Space-time multi-dimensional perception driven cloud environment dynamic data replication optimization method and related device

Abstract

The application discloses a kind of spatiotemporal multidimensional perception driven cloud environment dynamic data replication optimization method and related device, comprising: judging whether need to carry out data file replication operation;When need to carry out data file replication operation, then the heat of each data file is calculated;The score of each storage node is calculated;The highest data file of heat is copied and stored in the highest storage node of score, the method and related device can solve the problem of network load imbalance and storage resource utilization decline caused by user scale expansion of cloud platform.

Description

Technical Field

[0001] This invention belongs to the field of distributed computing resource optimization technology, and relates to a method and related apparatus for dynamic data replication optimization in cloud environments driven by spatiotemporal multidimensional perception. Background Technology

[0002] Cloud computing has revolutionized enterprise operations through a global resource-sharing mechanism. In the cloud service ecosystem, service providers and users establish contractual relationships through Service Scope Agreements (SLAs). These agreements clearly define core elements such as service quality, uptime, availability, and pricing mechanisms, serving as a crucial framework for regulating the rights and responsibilities of both parties and ensuring the reliability, efficiency, and cost-effectiveness of service delivery. Improving cloud service quality hinges on the application of redundancy technologies, with service replication playing a central role. Data replication also reduces communication overhead by minimizing round-trip time (RTT). This data localization strategy not only accelerates data access but also improves network operating efficiency. Users can access the same data in parallel across multiple nodes, significantly enhancing network speed and reliability.

[0003] Cloud computing heavily relies on data replication technology to improve service quality and ensure data availability, fault tolerance, and operational efficiency. This technology achieves redundant storage by creating data replicas among distributed storage nodes, but its implementation faces several challenges: a) Consistency Maintenance Mechanism: Maintaining data consistency among replicas requires effective propagation of update operations. Centralized strategies restrict modification to the main logical file, while distributed strategies allow updates from any physical replica. There is a trade-off between immediate update and delayed update strategies regarding access response latency and temporary data inconsistencies.

[0004] b) Replication Strategy Classification System: Replication strategies can be divided into two categories: static and dynamic. Static strategies pre-set replication parameters, while dynamic strategies adjust dynamically based on network conditions and user access patterns. Although static strategies are simple to implement, dynamic strategies are more suitable for complex network scenarios such as cloud environments.

[0005] The above issues have led to problems such as uneven network load and decreased storage resource utilization on cloud platforms due to the expansion of user scale. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method and related apparatus for dynamic data replication optimization in cloud environments driven by spatiotemporal multidimensional perception. This method and related apparatus can solve the problems of uneven network load and decreased storage resource utilization caused by the expansion of user scale in cloud platforms.

[0007] To achieve the above objectives, this invention discloses a spatiotemporal multidimensional perception-driven dynamic data replication optimization method for cloud environments, comprising: Determine whether a data file copy operation is necessary; When a data file copying operation is required, the popularity of each data file is calculated; the score of each storage node is calculated. The most popular data file is copied and stored to the storage node with the highest score.

[0008] Furthermore, the process of determining whether a data file copying operation is necessary is as follows: When entering the replication cycle, a data file copying operation is required; When the availability of a data file drops below the user demand threshold agreed upon in the SLA, a data file copy operation is required. If the service quality indicators fail to meet user requirements, a data file copying operation is required.

[0009] Furthermore, the first n Data files within a period j The popularity is: (1)

[0010] in, For local areas l In the n Data files within the period j The total number of requests is used to quantify file access demand within a specific time and space range; For the region l In the n The total number of requests to all data files within a period reflects the activity level of data access at the regional level; For local areas l Chinese data file j The capacity size; For the region l Average size of internal data files; For the file j In the region l The number of existing copies; For the region l The average number of replicas; These are the weighting coefficients.

[0011] Furthermore, the storage node i The rating is: (10) in, Indicates storage node i Data transmission rate; Indicates the storage node in the previous time interval iTotal number of visits; Indicates storage node i The probability of failure; Indicates storage node i The current workload; These are binary control parameters.

[0012] Furthermore, the binary control parameters for: (11) (12) (13).

[0013] Furthermore, before copying and storing the most popular data file to the storage node with the highest rating, the process also includes: Determine the minimum number of replicas; Determine whether the number of replicas deployed in the storage node is greater than or equal to the minimum number of replicas; If the number of replicas deployed in the storage node is greater than or equal to the minimum number of replicas, then one of the replicas in the storage node is deleted.

[0014] Furthermore, minimal copy for: (9) (8) (7) (6) in, For file size, For the transmission rate of storage nodes, The request rate of the storage node. This is the maximum response time for the SLA. The minimum number of replicas required for availability. This represents the average failure probability at the network level.

[0015] This invention discloses a spatiotemporal multidimensional perception-driven dynamic data replication optimization system for cloud environments, comprising: The decision module is used to determine whether a data file copying operation is required; The calculation module is used to calculate the popularity of each data file and the score of each storage node when a data file copying operation is required. The copy module is used to copy the most popular data files to the storage node with the highest score.

[0016] This invention discloses a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the spatiotemporal multidimensional perception-driven dynamic data replication optimization method for cloud environments.

[0017] This invention discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the spatiotemporal multidimensional perception-driven dynamic data replication optimization method for cloud environments.

[0018] The present invention has the following beneficial effects: The spatiotemporal multidimensional perception-driven dynamic data replication optimization method and related device for cloud environments described in this invention calculates the popularity of each data file and the score of each storage node during specific operation. The data file with the highest popularity is copied and stored in the storage node with the highest score, so as to solve the problem of uneven network load and decreased storage resource utilization caused by the expansion of user scale in cloud platforms. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a model architecture diagram of the present invention; Figure 2 This is a flowchart of the method of the present invention; Figure 3 A flowchart of the replication step algorithm; Figure 4 A flowchart for the copy placement algorithm; Figure 5 A distribution chart of removal rate and average number of file copies; Figure 6 A graph showing the distribution of time intervals and storage space; Figure 7 A graph showing the distribution of CloudLet quantity and load balancing rate; Figure 8 This is a graph showing the distribution of CloudLet quantity and response time. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] In the description of this invention, it should be understood that the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0023] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0024] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. Additionally, the character " / " in this invention generally indicates that the preceding and following objects have an "or" relationship.

[0025] It should be understood that although terms such as first, second, third, etc., may be used in the embodiments of the present invention to describe the preset range, these preset ranges should not be limited to these terms. These terms are only used to distinguish the preset ranges from one another. For example, without departing from the scope of the embodiments of the present invention, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.

[0026] Depending on the context, the word "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0028] The accompanying drawings illustrate various structural schematic diagrams according to embodiments disclosed in this invention. These drawings are not to scale, and some details have been enlarged for clarity, and some details may have been omitted. The shapes of the various regions and layers shown in the drawings, as well as their relative sizes and positional relationships, are merely exemplary and may deviate from reality due to manufacturing tolerances or technical limitations. Furthermore, those skilled in the art can design regions / layers with different shapes, sizes, and relative positions as needed.

[0029] Example 1 To address the inherent defects of traditional Cayley trees, this invention makes key improvements to the Cayley tree architecture: removing the original central node and interconnecting the second-layer nodes to form a ring structure. The improved network architecture has the following advantages when used as the underlying topology of a cloud network. 1) Eliminate the risk of single point of failure; In the traditional Cayley tree architecture, the root node serves as the core intersection of all subtrees. If it fails due to natural disasters, human error, or network attacks, it will cause network partitioning, and communication between subtrees will be completely interrupted. The improved architecture ensures that the failure of any second-level node will not affect the overall network connectivity through a distributed second-level node interconnection mechanism.

[0030] 2) Alleviate network bottleneck effects; The improved architecture distributes centralized computing tasks and request-response functions to a second-layer node cluster, effectively avoiding the problem of the root node becoming a performance bottleneck in the traditional architecture through a load balancing mechanism.

[0031] 3) Simplify network management complexity; Each second-level node only needs to manage a significantly smaller subtree domain. Compared to the complex management task of the root node coordinating all nodes in the network in the original architecture, the improved solution greatly reduces the difficulty of network operation and maintenance.

[0032] like Figure 1 As shown, this invention constructs a hierarchical data replica management model based on an improved Cayley tree structure. This architecture divides the cloud network into several regions, each with a Region Replica Manager (RRM) for coordination and control. Within a region, the Local Replica Manager (LLM) oversees multiple data centers containing racks and storage devices. When a user initiates a file access request, the LLM first searches the local storage node. If the target file does not exist on the local storage node, the LLM forwards the request to its RRM. The RRM then searches the entire region. If the target file is still not located, the RRM initiates a global-local collaborative search with RRMs in other regions, thereby ensuring reliable access to data files.

[0033] It should be noted that cross-regional data access inevitably increases the round-trip time (RTT) of requests, leading to longer response times and degraded cloud service performance. Therefore, restricting high-frequency access requests to local processing is crucial, especially for frequently accessed data files that are of interest to most network users. This invention provides an intelligent replication strategy based on dynamic access pattern analysis. Its core objective is to replicate and deploy high-frequency data files to the optimal-scoring storage node, effectively improving overall network performance by balancing the load distribution among storage nodes. Simultaneously, the model introduces a low-frequency data replica eviction mechanism, freeing up storage space by removing outdated replicas to ensure resource availability for newly added high-frequency data replicas.

[0034] This model employs an update strategy that separates logical data files from physical data copies. Specifically, when the replication process begins, the logical data file is modified, and the changes are subsequently propagated to the physical copy. This mechanism defines time-constrained data freshness, meaning that users can accept version differences in the physical copy data within a specific time interval. When requesting access to a modified data file, users can choose to obtain the latest version or accept a historical version within a limited timeframe, depending on their needs.

[0035] Based on the above, refer to Figure 2 The spatiotemporal multidimensional perception-driven dynamic data replication optimization method for cloud environments described in this invention includes the following steps: The workflow of the dynamic data replication model is as follows: 1) real-time access mode monitoring and heat analysis; 2) storage node scoring based on multi-dimensional evaluation indicators; 3) adaptive replica distribution optimization; 4) dynamic replica eviction and space reclamation. This process ensures that the system continuously adapts to the dynamically changing network environment through a closed-loop feedback mechanism.

[0036] a) Decision on when to trigger replication This invention is based on a static time-sharded network activity scheduling model. At the beginning of a cycle, it performs replication of frequently accessed data files and node deployment. The setting of the interval length is a key decision variable that directly affects network performance: too long an interval will cause access requests to concentrate on specific nodes, leading to load imbalance and increased response latency; too short an interval will increase the computational overhead of the replica manager.

[0037] In addition to the regular replication cycle, there are two exceptions that require triggering an emergency replication operation: a1) Availability threshold triggering mechanism: When the availability of a data file drops below the user demand threshold agreed upon in the SLA; a2) Service quality degradation triggering mechanism: When service quality indicators such as response latency fail to meet user requirements.

[0038] b) Data file copying selection strategy b1) File popularity calculation model As shown in equation (1), the present invention establishes the first n Data files within a period j The heat change calculation model is as follows: (1)

[0039] in, For local areas l In the n Data files within the period j The total number of requests is used to quantify file access demand within a specific time and space range; For the region l In the n The total number of requests to all data files within a period reflects the activity level of data access at the regional level; For local areas l Chinese data file j Size (KB); For the region l Average size of internal data files (KB); For the file j In the region l The number of existing copies; For the region l The average number of replicas; This is the weighting coefficient (value range 0-1).

[0040] b2) Access mode weighting mechanism Local access: Optimal response time, but geographically limited; Remote access: Divided into same-region remote access (weighted by 0.5) and cross-region access (weighted by 1). Formula (1) is introduced The ratio is used to prevent excessive copying. This design uses a negative feedback mechanism to suppress the unlimited growth of copies of popular files. The global heat aggregation calculation is shown in Equation (2): (2) b3) Dynamic copy management rules when ΔP_j>+1 When creating ΔP_j Create a new copy and reset the decimal part of the original file's popularity to zero; when ΔP_j<-1 And the current number of replicas >R_cmin If the condition is met, one copy will be randomly removed, and the integer part of the popularity score will be reset.

[0041] b4) Reliability Assurance Model The annual failure rate of a single node is: (3) in, AFR For annual failure rate MTTF Mean time between failures (MTBF) is the average time between failures (MTBF).

[0042] File unavailability is defined as the probability that all replica storage nodes fail simultaneously; availability is: (4) The network-level average failure probability is used for optimization calculation, namely: (5) in, N This represents the total number of network storage nodes.

[0043] The optimized availability model is as follows: (6) b5) Performance Constraint Model Formula (7) calculates the minimum number of replicas required to guarantee availability; (7) Consider file size Node transmission rate Request rate and SLA maximum response time ,get: (8) The final minimum number of replicas is the maximum of the two values: (9) The replication algorithm is shown in Table 1: Table 1

[0044] c) Selection of storage location for the copy data file The model proposed in this invention stores the replica data file on the storage node with the highest score. i The rating is: (10) in, Indicates storage node i Data transfer rate (unit: MB / s); Indicates the storage node in the previous time interval i Total number of visits; This represents the storage node calculated based on formula (3). i Failure probability; Indicates storage node i The current workload; For binary control parameters, when When the value is 0, the storage node score is forcibly reset to zero. This parameter includes two status detection conditions: (11) The first condition is defined as follows: (12) When storage node i If the data transmission time is lower than the set threshold, then the value is set to 0. The second condition is used to detect whether the node's workload meets the threshold requirement, that is: (13) The core mechanism of improving data access performance by optimizing replica layout includes: (1) comprehensively considering the access frequency and bandwidth resources of data files in the region; (2) dynamically adjusting the replica distribution according to the request volume of each data file in the region; and (3) prioritizing the deployment of replicas in the region where the access request originates, effectively reducing round-trip time (RTT) and response time.

[0045] This model ensures that the data copy is always located closest to the user by analyzing access demand and bandwidth together, thereby achieving: (1) reduced data access latency; (2) increased system throughput; and (3) optimized user experience.

[0046] The specific copy placement algorithm is shown in Table 2: Table 2

[0047] Confirmatory test To verify the effectiveness of the model, comparative experiments were conducted with three other model algorithms: CDRM, DROPS, and PEPRv2. The simulation network consisted of 15 data centers distributed across 5 geographical locations, with each data center configured with 10 storage nodes (parameters detailed in Table 3) and 45 virtual machines (parameters detailed in Table 4). Cloud tasks (cloudlets) were received and assigned to virtual machines via a task agent; the attributes of the cloud tasks are shown in Table 5.

[0048] Table 3

[0049] Table 4

[0050] Table 5

[0051] 1. Storage space consumption Cloud storage space consumption can be evaluated by the total number of files stored in the storage nodes. In all comparative models, the number of original files remains constant in each simulation, but increasing the number of file copies directly leads to increased storage space. Therefore, controlling the number of copies can effectively manage cloud storage capacity. Figure 5 This displays how the average number of replicas created within the cloud environment changes with the removal rate parameter (i.e., α in Formula 1) and the average failure probability of storage nodes (defined in Formula 5).

[0052] from Figure 5 It can be seen how the model behavior responds to changes in the average failure probability of storage nodes when the removal rate parameter (α) is fixed. From Figure 5 As can be seen, the model creates more file replicas when the average failure probability of storage nodes increases. This dynamic adjustment mechanism ensures that the availability of data files remains at the expected level. When the average failure probability of storage nodes remains constant, increasing the α value causes the model to more strictly eliminate infrequently accessed "cold" replicas. This demonstrates that the model can achieve a precise balance between storage efficiency and data availability by dynamically adjusting the replica creation and removal rate.

[0053] like Figure 6 As shown, the storage space consumption (evaluated by the average number of replicas) was compared over different time intervals. Simulation parameters were set as follows: α = 0.2 in Equation 1, and the average failure probability of storage nodes = 0.04. The results show that the model creates a higher number of replicas than similar models. Although this implies higher storage consumption, the resulting high data availability significantly improves network reliability, demonstrating the model's advantage in building highly robust networks.

[0054] Table 6

[0055] As shown in Table 6, the average number of replicas in the model responds more dynamically to the average failure probability of storage nodes. As the failure probability increases, the model generates more replicas to ensure data availability, demonstrating its robustness under different network conditions.

[0056] 2. Load balancing The variance of workload reflects the phenomenon that some storage nodes in a network are overloaded due to a large number of user requests, while other nodes are idle because the stored data is not frequently accessed. Therefore, accurately allocating data files to storage nodes to achieve load balancing is crucial. Figure 7 The load balancing performance of different models was compared under a given total number of network tasks (number of Cloudlets).

[0057] To quantify workload balance, we calculate the standard deviation σ of the storage node workload using Equation 14: (14) in, x i For storage nodes i The workload; µ This represents the average network workload. N This represents the total number of storage nodes.

[0058] A smaller standard deviation indicates a more balanced workload distribution. For visual clarity, the standard deviation is converted to a workload balance rate. For example... Figure 8 As shown, the model's load balancing rate is significantly better than other comparative models, reducing the risk of network congestion and improving the overall system throughput. This advantage stems from the model's intelligent distribution strategy for popular data files, effectively reducing node idleness or overload.

[0059] 3. Response time Figure 8 The model's response time performance was compared with other models. The two core factors affecting response time are as follows: 1) Distance between the user and the storage node Prioritizing the storage node closest to the user to respond to requests can shorten the round-trip time (RTT) of requests, thereby reducing response time. The model significantly optimizes this metric through regionalized data replication deployment.

[0060] 2) Network load distribution Each storage node can only handle a limited number of requests simultaneously. If the load is uneven, users must wait for the nodes to complete their preceding tasks, leading to longer response times. Figure 7 As shown, the model's load balancing characteristics reduce this type of waiting time.

[0061] like Figure 8 As shown, the model's significant advantages in resource allocation efficiency and overall system performance are primarily due to its intelligent data layout strategy: by deploying replicas in areas with high user access density, the physical distance between users and storage nodes is reduced, thereby lowering round-trip latency (RTT); balanced load distribution avoids node overload and reduces task queuing time.

[0062] 5. Evaluation of Extended Performance Metrics As shown in Table 7, the model demonstrates significant advantages in packet delivery rate (PDR), throughput, time complexity, and space complexity.

[0063] 1) The PDR reached 98.7%, which is 12% higher than the best comparison model, verifying its high reliability communication capability. 2) Throughput reached 1.2 GB / s, with a 19% improvement in efficiency, attributed to optimized replica distribution reducing cross-domain transfers.

[0064] 3) Time complexity preserved O(n log n) At this level, it still possesses real-time decision-making capabilities even in large-scale node scenarios.

[0065] 4) Space complexity is controlled within O(n) Linear growth adapts to the dynamic expansion needs of cloud environments.

[0066] The results show that the model achieves efficient utilization of computing and storage resources while ensuring service quality.

[0067] Table 7

[0068] In summary, this invention proposes a model that employs a dynamic full data replication strategy in a cloud environment. Replicas are copied based on file access patterns and deployed on storage nodes with optimal configurations. Simultaneously, response latency is reduced through data locality optimization, and data modification operations are supported, ensuring replica consistency under time-constrained freshness requirements.

[0069] Experimental results show that the model performs excellently in load balancing, with an average network performance improvement of 5% compared to the best baseline model in the simulation environment. In addition, thanks to the locality-aware replica layout strategy and load balancing distribution, the model significantly reduces system response time.

[0070] Example 2 The spatiotemporal multidimensional perception-driven dynamic data replication optimization system for cloud environments described in this invention includes: The decision module is used to determine whether a data file copying operation is required; The calculation module is used to calculate the popularity of each data file and the score of each storage node when a data file copying operation is required. The copy module is used to copy the most popular data files to the storage node with the highest score.

[0071] The module division in this embodiment is illustrative and represents only one logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional modules in each embodiment of this application can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0072] Example 3 A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of a spatiotemporal multi-dimensional perception-driven dynamic data replication optimization method for cloud environments. For example, the steps include: determining whether a data file replication operation is needed; when a data file replication operation is needed, calculating the popularity of each data file; calculating the score of each storage node; and replicating and storing the data file with the highest popularity to the storage node with the highest score. The memory may include main memory, such as high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device. The processor, network interface, and memory are interconnected via an internal bus, which may be an industry-standard architecture bus, a peripheral component interconnection standard bus, an extended industry-standard architecture bus, etc. The bus may be classified as an address bus, a data bus, a control bus, etc. The memory stores the program; specifically, the program may include program code, which includes computer operation instructions. The memory may include main memory and non-volatile memory, and provides instructions and data to the processor.

[0073] Example 4 A computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of a spatiotemporal multidimensional perception-driven dynamic data replication optimization method for cloud environments. For example, the steps include: determining whether a data file replication operation is needed; when a data file replication operation is needed, calculating the popularity of each data file; calculating the score of each storage node; and replicating and storing the data file with the highest popularity to the storage node with the highest score. Specifically, the computer-readable storage medium includes, but is not limited to, volatile memory and / or non-volatile memory. The volatile memory may include random access memory (RAM) and / or cache memory, etc. The non-volatile memory may include read-only memory (ROM), hard disk, flash memory, optical disk, magnetic disk, etc.

[0074] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0075] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0076] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0077] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0078] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and disclosure of the invention. This application is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the following claims.

[0079] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

[0080] The above description is merely a preferred embodiment of the present invention and does not constitute any limitation on the present invention. Any simple modifications, alterations, or equivalent structural changes made to the above embodiments based on the technical essence of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for optimizing dynamic data replication in a cloud environment driven by spatiotemporal multidimensional perception, characterized in that, include: Determine whether a data file copy operation is necessary; When a data file copying operation is required, the popularity of each data file is calculated. Calculate the score for each storage node; The most popular data file is copied and stored to the storage node with the highest score.

2. The method for optimizing dynamic data replication in a cloud environment driven by spatiotemporal multidimensional perception according to claim 1, characterized in that, The process of determining whether a data file copying operation is necessary is as follows: When entering the replication cycle, a data file copying operation is required; When the availability of a data file drops below the user demand threshold agreed upon in the SLA, a data file copy operation is required. If the service quality indicators fail to meet user requirements, a data file copying operation is required.

3. The method for optimizing dynamic data replication in a cloud environment driven by spatiotemporal multidimensional perception according to claim 1, characterized in that, No. n Data files within a period j The popularity is: (1) in, For local areas l In the n Data files within the period j The total number of requests is used to quantify file access demand within a specific time and space range; For the region l In the n The total number of requests to all data files within a period reflects the activity level of data access at the regional level; For local areas l Chinese data file j The capacity size; For the region l Average size of internal data files; For the file j In the region l The number of existing copies; For the region l The average number of replicas; These are the weighting coefficients.

4. The method for optimizing dynamic data replication in a cloud environment driven by spatiotemporal multidimensional perception according to claim 1, characterized in that, The storage node i The rating is: (10) in, Indicates storage node i Data transmission rate; Indicates the storage node in the previous time interval i Total number of visits; Indicates storage node i The probability of failure; Indicates storage node i The current workload; These are binary control parameters.

5. The method for optimizing dynamic data replication in a cloud environment driven by spatiotemporal multidimensional perception according to claim 4, characterized in that, The binary control parameters for: (11) (12) (13)。 6. The method for optimizing dynamic data replication in a cloud environment driven by spatiotemporal multidimensional perception according to claim 1, characterized in that, Before copying and storing the most popular data file to the storage node with the highest score, the process also includes: Determine the minimum number of replicas; Determine whether the number of replicas deployed in the storage node is greater than or equal to the minimum number of replicas; If the number of replicas deployed in the storage node is greater than or equal to the minimum number of replicas, then one of the replicas in the storage node is deleted.

7. The method for optimizing dynamic data replication in a cloud environment driven by spatiotemporal multidimensional perception according to claim 6, characterized in that, Minimum copy for: (9) (8) (7) (6) in, For file size, For the transmission rate of storage nodes, The request rate of the storage node. This is the maximum response time for the SLA. The minimum number of replicas required for availability. This represents the average failure probability at the network level.

8. A dynamic data replication and optimization system for cloud environments driven by spatiotemporal multidimensional perception, characterized in that, include: The decision module is used to determine whether a data file copying operation is required; The calculation module is used to calculate the popularity of each data file when a data file copying operation is required. Calculate the score for each storage node; The copy module is used to copy the most popular data files to the storage node with the highest score.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the spatiotemporal multidimensional perception-driven dynamic data replication optimization method for cloud environments as described in any one of claims 1-7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the spatiotemporal multidimensional perception-driven dynamic data replication optimization method for cloud environments as described in any one of claims 1-7.

Images