Hybrid cloud data backup management method based on multi-objective optimal copy management policy

Abstract

Provided in the present invention are a data backup management method and system based on a multi-objective optimal copy management policy. The method comprises: respectively defining a reliability objective function, a sending delay objective function and a load balancing objective function for complete backup of data; acquiring an optimization objective function for data backup management on the basis of the reliability objective function, the sending delay objective function and the load balancing objective function; by using a decomposition-based multi-objective evolutionary algorithm, acquiring an optimal placement scheme matrix that meets the optimization objective function; and on the basis of the optimal placement scheme matrix, performing backup placement distribution of the data on each node. In the present invention, by comprehensively considering the reliability of backup data, the delay of a backup system and the stability of the backup system, and by using a decomposition-based multi-objective optimization algorithm, a backup copy optimal placement scheme for a gold copy is obtained, thereby reducing the risk of data loss and damage.

Description

Hybrid cloud data backup management method based on multi-objective optimal copy management strategy Technical Field

[0001] The present invention belongs to the field of cloud computing disaster recovery, and in particular relates to a hybrid cloud data backup management method based on a multi-objective optimal copy management strategy. Background Art

[0002] With the advent of the big data era, data has become a critical resource. Cloud data backup and recovery are crucial for governments and enterprises migrating to the cloud. This involves backing up data from local data centers to the cloud, or vice versa, to facilitate recovery in the event of data loss, corruption, or an attack. This is crucial for ensuring data security, improving data accessibility, reducing costs, and simplifying data management. However, the backup and recovery of massive amounts of data makes traditional data protection methods prohibitively expensive, time-consuming, and compromised.

[0003] Copy Data Management (CDM) is an efficient cloud-based backup and data protection solution and a crucial component for protecting critical information for governments, businesses, and individuals. An effective copy data management strategy can mitigate data loss caused by failures or natural disasters, ensure business continuity, address cybersecurity threats, and improve data recovery speed. Data backup management is crucial for ensuring data security and business continuity. Implementing an effective data backup strategy can reduce the risk of data loss, safeguard the interests of governments, businesses, and customers, and significantly impact the data stability, reliability, and response speed of the entire management system.

[0004] Backup data reliability is crucial in a replica management strategy. Reliability means that the backup data should be reliable and able to be restored in the event of any problem. Furthermore, during the backup and recovery process, it's also necessary to ensure that the backup data can be restored by repairing the data files in the event of an error.

[0005] Backup system latency is a key characteristic of replica management strategies. Cloud storage systems often store multiple copies of files on different nodes to ensure the security of stored information and reduce user access latency. In cloud storage systems, if a file is accessed infrequently and has too many copies, unnecessary storage overhead and costs will increase. On the other hand, if a file is accessed frequently and has too few copies, access latency will increase, failing to meet user needs.

[0006] In the replica management strategy, the load balancing of the backup system is also an important factor to consider. The server load balancing rate not only affects the server's resource utilization, but also has a significant impact on the entire system.

[0007] However, in actual replica management strategies, the three aspects pursued by the backup system may be potentially conflicting and cannot be optimized at the same time.

[0008] Summary of the Invention

[0009] The purpose of the present invention is to provide a hybrid cloud data backup management method based on a multi-objective optimal copy management strategy to make up for the shortcomings of the above-mentioned technical solutions. In the backup and recovery management system, backup and recovery of different types of data can be achieved through a single platform. In the backup module, the reliability of the backup data, the overall system latency and load balancing issues are taken into consideration at the same time, a multi-objective optimization model is established, and the optimal copy management strategy of the Pareto equilibrium of the three problems is obtained. The backup copies are backed up to the underlying storage space of the system based on the optimal strategy. This method improves the reliability of the backup data and the stability of the backup system while simplifying user operations and improving business efficiency, reduces the latency of the backup system, and ensures the stable operation of the business system.

[0010] On the one hand, to achieve the above-mentioned object, the present invention provides a hybrid cloud data backup management method based on a multi-objective optimal copy management strategy, comprising the following steps:

[0011] The reliability objective function, transmission delay objective function, and load balancing objective function of complete data backup in several storage nodes are defined respectively;

[0012] Obtaining an optimization objective function for data backup management based on the reliability objective function, the transmission delay objective function, and the load balancing objective function;

[0013] Using a decomposition-based multi-objective evolutionary algorithm to obtain an optimal placement solution matrix that satisfies the optimization objective function;

[0014] The data is allocated for backup placement on each storage node based on the optimal placement solution matrix.

[0015] Optionally, the reliability objective function is as follows:

[0016] in, Indicates the i-th data file k i Reliability, χ j is the jth storage node t in the storage cluster T j The failure rate of u storage node, is the storage node t jThe failure rate function, K represents the data file, m is the number of storage nodes, T represents a cluster consisting of m storage nodes, tk ij Represents node t j A copy of the above.

[0017] Optionally, the transmission delay objective function is as follows:

[0018] Among them, v j is the backup node t j Data transmission rate; c i For data file k i The size of represents k i At node t j Total visits on ; represents k i The backup copy is stored on the storage node t j The sending delay of the a-th access on .

[0019] Optionally, the load balancing objective function is as follows:

[0020] Among them, σ j For each backup node t j The weight of

[0021] Optionally, the optimization objective function is as follows:

[0022] Where X is the decision variable, here is the placement plan matrix TK, f1(X)=s1(TK), ρ(k i ) represents the i-th data file k i The size of Γ(t j ) indicates the j Storage nodes t j The capacity size.

[0023] Optionally, the process of using a decomposition-based multi-objective evolutionary algorithm to obtain an optimal placement solution matrix that satisfies the optimization objective function includes:

[0024] Initialize the population size and randomly generate a weight vector equal to the population size;

[0025] Decomposing the optimization objective function into a number of sub-problems equal to the number of populations based on the Tchebycheff decomposition algorithm;

[0026] Initialize the number of evolutions and randomly generate a placement scheme matrix with the same population size;

[0027] Initialize the frontier solution set and the reference point corresponding to the optimization objective function;

[0028] Initialize the neighborhood vector, obtain several weight vectors closest to the last weight vector, and build an index set based on several weight vectors;

[0029] A maximum number of iterations is set and iterations are performed, and the last weight vector and the corresponding placement solution matrix are iteratively updated based on the number of iterations. When the maximum number of iterations is reached, the iterations are stopped to obtain the optimal placement solution matrix.

[0030] Optionally, the process of updating each weight vector and the corresponding placement solution matrix based on the number of iterations includes:

[0031] Randomly obtain an index number based on the index set, and obtain a placement solution matrix corresponding to the index number;

[0032] Obtain a new placement solution matrix based on the corresponding placement solution matrix, and iteratively update the reference point, the population, and the non-dominated solution set based on the new placement solution matrix;

[0033] When the maximum number of iterations is reached, an updated non-dominated solution set is obtained and used as an optimal frontier solution set, a trade-off solution is selected from the optimal frontier solution set, and the trade-off solution is used as an optimal placement solution matrix.

[0034] Optionally, during the updating process of the reference point, but Represents the subproblem f k The reference point, Represents the new placement scheme matrix;

[0035] During the updating process of the population, for each neighborhood individual in the index set like but where g te represents the subproblem obtained based on the Tchebycheff decomposition algorithm, and λ represents the neighborhood individual The weight vector of .

[0036] Optionally, during the update of the non-dominated solution set, the dominant solution in INL(h-1) is retained. The solution is to discard all the Dominated solution, if all solutions of INL(h-1) do not dominate Will Add INL(h), where INL(h-1) represents the non-dominated solution set after the last update. represents the new placement solution matrix, and INL(h) represents the current non-dominated solution set.

[0037] On the other hand, in order to achieve the above-mentioned purpose, the present invention provides a hybrid cloud data backup management system based on a multi-objective optimal copy management strategy, characterized in that the system includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor, and when the processor executes the computer program, it implements the above-mentioned hybrid cloud data backup management method based on the multi-objective optimal copy management strategy.

[0038] Compared with existing technologies, the present invention's hybrid cloud data backup and recovery management method, based on a multi-objective optimal replica management strategy, comprehensively considers the reliability of backup data, the latency of the backup system, and the stability of the backup system. Through a decomposition-based multi-objective optimization algorithm, it derives the optimal placement plan for backup copies of the golden copy, thereby reducing the risk of data loss and corruption. Users can configure backup and recovery strategies through this method, enabling flexible and rapid data backup and recovery, helping administrators quickly locate faults and maximize system efficiency and performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present invention. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0040] FIG1 is a flow chart of a multi-objective optimal replica management method according to an embodiment of the present invention;

[0041] FIG2 is a schematic diagram of the design principle of a hybrid cloud data backup and recovery management system based on a multi-objective optimal copy management strategy in an embodiment of the present invention;

[0042] FIG3 is an architecture diagram of a hybrid cloud data backup and recovery management system based on a multi-objective optimal copy management strategy in an embodiment of the present invention;

[0043] FIG4 is a schematic diagram of a multi-target backup copy management strategy design based on the CDM golden copy concept in an embodiment of the present invention;

[0044] FIG5 is a roadmap of a multi-objective optimal replica management method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0045] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, unless there is a conflict, the embodiments of the present disclosure and the features described in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0046] As shown in Figures 1-5, this embodiment provides a hybrid cloud data backup management method and system based on a multi-objective optimal copy management strategy. The method, as shown in Figure 1, includes:

[0047] S100: defining a reliability objective function, a transmission delay objective function, and a load balancing objective function for complete data backup in a plurality of storage nodes;

[0048] S200: Acquire an optimization objective function for data backup management based on the reliability objective function, the transmission delay objective function, and the load balancing objective function;

[0049] S300: using a decomposition-based multi-objective evolutionary algorithm to obtain an optimal placement solution matrix that satisfies the optimization objective function;

[0050] S400: Allocate backup placement of data on each storage node based on the optimal placement solution matrix.

[0051] The working principle of the system is shown in Figure 2, including:

[0052] Step 1: User and permission management settings. Implement user authentication and permission management functions, and manage the backup and recovery system throughout its lifecycle.

[0053] Step 1.1: Implement user authentication. When a user logs in, they need to compare their account name and password with the one stored in the system. If the two match, the user is authenticated. Otherwise, the user is denied access and is required to re-authenticate.

[0054] Step 1.2: Implement the permission management function. Set role permissions for different users, such as ordinary users, operation and maintenance administrators, etc., and set permissions for different user roles.

[0055] Step 2: Backup strategy management: Set the backup method and backup strategy to schedule various backup tasks.

[0056] Step 2.1: Backup mode settings. The present invention supports two backup modes: real-time backup and scheduled backup. Users can choose the corresponding backup mode according to different needs.

[0057] Step 2.2: Set the backup strategy. This system supports three backup strategies: full backup, incremental backup, and differential backup. When performing a first backup, a full backup is performed by default. For subsequent backup operations, users can select a backup strategy based on their needs.

[0058] Step 3: Data Backup. Data backup begins. An encrypted connection is established between the data node and the backup node, and data is transferred. Data is backed up according to the requested backup method. If the user chooses to back up to third-party backup software, proceed to Step 3.1. If the user chooses to back up to the system's underlying storage space, proceed to Step 3.2.

[0059] Step 3.1: The backup and recovery management system connects to the third-party backup software and directly backs up data to the third-party backup software storage space based on the backup policy set by the user.

[0060] Step 3.2: The system's underlying storage space adopts a multi-target optimal backup copy management solution. Based on the CDM golden copy concept, the steps and principles of designing a multi-target backup copy management strategy are shown in Figures 3 and 4 respectively:

[0061] For a cluster T consisting of m heterogeneous and independent storage nodes, T=(t1,t2,…,t m ), the user sets his data file K, K=(k1,k2,…,k n ), backed up to the storage system.

[0062] The same replica cannot be placed on the same storage node. The placement scheme of data replicas in the backup node is represented by the matrix TK:

[0063] in,

[0064] Step 3.2.1 is shown in Figure 5. From the three aspects of backup system reliability, latency, and load balancing, set the corresponding objective optimization function:

[0065] Step 3.2.1.1 Reliability objective function design. Let the j-th storage node t in the storage cluster T be j The failure rate is χ j , the failure rate of a storage node is related to its usage time u (the unit of u is hours), then the storage node t j The failure rate function is Data file K, K=(k1,k2,…,k n ), for each of its data files ki If there is at least one complete backup copy in the m storage nodes, then the i-th data file k i Is reliable. i The reliability of is defined as:

[0066] The reliability objective function of the entire system for a complete backup of data file K is defined as:

[0067] Step 3.2.1.2 Design of latency objective function. Let the jth storage node t in the storage cluster T be j The total network access delay is Tim_del j (Including processing delay, queuing delay, sending delay, and propagation delay). Considering the data file response time and transmission time, k i The backup copy is stored on the storage node t j The sending delay of the a-th access on:

[0068] Among them, v j is the backup node t j Data transmission rate; c i For data file k i The size of k i The access to the replicas on different backup nodes is independent of each other, then k i The total access delay function is defined as:

[0069] in, represents k i At node t j The total number of accesses on the system. Then the average delay objective function of the entire system for data file K is defined as:

[0070] Step 3.2.1.3 Load balancing objective function design. Using weighted round-robin method, each backup storage node t j , according to its hardware carrying capacity, attach the corresponding weight σ j ,σ j satisfy The present invention sets all storage nodes t j The absolute value of the difference between the actual load and the theoretical load calculated based on the hardware carrying capacity is used as the standard for measuring system load balancing. The smaller the standard deviation, the more balanced the load. Define the load balancing objective function:

[0071] Step 3.2.2 defines the optimization objective function for the multi-objective replica management strategy. When performing backups, we must provide reliable cloud backup services, guarantee latency, and avoid failures caused by excessive load on a storage node. Based on this, we define the optimization objective function:

[0072] Where X is called the decision variable, and here it is the placement plan matrix TK. The optimal placement plan matrix can be found through multi-objective optimization. f1(X)=s1(TK), ρ(k i ) represents the i-th data file k i The size of Γ(t j ) represents the jth storage node t j The capacity of ρ(k i ) and Γ(t j ) should be calculated in the same unit. j Indicates the weight assigned to the storage node based on its hardware carrying capacity.

[0073] Step 3.2.3 uses a decomposition-based multi-objective evolutionary algorithm to solve the above multi-objective problem. The specific steps are:

[0074] Step 3.2.3.1 Initialize the population size to P and randomly generate P weight vectors uniformly distributed by the Tchebycheff decomposition algorithm p=1,2,…,P, where l is the number of objective functions.

[0075] Step 3.2.3.2: Use the Tchebycheff decomposition algorithm to decompose the multi-objective problem consisting of l objectives into P sub-problems, each of which is:

[0076] in, z * is the reference point corresponding to the multi-objective function. is such that the subproblem f k (X) is the minimum value that holds, indicating the subproblem f k (X) reference point.

[0077] Step 3.2.3.3 Initialize the evolution times h = 0, randomly generate P placement scheme matrices {X 1 (0),...,X p (0),…X P (0)}. Among them, X p (0) represents the pth placement plan matrix at initialization, and the feasible region is the range of values ​​in the placement plan matrix that meets the constraints.

[0078] Step 3.2.3.4 Initialize the Pareto frontier solution set INL(0)=Φ and initialize the reference point.

[0079] Step 3.2.3.5 Initialize the neighborhood vector and find the distance p-th weight vector The nearest Neb weight vectors will be combined with The index set of the nearest Neb weight vectors is denoted as B(p)={p1,p2,...,p Neb}, p Neb =1,2,...,P.

[0080] Step 3.2.3.6 Set the maximum number of iterations H, start the iteration from h = 1, set h = H when the iteration ends, for each weight vector and its corresponding individual p=1,2,…,P is updated according to the number of iterations h:

[0081] Step 3.2.3.6.1 Evolutionary algorithm updates individuals:

[0082] Randomly select index numbers c and d from B(p) and find their corresponding individuals and Depend on and Producing new solutions

[0083] Among them, ξ is the evolution parameter.

[0084] Step 3.2.3.6.2 Update reference point z * :For each component k=1,2,L,l, if Represents the subproblem f k reference point.

[0085] Step 3.2.3.6.3 Update the population: For each neighborhood individual in B(p) b∈B(p), if but where g te represents the subproblem obtained based on the Tchebycheff decomposition algorithm, and λ represents the neighborhood individual The weight vector of .

[0086] Step 3.2.3.6.4 Update the non-dominated solution set INL(h): In INL(h), retain the dominant solution set in INL(h-1) The solution is to discard all the INL(h-1) Dominated solution, if all solutions of INL(h-1) do not dominate Will Add INL(h).

[0087] Step 3.2.3.6.5 If h = H, the obtained non-dominated solution set INL(H) is the Pareto optimal frontier solution set of the placement scheme matrix TK, the iteration terminates, and go to step 3.2.3.6; otherwise, h = h + 1, go to step 3.2.3.1

[0088] Step 3.2.3.6.6 Select a trade-off solution TK based on the decision maker's decision preference or randomly from the Pareto optimal solution set optimal , the TK optimal That is, the Pareto optimal placement matrix that satisfies the three objectives of reliability, delay, and load balancing. optimal Complete the placement and allocation of backup copies on each node.

[0089] Step 3.2.4: After the backup is completed, the backup system notifies the management node that the data backup is complete. The management node generates a backup log based on the backup information and archives it, and notifies the user that the backup is complete.

[0090] Step 4: Data recovery. The backup system receives the data recovery request and uses the backup selection algorithm to calculate the appropriate backup data according to the recovery time point in the request and perform data recovery. The specific steps are:

[0091] Step 4.1: The backup node transmits the data to the recovery node. After establishing a connection between the sender and receiver, the sender first encrypts the data. The receiver decrypts the data and performs a checksum check on the decrypted data. If the checksum passes, the data transmission is correct and can be written to the disk. If data errors occur during the transmission, the receiver can request that the sender resend the erroneous data.

[0092] Step 4.2: After data recovery is completed, the recovery node notifies the management node that recovery is complete. The management node generates a recovery log based on the data recovery information and archives it, and notifies the user that data recovery is complete.

[0093] Step 5: Log management. Create a log file for each backup task process and restore task process to track task process information, including the time of periodic backup and restore, task error information, etc., to facilitate task record retrieval and error location.

[0094] Step 6: Monitoring and management. The backup and recovery management system provides a unified monitoring screen to monitor the entire backup and recovery process. Users can view data backup and recovery status in real time and monitor the available capacity of the system's underlying storage space and the status of each storage node.

[0095] Based on the concept of multi-objective optimization, this invention abstracts the reliability, latency, and load balancing issues of the backup system into a multi-objective optimization problem. It simultaneously optimizes these three aspects of system performance using vectors. While maintaining load balancing, it improves system reliability and minimizes latency. The resulting Pareto optimal solution enables the system to achieve optimal performance under multiple objectives, thereby improving system performance in multiple areas.

[0096] Compared with the prior art, the present invention has the following advantages and effects:

[0097] (1) This patent invention designs a hybrid cloud data backup and recovery management system that integrates different types of data and backup space on a single system platform. Users can operate through the management interface on the platform and back up different types of data on the local / cloud to third-party storage software or the system's underlying storage space according to their needs, simplifying user operations and improving business efficiency.

[0098] (2) The hybrid cloud data backup and recovery management system invented by this patent based on a multi-objective optimal copy management strategy comprehensively considers the reliability of backup data, the latency of the backup system, and the stability of the backup system. Through a multi-objective optimization algorithm based on decomposition, it obtains the optimal placement plan for the backup copy of the golden copy, thereby reducing the risk of data loss and damage.

[0099] (3) The hybrid cloud data backup and recovery management system invented by this patent is based on a multi-objective optimal copy management strategy. It is managed with a graphical interface, records log information in real time, and provides a unified monitoring platform. Users can set backup and recovery strategies, and can flexibly and quickly perform data backup and recovery, helping administrators quickly locate faults and maximize the efficiency and performance of the system.

[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A data backup management method based on a multi-objective optimal copy management strategy, characterized in that: The following steps are involved: The reliability objective function, transmission delay objective function, and load balancing objective function of complete data backup in several storage nodes are defined respectively; Acquire an optimization objective function for data backup management based on the reliability objective function, the transmission delay objective function, and the load balancing objective function; Using a decomposition-based multi-objective evolutionary algorithm to obtain an optimal placement solution matrix that satisfies the optimization objective function; The data is allocated for backup placement on each storage node based on the optimal placement solution matrix.

2. The data backup management method based on the multi-objective optimal copy management strategy according to claim 1 is characterized in that: The reliability objective function is as follows: in, Indicates the i-th data file k i The reliability of j is the jth storage node t in the storage cluster T j The failure rate of u storage node, is the storage node t j The failure rate function, K represents the data file, m is the number of storage nodes, T represents a cluster consisting of m storage nodes, tk ij Represents node t j A copy on .

3. The data backup management method based on the multi-objective optimal copy management strategy according to claim 1 is characterized in that: The transmission delay objective function is as follows: Among them, v j is the backup node t j Data transmission rate; c i For data file k i The size of Indicates k i At node t j Total visits on ; Indicates k i The backup copy is stored on the storage node t j The sending delay of the a-th access on .

4. The data backup management method based on the multi-objective optimal copy management strategy according to claim 1 is characterized in that: The load balancing objective function is as follows: Among them, σ j For each backup node t j The weight of 5. The data backup management method based on the multi-objective optimal copy management strategy according to claim 1 is characterized in that: The optimization objective function is as follows: max S(X) = [f1(X), f2(X), f3(X)] T Where X is the decision variable, here is the placement plan matrix TK, f1(X) = s1(TK), ρ(k i ) represents the i-th data file k i The size of Γ(t j ) represents the jth storage node t j The capacity size.

6. The data backup management method based on the multi-objective optimal copy management strategy according to claim 1 is characterized in that: The process of using a decomposition-based multi-objective evolutionary algorithm to obtain an optimal placement solution matrix that satisfies the optimization objective function includes: Initialize the population size and randomly generate a weight vector equal to the population size; Decomposing the optimization objective function into a number of sub-problems equal to the number of populations based on the Tchebycheff decomposition algorithm; Initialize the number of evolutions and randomly generate a placement scheme matrix with the same population size; Initialize the frontier solution set and the reference point corresponding to the optimization objective function; Initialize the neighborhood vector, obtain several weight vectors closest to the last weight vector, and construct an index set based on several weight vectors; The maximum number of iterations is set and it is iterated. The last weight vector and the corresponding placement solution matrix are iteratively updated based on the number of iterations. When the maximum number of iterations is reached, the iteration is stopped to obtain the optimal placement solution matrix.

7. The data backup management method based on the multi-objective optimal copy management strategy according to claim 6 is characterized in that: The process of updating each weight vector and the corresponding placement scheme matrix based on the number of iterations includes: Randomly obtain an index number based on the index set, and obtain a placement scheme matrix corresponding to the index number; Acquire a new placement solution matrix based on the corresponding placement solution matrix, and iteratively update the reference point, the population, and the non-dominated solution set based on the new placement solution matrix; When the maximum number of iterations is reached, an updated non-dominated solution set is obtained and used as an optimal frontier solution set, a trade-off solution is selected from the optimal frontier solution set, and the trade-off solution is used as an optimal placement solution matrix.

8. The hybrid cloud data backup management method based on multi-objective optimal copy management strategy according to claim 7 is characterized in that: During the updating process of the reference point, if but Represents the subproblem f k The reference point, Represents the new placement scheme matrix; During the population update process, for each neighborhood individual in the index set like but where g te represents the subproblem obtained based on the Tchebycheff decomposition algorithm, and λ represents the neighborhood individual The weight vector of .

9. The hybrid cloud data backup management method based on multi-objective optimal copy management strategy according to claim 7 is characterized in that: In the updating process of the non-dominated solution set, the dominant solution in INL(h-1) is retained. The solution is to discard all the Dominated solution, if all solutions of INL(h-1) do not dominate Will Add INL(h), where INL(h-1) represents the non-dominated solution set after the last update. represents the new placement solution matrix, and INL(h) represents the current non-dominated solution set.

10. A hybrid cloud data backup management system based on a multi-objective optimal copy management strategy, characterized in that: The system comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, and the processor implements the method according to any one of claims 1 to 9 when executing the computer program.

Images