceeds the set threshold θ F When , the dynamic frequency coupling mechanism is triggered, and the lifting operation is performed on all its direct downstream nodes u∈succ(v): β u =β u +η·α v , where θ F is the update frequency threshold, which is set to the mean of the update frequency within the sliding window plus one standard deviation, η is the frequency coupling gain factor, which is used to control the intensity of the frequency-driven impact enhancement, and sec(v) is the set of directly downstream dependent nodes of node v;

[0062] Finally, all nodes v∈V (α v ,β v ) forms an active graph to drive subsequent backup and recovery strategies, (α v ,β v ) represents an ordered pair of two core attributes of node v in the activity graph, α v is the time sensitivity weight, reflecting the update activity of the data stream corresponding to node v, and is the normalized dynamic update frequency value. The influencing factor β v Reflecting the structural importance of node v in cross-data chain dependency, it is based on the result after correction of betweenness centrality and dependency depth, (α v ,β v) is defined as the activity feature vector of each node v in the activity graph, which is used to jointly evaluate its backup priority and recovery strategy weight.

[0063] S2 specifically includes:

[0064] S21, dynamic window calculation: Based on the time sensitivity weight of the data links in the activity graph, the backup window length of each data link is dynamically determined: Among them, W i represents the actual backup time window length of data link i, T base The benchmark time window duration set by the system is 5 seconds. i is the time sensitivity weight of data link i, obtained by normalizing the update frequency, This is a dynamic adjustment coefficient that reflects the sensitivity of activity adjustment. It supports adaptive adjustment based on system load. The value range is 1.0-3.0 and can be dynamically increased when the load is high. The default value is 2.0.

[0065] S22, atomic snapshot trigger: In each time window, if an update event is detected at any node in the associated data chain, the window countdown is immediately reset. When the window is closed, the following operations are performed:

[0066] Adopting the two-phase locking protocol (2PL), a shared lock is added to the associated nodes when freezing the data state;

[0067] A consistent snapshot is performed on the data status of all associated nodes, and the same version number is bound.

[0068] The "associated data chain" here represents a set of nodes with direct or indirect dependencies built through a dependency graph, that is, in the cross-data flow dependency graph G = (V, E) built based on the call chain identifier, starting from a certain node, the set of all downstream or upstream nodes that can be traversed through the edge E. In short, the associated data chain is the complete subgraph or subchain structure that forms the dependency path in the active graph. When the snapshot is triggered, all nodes on the chain are regarded as "associated nodes" and need to be frozen uniformly and participate in atomic backup.

[0069] S23, identifier encoding:

[0070] Generate a globally unique spatiotemporal consistency identifier for each snapshot, with the following encoding structure:

[0071] SID=[t start ]-[t end ]-[H topo ]-[CRC]; the meaning of each field is as follows:

[0072] t startThe snapshot window start timestamp, t end The closing timestamp for the snapshot window, H topo Represents the topological hash value, which is the result of the hash function calculation of the influence factor sequence of each node in the current data link. CRC is a 32-bit cyclic redundancy check code (CRC32) used for tamper-proof verification.

[0073] The method for generating the topological hash value is expressed as: in, is the cross-data chain influence factor of the kth node in the data chain, n represents the number of nodes contained in the data chain, and SHA256(·) represents the standard secure hash algorithm used to generate irreversible topological fingerprints.

[0074] S3 specifically includes:

[0075] S31, Topology Reconstruction: Parsing the Topological Hash Value H in the Spatiotemporal Consistency Identifier topo , find the matching items in the preset topology fingerprint library, and restore the dependency graph structure G=(V,E) in the original active graph based on it, and add the cross-data chain impact factor corresponding to each node at the time of recovery Indicates the real-time impact factor of node v during recovery, which can be recalculated based on the current system status or obtained by looking up a table.

[0076] S32, recovery priority sorting: Based on all nodes v∈V in the graph structure G, sort them in descending order of impact factors to generate the main recovery queue Q main :

[0077] Sort indicates a sorting operation, arranging nodes in descending order of impact factors, and descending indicates descending order;

[0078] At the same time, all nodes with zero in-degree are identified to form a parallel recovery subqueue:

[0079] Q parallel ={v∈V|in-degree(v)=0}, in-degree(v) represents the in-degree of node v, that is, the number of predecessor nodes it depends on.

[0080] S33, performs recovery in stages, scheduling recovery operations in three stages:

[0081] Phase i: Primary node recovery and preloading: From Q main Select all nodes in the queue that meet the following conditions: Trigger its state recovery and synchronously start the preloading process of its direct dependencies to accelerate subsequent unlocking operations. Here, θ is the high impact factor threshold, which is between 0.7 and 0.9 and is set according to the criticality of the system;

[0082] Phase II: Parallel node recovery: parallel All nodes in the queue start multi-threaded concurrent recovery tasks and continuously monitor the recovery completion status of the main queue nodes in their dependency paths to dynamically unlock downstream dependencies.

[0083] Phase III: Circular Dependency Processing: If a loop is detected in the dependency graph Execute the following loop removal strategy:

[0084] Calculate each edge e in the ring ij ∈C corresponding edge influence factor Take the edge corresponding to the minimum value as the cutting target: Remove the edge from the graph and generate a conflict marker:

[0085] ConflictTag=[CycleID]_Hash(e * )_Timestamp, where Represents edge e ij The impact factor of the inherited source node, Hash(e * ) represents the unique hash fingerprint of the severed edge, Timestamp represents the current timestamp, and ConflictTag is the conflict tag record, which is used for log auditing and manual intervention.

[0086] The present invention encompasses any alternatives, modifications, equivalents, and solutions that fall within the spirit and scope of the present invention. To provide a thorough understanding of the present invention, specific details are described in detail below in connection with the preferred embodiments of the present invention, but those skilled in the art will be able to fully understand the present invention without these detailed descriptions. Furthermore, to avoid unnecessary confusion regarding the essence of the present invention, well-known methods, processes, procedures, components, and circuits have not been described in detail.

[0087] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A backup and recovery management method based on multiple data, characterized in that: The following steps are involved: S1. Dynamic Data Activity Perception: Real-time monitoring of the update frequency and associated dependencies of multiple data streams, generating an activity map that includes time-sensitivity weights and cross-data chain impact factors; S2. Spatiotemporal consistency backup: Dynamically divide time windows based on the active graph, perform atomic snapshot backups of the associated data chain when the time window closes, and generate spatiotemporal consistency identifiers. S3. Topology-driven recovery: Reversely parse the topology of the linked data chain based on spatiotemporal consistency identifiers, prioritize the recovery of high-impact factor nodes, and reconstruct dependent paths in parallel.

2. The backup and recovery management method based on multiple data according to claim 1, characterized in that: The S1 includes quantification of the update frequency, specifically including creating a sliding time window for each data stream, counting the number of data updates in the sliding time window and superimposing a time decay factor to generate an update frequency value.

3. The method for managing backup and recovery of multiple data according to claim 2, wherein: The S1 also includes parsing of associated dependencies, specifically extracting call chain identifiers embedded in the data flow, building a dependency graph across data flows, and calculating the betweenness centrality of each node in the dependency graph as an initial impact factor.

4. The method for managing backup and recovery of multiple data according to claim 3, wherein: The update frequency value is normalized to generate a time sensitivity weight, and the cross-data chain impact factor is iteratively corrected based on the betweenness centrality and dependency depth in the dependency graph to finally generate an activity graph.

5. The method for managing backup and recovery of multiple data according to claim 4, wherein: The correction of the cross-data link impact factor is expressed as: Among them, β v is the corrected cross-data link impact factor, C B (v) is the betweenness centrality value of node v in the dependency graph, Denotes the dependency depth penalty factor term, d v represents the dependency depth, i.e. the shortest path length from a node to a source node, and γ represents the dependency depth penalty factor.

6. The method for managing backup and recovery of multiple data according to claim 5, wherein: The generation of the activity graph specifically includes normalizing the update frequency value of the data corresponding to each node to obtain a time sensitivity weight under a unified scale, characterizing the activity level of the data in the current time period, combining the betweenness centrality of each node in the dependency graph with its dependency depth in the topological structure, and applying a dependency depth correction factor to the centrality index to obtain the initial cross-data chain influence factor of each node, suppressing the influence weight of the deep indirect dependent node in the global topology, and when the time sensitivity weight of the node exceeds the preset threshold, the dynamic frequency coupling mechanism is activated to adjust the gain of the cross-data chain influence factor of its directly downstream dependent node, strengthen the driving effect of the highly active node on the downstream node in its dependent path, and combine the time sensitivity weight of each node with the corrected cross-data chain influence factor to form an activity graph that supports spatiotemporal consistency backup and topology-driven recovery strategy.

7. The backup and recovery management method based on multiple data according to claim 2, characterized in that: The S2 specifically includes: S21, dynamic window calculation: Determine the backup time window length of each data link based on the time sensitivity weight value in the activity graph; S22, atomic snapshot triggering: When an update event is detected for any node in the associated data chain, the backup time window countdown for the chain is reset. When the backup time window closes, the states of all associated nodes are frozen and a consistent snapshot is generated. The associated data chain represents a set of nodes with direct or indirect dependencies constructed through a dependency graph. S23, identifier encoding: concatenate the snapshot timestamp, data link topology hash value and check code in a preset format to generate a spatiotemporal consistency identifier, where the topology hash value comes from the cross-data link influence factor distribution in the activity graph.

8. The backup and recovery management method based on multiple data according to claim 7, characterized in that: The length of each data link backup time window is calculated as: Among them, W i represents the actual backup time window length of data link i, T base is the set benchmark time window length, α i is the time sensitivity weight of data link i, is the dynamic adjustment coefficient.

9. The method for managing backup and recovery of multiple data according to claim 1, wherein: The S3 specifically includes: S31, topology reconstruction: parsing the topological hash value in the spatiotemporal consistency identifier, restoring the node dependency and cross-data chain impact factors recorded in the activity graph; S32, recovery priority sorting: Arrange nodes in descending order of cross-data chain impact factors to generate a main recovery queue, and simultaneously identify nodes in the topology that have no predecessor dependencies to form parallel recovery sub-queues; S33, perform recovery in stages.

10. The backup and recovery management method based on multiple data according to claim 9, characterized in that: The phased recovery includes three phases: Phase 1: Extract high-impact factor nodes above the preset threshold from the head of the main recovery queue and synchronously start preloading all their dependencies; Phase 2: Start multi-threaded concurrent recovery for nodes in the parallel recovery subqueue, while monitoring the dependency satisfaction status of nodes in the main queue. Phase 3: When a circular dependency is detected, the loop is forcibly resolved based on the impact factor weight and conflict markers are recorded.

Images