Discrete resource fair distribution method and device in multi-dimensional constraint network environment

Abstract

The invention provides a fair distribution method and device for discrete resources in a multi-dimensional constraint network environment, and relates to the technical field of network resource distribution, and the method comprises the steps: building a resource distribution model based on maximum and minimum fairness, representing a minimum benefit threshold value which can be obtained by a user through introducing a decision variable, and constructing a multi-dimensional resource constraint condition, the problem is converted into a solvable optimization model; and solving a fair distribution problem of discrete resources, firstly distributing the maximum bandwidth to all paths, and then reducing bandwidth distribution through iteration, so that the system meets various constraints, and meanwhile, keeping the fairness of resource distribution. The method has wide applicability and remarkable advantages, not only can meet the multi-dimensional resource constraint while ensuring fairness, but also can effectively cope with the challenge of computation complexity brought by discrete resource allocation, and in addition, the method can be popularized to various network environments with discrete resource units and multi-dimensional constraint characteristics, and has a wide application prospect. The method has important theoretical value and practical application prospect.

Description

Technical Field

[0001] The present application relates to the technical field of network resource allocation, and in particular to a method and apparatus for fair allocation of discrete resources in a multi-dimensional constrained network environment. Background Art

[0002] Against the backdrop of the rapid development of information technology, the complexity and importance of network resource allocation are becoming increasingly prominent. With the continuous emergence of new network technologies such as the Internet of Things, cloud computing, edge computing, and quantum networks, resource allocation has become a core factor affecting user experience and system performance. However, existing resource allocation algorithms primarily target continuous resources, and the related methods are relatively simple to solve. Research on discrete resources is relatively insufficient and faces numerous challenges. First, discrete resource allocation typically requires integer programming modeling, and the multidimensional nature of network resource constraints makes this type of problem NP-hard, making traditional optimization methods difficult to solve efficiently in large-scale network environments. Second, existing algorithms often overlook the fairness of resource allocation, making it difficult to meet the fairness and diversity requirements of practical applications. Therefore, a new discrete resource allocation method that balances multidimensional constraints and fairness is urgently needed to effectively address these challenges. Summary of the Invention

[0003] The present application aims to solve one of the technical problems in the related art at least to a certain extent.

[0004] To this end, the first objective of this application is to propose a method for fair allocation of discrete resources in a multi-dimensional constrained network environment.

[0005] The second objective of this application is to propose a device for fair allocation of discrete resources in a multi-dimensional constrained network environment.

[0006] The third objective of this application is to provide an electronic device.

[0007] The fourth object of this application is to provide a computer-readable storage medium.

[0008] A fifth object of this application is to provide a computer program product.

[0009] To achieve the above objectives, the first embodiment of the present application proposes a method for fair allocation of discrete resources in a multi-dimensional constrained network environment, including:

[0010] A resource allocation optimization model based on maximum-min fairness is constructed. This model introduces binary decision variables and a minimum benefit threshold variable to represent the path resource allocation scheme for each source-destination pair. With the goal of maximizing the minimum benefit threshold, a benefit calculation rule for each source-destination pair is set, and node and link resource constraints are established.

[0011] Initialize a resource allocation plan, set the bandwidth of each path to the maximum allocatable value, calculate the benefits of each source-destination pair and the resource utilization of nodes and links according to the allocation plan, and identify the nodes or links with the most serious resource constraint violations;

[0012] Selecting a path from a set of paths passing through the node or link, and reducing the bandwidth allocated to the path to reduce the amount of resources consumed by the path, wherein the path is selected based on a strategy of minimizing a minimum user benefit reduction and maximizing a resource violation mitigation amount;

[0013] The resource usage is updated and the path selection and resource adjustment process is repeated until all resource constraints are satisfied, and the minimum benefits and final resource allocation solutions for all source-destination pairs are output.

[0014] Optionally, the construction of a resource allocation optimization model based on maximum-minimum fairness introduces a binary decision variable and a minimum benefit threshold variable for representing the path resource allocation scheme for each source-destination pair, sets a benefit calculation rule for each source-destination pair with the goal of maximizing the minimum benefit threshold, and establishes node resource constraints and link resource constraints, including:

[0015] Suppose there are N source-destination SD pairs in the network, and the i∈{1,2,…,N}th SD pair has K i candidate paths, each path is allocated 0 to W units of bandwidth;

[0016] Introducing the profit matrix w ijk , which is represented by the j∈{1,2,…,K i The benefits brought by allocating k∈{1,2,…,W} unit bandwidth to each path;

[0017] Introduce binary decision variable x ijk ∈{0,1}, indicating whether k units of bandwidth are allocated to the jth path of the i-th SD pair, satisfying the allocation uniqueness constraint:

[0018]

[0019] The continuous variable z is introduced to represent the minimum benefit threshold that can be obtained among all SD pairs in the network. The optimization goal is to maximize this minimum benefit. The formula is:

[0020] maxz

[0021] Assume that the total benefit of the i-th SD pair is represented by the weighted sum of each path, and is required to be no less than the minimum benefit threshold:

[0022]

[0023] According to the network node resource constraints, let the node set be node The available resources are Q n , define the variable q ijn It represents the resource consumption at node n when one unit of bandwidth is allocated to the jth path of the i-th SD pair. Then the node resource constraint is:

[0024]

[0025] For link resource constraints, let the link set be link The available resources are C m , define the variable c ijm Denotes the resource consumption on link m when one unit of bandwidth is allocated to the jth path of the i-th SD pair. Then the link resource constraint is:

[0026]

[0027] The final resource allocation optimization model based on maximum-minimum fairness is as follows:

[0028]

[0029] The resource allocation optimization model is an integer programming problem with strong NP-hardness, and is applicable to the fair scheduling of discrete resources under multi-dimensional constraints.

[0030] Optionally, the initialization of the resource allocation plan sets the bandwidth of each path to the maximum allocatable value, calculates the benefits of each source-destination pair and the resource utilization of nodes and links according to the allocation plan, and identifies the nodes or links with the most serious resource constraint violations, including:

[0031] For each SD pair i∈{1,2,…,N} and each corresponding path j∈{1,2,…,K i}, initialize its resource allocation variable to x ijW =1, the rest x ijk =0(k≠W);

[0032] Under the initial resource allocation scheme, calculate the total profit value of each SD pair i , the formula is:

[0033]

[0034] Calculate the resource utilization of each node and link, where each node With each link Resource utilization and They are:

[0035]

[0036] By comparing the resource utilization of all nodes and links, we can identify the location where the constraint violation is most serious in the current resource allocation scheme, which is defined as:

[0037]

[0038] like This indicates that at least one resource constraint has been violated. This node or link is the location with the most severe violation in the current resource allocation scheme. In subsequent iterations, the bandwidth passing through this node or link will be adjusted first to reduce its resource consumption.

[0039] Optionally, a path is selected from the set of paths through the node or link, and the bandwidth allocated to the path is reduced to reduce the amount of resources consumed by the path. The path selection is based on a strategy of minimizing the minimum user benefit reduction and maximizing the resource violation mitigation, including:

[0040] After identifying the node or link that violates the most serious resource constraint, determine the set of all paths passing through the node or link L = {l ij}, where l ij represents the jth candidate path for the i-th source-destination pair;

[0041] Calculate the minimum benefit z for all SD pairs i∈{1,2,…,N} current =min i {profit i};

[0042] For each candidate path l in the set L ij , simulate reducing its bandwidth from k to k-1, and calculate the profit newprofit of the i-th SD pair i , the decrease in the minimum benefit of all SDs is Δz ij , resource violation mitigation amount The calculation formulas are:

[0043] newprofit i =profit i -(w ijk -w ij(k-1) )

[0044] Δz ij =z current -min{{profit p |p≠i}∪{newprofit i}}

[0045]

[0046] Based on the simulation results, a path is selected from the set L using the first strategy or the second strategy. And reduce its bandwidth by 1, that is, if the selected path is the i * The jth SD pair * paths, and k are assigned * Unit bandwidth Then assign k * -1 unit bandwidth, and at the same time

[0047] Optionally, based on the simulation results, a path is selected by a first strategy. include:

[0048] Choose Δz ij The minimum path is If there are multiple parallel candidate paths, the corresponding user simulation profit newprofit will be selected first i The largest path as

[0049] Optionally, based on the simulation results, a path is selected by a second strategy. include:

[0050] Calculate the ratio Select The minimum path is If there are multiple parallel candidate paths, the corresponding user simulation profit newprofit will be selected first i The largest path as

[0051] Optionally, the step of outputting the minimum benefits and final resource allocation solutions for all source-destination pairs until all resource constraints are satisfied includes:

[0052] When judging If it does not hold, stop the resource adjustment process and calculate the final total profit of all SD pairs i∈{1,2,…,N} i , and take the minimum value as the final minimum benefit value z = min i {profit i};

[0053] Output each decision variable x ijk The value of is used as the allocation scheme for discrete resources, and the minimum benefit value z after maximization is output.

[0054] To achieve the above-mentioned objectives, a second embodiment of the present application proposes a device for fair allocation of discrete resources in a multi-dimensional constrained network environment, comprising:

[0055] A modeling module is used to construct a resource allocation optimization model based on maximum-minimum fairness. It introduces binary decision variables and a minimum benefit threshold variable to represent the path resource allocation plan for each source-destination pair. With the goal of maximizing the minimum benefit threshold, it sets the benefit calculation rules for each source-destination pair and establishes node resource constraints and link resource constraints.

[0056] an initialization and identification module, configured to initialize a resource allocation scheme, set the bandwidth of each path to the maximum allocatable value, calculate the benefits of each source-destination pair and the resource utilization of nodes and links according to the allocation scheme, and identify the nodes or links with the most severe resource constraint violations;

[0057] a path adjustment module, configured to select a path from a set of paths passing through the node or link, and reduce the bandwidth allocated to the path to reduce the amount of resources consumed by the path, wherein the path selection is based on a strategy of minimizing the minimum user benefit reduction and maximizing the amount of resource violation mitigation;

[0058] The output module is used to update the resource usage and repeatedly execute the path selection and resource adjustment process until all resource constraints are satisfied, and then output the minimum benefits of all source-destination pairs and the final resource allocation plan.

[0059] To achieve the above-mentioned purpose, a third embodiment of the present application provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor;

[0060] The memory stores computer-executable instructions;

[0061] The processor executes the computer-executable instructions stored in the memory to implement the method as described in any one of the first aspects.

[0062] To achieve the above-mentioned purpose, the fourth embodiment of the present application proposes a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are executed by a processor, they are used to implement the method as described in any one of the first aspects.

[0063] To achieve the above-mentioned objectives, the fifth embodiment of the present application proposes a computer program product, which implements any one of the methods in the first aspect when executed by a processor.

[0064] The technical solutions provided by the embodiments of this application bring at least the following beneficial effects:

[0065] This application has wide applicability and significant advantages. It can not only meet multi-dimensional resource constraints while ensuring fairness, but also effectively cope with the computational complexity challenges brought by allocating discrete resources. In addition, this application can be extended to various network environments with discrete resource units and multi-dimensional constraints. It has important theoretical value and practical application prospects.

[0066] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0068] Figure 1 This is a flow chart of a method for fair allocation of discrete resources in a multi-dimensional constrained network environment provided by an embodiment of the present application. DETAILED DESCRIPTION

[0069] The following describes in detail embodiments of the present application, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0070] In response to the problem of fair allocation of discrete resources in a multi-dimensional constrained network environment, an embodiment of the present application proposes a method for fair allocation of discrete resources in a multi-dimensional constrained network environment. First, a resource allocation model based on max-min fairness is established. By introducing decision variables to represent the minimum benefit threshold that a user can obtain, and constructing multi-dimensional resource constraints, the problem is converted into a solvable optimization model; then the fair allocation problem of discrete resources is solved, the maximum bandwidth is allocated to all paths, and then the bandwidth allocation is iteratively reduced so that the system meets various constraints while maintaining the fairness of resource allocation.

[0071] Figure 1 This is a flow chart of a method for fair allocation of discrete resources in a multi-dimensional constrained network environment provided by an embodiment of the present application. Figure 1 Specifically, the method includes the following steps:

[0072] Step 101: Construct a resource allocation optimization model based on maximum-minimum fairness, introduce binary decision variables and minimum benefit threshold variables for representing the path resource allocation scheme for each source-destination pair, set the benefit calculation rules for each source-destination pair with the goal of maximizing the minimum benefit threshold, and establish node resource constraints and link resource constraints.

[0073] In an embodiment of the present application, step 101 is used to construct a resource allocation optimization model based on maximum-minimum fairness. The model is used to optimize the discrete resource allocation strategy between multiple paths while satisfying multi-dimensional resource constraints, maximize the minimum benefit value among all communication demand users, and achieve a balance between the fairness of resource allocation and the overall performance of the system.

[0074] The above modeling process is described in detail below using quantum network as an example.

[0075] A quantum network is a novel network architecture based on the principles of quantum mechanics, designed to enable the long-distance transmission and distributed processing of quantum information. Its basic components include quantum nodes and quantum channels. Quantum nodes are typically equipped with quantum memory and quantum processors, capable of performing quantum operations, storing quantum bits (qubits), and participating in quantum protocols. Quantum channels, which transmit entangled photons via optical fibers, free space, and other means, establish quantum entanglement connections between different quantum nodes.

[0076] Compared to traditional communication networks, quantum networks possess significantly different characteristics. Their core resources are qubits and quantum entanglement, which are not only discrete and finite but also possess physical properties such as non-cloning, non-relayability, and susceptibility to decoherence. Furthermore, quantum operations are significantly probabilistic and uncertain. These characteristics present unprecedented challenges in resource scheduling, routing planning, and load balancing. Therefore, achieving efficient and fair resource allocation in quantum networks requires careful consideration of resource discreteness and fairness objectives, while also satisfying complex, multi-dimensional physical constraints.

[0077] Assume that there are several source-destination pairs (SD pairs) in a quantum network that need to communicate. Each SD pair can choose multiple candidate paths for communication. Since network resources (such as the quantum bit storage capacity of each node and the quantum channel capacity of each link) are limited and discrete, how to allocate appropriate resources to all SD pairs while satisfying multi-dimensional resource constraints has become an urgent problem to be solved. This problem not only requires full utilization of limited network resources, but also takes into account fairness between different SD pairs to avoid some users being at a disadvantage for a long time due to resource constraints.

[0078] To achieve the above objectives, the present invention constructs a mathematical optimization model in the following manner:

[0079] For N source-destination (SD) pairs in a quantum network, let the i∈{1,2,…,N}th SD pair have K i candidate paths, each path is assigned 0 to W quantum links;

[0080] Introducing the profit matrix w ijk , which is represented by the j∈{1,2,…,K i The benefits brought by allocating k∈{1,2,…,W} quantum links to each path;

[0081] Introduce binary decision variable x ijk ∈{0,1}, indicating whether k quantum links are allocated to the j-th path of the i-th SD pair, satisfying the uniqueness constraint of quantum link allocation:

[0082]

[0083] A continuous variable z is introduced to represent the minimum gain threshold that can be obtained among all SD pairs in the quantum network. The optimization goal is to maximize this minimum gain. The formula is:

[0084] maxz

[0085] Assume that the total benefit of the i-th SD pair is represented by the weighted sum of each path, and is required to be no less than the minimum benefit threshold:

[0086]

[0087] In addition, with respect to multi-dimensional resource constraints in the network, the embodiments of the present application also consider two types of key constraints: node resource constraints and link resource constraints.

[0088] Specifically, in view of the network node resource constraints, the embodiment of the present application sets the node set as node The quantum bit capacity is Q n , define the variable q ijn Denotes the quantum bit consumption per unit link of the j-th path of the i-th SD pair on node n, and the node resource constraint is:

[0089]

[0090] In view of the link resource constraint, the embodiment of the present application sets the link set as link The capacity is C m , define the variable c ijm Indicates whether the jth path of the i-th SD pair passes through link m, then the link resource constraint is:

[0091]

[0092] The final resource allocation optimization model based on maximum-minimum fairness is as follows:

[0093]

[0094] The resource allocation optimization model, a strongly NP-hard integer programming problem, is applicable to the fair scheduling of discrete resources under multidimensional constraints. Through the above method, the fair allocation problem of discrete network resources is transformed into a solvable integer programming model. This model not only integrates fairness objectives and multidimensional resource constraints, but also provides a theoretical foundation for subsequent algorithm design and efficient solution. It is worth emphasizing that this model has good scalability and versatility, and is applicable to discrete resource allocation problems in various network environments, such as quantum networks, traditional communication networks, edge computing, and cloud computing. For other types of networks or discrete resources, the model can be flexibly applied by simply calculating the corresponding payoff matrix and replacing the multidimensional resource constraints with the constraints of the specific scenario.

[0095] Step 102: Initialize the resource allocation plan, set the quantum link of each path to the maximum allocatable value, calculate the benefits of each source-destination pair and the resource utilization of nodes and links according to the allocation plan, and identify the nodes or links with the most serious resource constraint violations.

[0096] It should be noted that the resource allocation optimization model constructed in this application is a maximum-minimum variant of the multidimensional multi-choice knapsack problem (MMKP). MMKP is a strong NP-hard problem and there is no pseudo-polynomial time optimal solution algorithm. Since MMKP is a special case of the resource allocation optimization model when N=1, the resource allocation optimization model is also a strong NP-hard problem. Therefore, it is urgent to design an efficient algorithm that can obtain a high-quality approximate solution while ensuring algorithm performance.

[0097] After constructing the resource allocation optimization model based on maximum-minimum fairness, next, the embodiment of the present application will solve the model based on steps 102-104.

[0098] Specifically, after the resource allocation model is established, the embodiment of the present application first performs a global initialization resource allocation based on step 102. Assume that each source-destination pair i∈{1,2,…,N} has K i optional paths, then for each path j∈{1,2,…,K i}, initialize its resource allocation variable to x ijW =1, the rest x ijk =0(k≠W).

[0099] Each path is initialized to allocate the maximum amount of bandwidth (e.g., the maximum number of quantum links). This initialization strategy ensures that each path obtains the maximum potential benefit under unconstrained conditions, providing a benchmark state for subsequent iterations.

[0100] After initialization, the embodiment of the present application calculates the total profit value profit of each source-destination pair i , the benefit value is defined as the sum of the weighted benefits obtained by the i-th SD pair through all paths under the current resource allocation scheme. The calculation formula is as follows:

[0101]

[0102] Furthermore, in order to evaluate the pressure of the current allocation on network resources, the embodiment of the present application evaluates the resource utilization of all network nodes and links. The link set is Then each node Indicates its resource consumption relative to the available resources Q n The proportion is calculated as follows:

[0103]

[0104] Likewise, each link Resource utilization Indicates its total usage relative to the available resource C m The proportion is calculated as follows:

[0105]

[0106] Based on the above calculation results, this application further compares the resource utilization of all nodes and links to identify the locations where resource constraints are most severely violated under the current resource allocation scheme, namely:

[0107]

[0108] like This indicates that the resource usage of at least one node or link exceeds the limit, violating the resource constraint. In this case, the embodiment of the present application regards this node or link as the "location with the most serious resource conflict" and prioritizes the set of paths passing through this resource location in subsequent iterations as the focus of resource adjustment.

[0109] Through the above steps, this application can identify the main constraint bottlenecks faced by the current resource configuration in the initial stage, provide structured input information for subsequent execution path fallback, resource reduction and iterative adjustment, and effectively improve the convergence efficiency and fairness control capabilities of the model solution.

[0110] Step 103 : Select a path from the set of paths passing through the node or link, and reduce the bandwidth allocated to the path to reduce the amount of resources consumed. The path selection is based on the strategy of minimizing the user's minimum benefit reduction and maximizing the resource violation mitigation.

[0111] After identifying the node or link that has the most serious resource constraint violation in step 102, the embodiment of the present application uses it as the resource conflict core and further determines the set of all paths passing through the node or link, which is recorded as L={l ij}, where l ij represents the jth candidate path for the i-th source-destination pair. This path set is the candidate target set for this round of resource adjustment.

[0112] Calculate the minimum benefit z for all SD pairs i∈{1,2,…,N} current =min i {profit i}.

[0113] For each candidate path l in the set L ij , the embodiment of the present application simulates reducing the number of quantum links from k to k-1, and calculates the corresponding three key indicators: the profit newprofit of the i-th SD pair i , the decrease in the minimum benefit of all SDs is Δz ij , and the amount of resource violation mitigation They are defined as follows:

[0114] The profit of the i-th SD pair is newprofit i : Measures the benefits of users after the path is adjusted. The calculation formula is:

[0115] newprofit i =profit i -(w ijk -w ij(k-1) )

[0116] The decrease Δz of all SDs to the minimum return ij : Measures the negative impact of the path adjustment on fairness. The calculation formula is:

[0117] Δz ij =z current -min{{profit p |p≠i}∪{newprofit i}}

[0118] Resource Violation Mitigation Amount Measures the contribution of the path adjustment to alleviating the current resource overload node or link. The calculation formula is:

[0119]

[0120] According to the above simulation indicators, the embodiment of the present application adopts two types of path selection strategies to select the optimal path from the path set L. And perform resource rollback operations. Among them:

[0121] The first strategy: focus on fairness first. This strategy takes minimizing the loss of benefits as the primary goal and selects the path set that makes Δz ij The minimum path is If there are multiple parallel candidate paths, the corresponding user simulation profit newprofit will be selected first i The largest path as To improve the fairness of overall distribution.

[0122] The second strategy: balance fairness and feasibility. This strategy considers both resource mitigation efficiency and minimum revenue loss, and calculates the comprehensive ratio of each path. Select The minimum path is In order to reduce the system pressure and control the loss of user income. If there are multiple parallel candidate paths, the corresponding user simulation income newprofit will also be given priority. i The largest path as

[0123] Once the optimal path is determined Start executing resource fallback operation. Assume that the path is currently assigned k * quantum link Then modify it to allocate k * -1 quantum link, and at the same time The remaining paths retain their original assignments.

[0124] By executing this step, the algorithm can alleviate the areas with the most serious resource violations in each round of iteration, while seeking the smallest sacrifice in the benefit space, thereby gradually converging to a feasible solution that takes into account both fairness and feasibility.

[0125] It should be pointed out in particular that the path selection and fallback mechanism of this application is not only applicable to quantum network scenarios, but also has good adaptability and scalability for other types of network systems and discrete resource allocation problems. The algorithm framework proposed in this application can be flexibly applied by simply mapping the resource types, benefit indicators and resource constraints in the specific network to the general variable structure and formula of the model of this application. For example: in traditional communication networks, link bandwidth, processing power, etc. can be modeled as discrete resources; in edge computing or cloud computing scenarios, task allocation and server capacity can be used as resource constraint dimensions; the first strategy and the second strategy can also be used to evaluate the benefit changes and resource release efficiency brought about by task migration, so as to achieve controllable and balanced scheduling process.

[0126] Through the adaptive adjustment of strategy simulation and objective function design, the method of this application can support the fair scheduling optimization problem of complex discrete resources in different types of network environments.

[0127] Step 104 : Update resource usage and repeat the path selection and resource adjustment process until all resource constraints are satisfied, and then output the minimum benefits and final resource allocation solutions for all source-destination pairs.

[0128] After executing the path fallback operation, the embodiment of the present application updates the resource usage status of all nodes and links in real time. Specifically, for each round of adjusted resource allocation plan, the resource utilization of each node and each link is recalculated. After the update is completed, it is re-determined whether there are still nodes or links that violate resource constraints, that is, whether there are:

[0129]

[0130] If the judgment result is still This means that there are still nodes or links with excessive resources. The embodiment of the present application continues to perform the path set identification, path simulation evaluation and resource fallback operations in step 103, and repeats the resource update and judgment until the resource utilization of all nodes and links does not exceed their corresponding capacity.

[0131] When judging If it is not true, that is, all resource usage in the network is within the allowed range, it means that the current resource allocation scheme has satisfied all resource constraints. At this point, the embodiment of the present application terminates the path adjustment iteration process and enters the final result output stage.

[0132] Under the final feasible allocation scheme, calculate the final total profit of all SD pairs i∈{1,2,…,N} i , extract the minimum benefit value from all source-destination pairs as the evaluation index of the final system fairness, that is: z = min i {profit i}.

[0133] Finally, output each decision variable x ijk The value of is taken as the discrete resource allocation result of all paths, and the maximized minimum benefit value z is output at the same time, which is the optimal solution or high-quality approximate solution of the optimization model proposed in this application under the current network conditions.

[0134] Through the above-mentioned iterative optimization and termination judgment mechanism, the embodiment of the present application can achieve fair and efficient discrete resource scheduling in a complex environment with limited resources and multi-dimensional constraints, taking into account both user benefit fairness and system resource feasibility, and has good engineering applicability and cross-scenario scalability.

[0135] Furthermore, to verify the applicability and performance of the proposed discrete resource fair allocation algorithm in a practical quantum network environment, the present invention constructed multiple simulated network scenarios and systematically evaluated the allocation effect under different network scales and different numbers of source-destination pairs (SD pairs). The comparison methods included the Weighted Round Robin (WRR) algorithm, the Subgradient Method, and the commercial optimizer Gurobi, with Gurobi being considered as a reference for the near-optimal solution.

[0136] During the simulation, the throughput of the quantum network is used as the performance evaluation indicator. This indicator is defined as the expected number of quantum links that can be established between each SD pair in the network. Based on this indicator, the embodiment of the application constructs the profit matrix w ijk , the specific calculation formula is as follows:

[0137]

[0138] in, Represents the jth path l of the i-th SD pair ij The probability of generating at least r successful links, l° represents the subpath of l without the first and last nodes, p n is the success probability of entanglement exchange at node n.

[0139] The experimental results show that the maximum and minimum benefits of the algorithm proposed in this application are significantly better than those of the WRR algorithm and the subgradient algorithm under all network sizes and numbers of SD pairs, and are very close to the optimal solution given by Gurobi. For example, in a scenario where the network size is 250 nodes, 600 edges, and there are 5 SD pairs communicating at the same time, compared with the optimal solution given by Gurobi, the median relative error of the second strategy of the algorithm proposed in this application is only 0.02%, the first strategy is 3.20%, while the subgradient method and WRR are 11.57% and 55.67% respectively. In terms of running time, WRR is the fastest (about 10 -1 ms level), the algorithm proposed in this application is second (both strategies are about 10 0 ms level), Gurobi (about 10 1 ms level) and sub-gradient method (about 10 2 ms level) is slower.

[0140] Of the two strategies in the algorithm proposed in this application, the second strategy significantly reduces the error rate at the expense of slightly longer runtime compared to the first strategy, demonstrating the effectiveness of the trade-off between maximum-minimum fairness and the degree of constraint violation. Overall, the algorithm proposed in this application achieves high computational efficiency while ensuring high solution quality, making it suitable for latency-sensitive scenarios such as practical quantum networks that require discrete resource allocation.

[0141] In summary, the discrete resource allocation method proposed in this application can not only obtain high-quality allocation results under strict resource constraints, but also has high computational efficiency and strong stability. It is particularly suitable for scenarios such as quantum networks that are sensitive to delays and have highly discrete resources, and has significant practical application value and promotion potential.

[0142] To implement the above embodiment, the present application also proposes a device for fair allocation of discrete resources in a multi-dimensional constrained network environment. The device includes:

[0143] The modeling module is used to build a resource allocation optimization model based on maximum-minimum fairness. It introduces binary decision variables and minimum benefit threshold variables to represent the path resource allocation plan for each source-destination pair. With the goal of maximizing the minimum benefit threshold, it sets the benefit calculation rules for each source-destination pair and establishes node resource constraints and link resource constraints.

[0144] The initialization and identification module is used to initialize the resource allocation plan, set the bandwidth of each path to the maximum allocatable value, calculate the benefits of each source-destination pair and the resource utilization of nodes and links according to the allocation plan, and identify the nodes or links with the most serious resource constraint violations;

[0145] A path adjustment module is used to select a path from a set of paths passing through nodes or links and reduce the bandwidth allocated to the path to reduce the amount of resources consumed. The path selection is based on the strategy of minimizing the minimum user benefit reduction and maximizing the amount of resource violation mitigation.

[0146] The output module is used to update resource usage and repeat the path selection and resource adjustment process until all resource constraints are met, and then output the minimum benefits and final resource allocation plan for all source-destination pairs.

[0147] Regarding the apparatus in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0148] In order to implement the above embodiments, the present application also proposes an electronic device, comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method provided by the above embodiments.

[0149] In order to implement the above embodiments, the present application also proposes a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the methods provided by the above embodiments.

[0150] In order to implement the above embodiments, the present application also proposes a computer program product, including a computer program, which implements the methods provided by the above embodiments when executed by a processor.

[0151] The collection, storage, use, processing, transmission, provision and disclosure of user personal information involved in this application are in compliance with relevant laws and regulations and do not violate public order and good morals.

[0152] It is important to note that personal information collected from users should be used for legitimate and reasonable purposes and should not be shared or sold beyond these legitimate uses. Furthermore, such collection / sharing should be conducted only after receiving the user's informed consent, including but not limited to notifying the user to read the user agreement / user notice and sign an agreement / authorization that includes the relevant user information before using the feature. Furthermore, any necessary steps must be taken to safeguard and secure access to such personal information and ensure that others with access to personal information comply with its privacy policy and procedures.

[0153] This application contemplates providing implementations that allow users to selectively block the use or access of personal information data. Specifically, this disclosure contemplates providing hardware and / or software to prevent or block access to such personal information data. Risks can be minimized by limiting data collection and deleting data once it is no longer needed. Furthermore, where applicable, such personal information can be de-identified to protect user privacy.

[0154] In the descriptions of the foregoing embodiments, the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are mutually inconsistent.

[0155] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of such features. Throughout the description of this application, "plurality" means at least two, for example, two, three, etc., unless otherwise specifically defined.

[0156] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present application belong.

[0157] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection having one or more wires (electronic devices), a portable computer disk cartridge (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). Furthermore, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing it in another suitable manner if necessary, and then storing it in a computer memory.

[0158] It should be understood that various parts of the present application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0159] Those skilled in the art will understand that all or part of the steps in the method of the above embodiment can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0160] In addition, the functional units in the various embodiments of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into a module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0161] The storage medium mentioned above may be a read-only memory, a magnetic disk, or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application. Persons skilled in the art may make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.

[0162] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this application can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of this application can be achieved. This is not limited herein.

[0163] The above specific embodiments do not constitute a limitation on the scope of protection of this application. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application shall be included within the scope of protection of this application.

Claims

1. A method for fair allocation of discrete resources in a multi-dimensional constrained network environment, characterized in that: The following steps are involved: A resource allocation optimization model based on maximum-min fairness is constructed. This model introduces binary decision variables and a minimum benefit threshold variable to represent the path resource allocation scheme for each source-destination pair. With the goal of maximizing the minimum benefit threshold, a benefit calculation rule for each source-destination pair is set, and node and link resource constraints are established. Initialize a resource allocation plan, set the bandwidth of each path to the maximum allocatable value, calculate the benefits of each source-destination pair and the resource utilization of nodes and links according to the allocation plan, and identify the nodes or links with the most serious resource constraint violations; Selecting a path from a set of paths passing through the node or link, and reducing the bandwidth allocated to the path to reduce the amount of resources consumed by the path, wherein the path is selected based on a strategy of minimizing a minimum user benefit reduction and maximizing a resource violation mitigation amount; The resource usage is updated and the path selection and resource adjustment process is repeated until all resource constraints are satisfied, and the minimum benefits and final resource allocation solutions for all source-destination pairs are output.

2. The method according to claim 1, characterized in that The resource allocation optimization model based on maximum-minimum fairness is constructed by introducing binary decision variables and minimum benefit threshold variables for representing the path resource allocation scheme for each source-destination pair. With the goal of maximizing the minimum benefit threshold, a benefit calculation rule for each source-destination pair is set, and node resource constraints and link resource constraints are established, including: Suppose there are N source-destination (SD) pairs in the network, and the i∈{1,2,…,N}th SD pair has K i candidate paths, each path is allocated 0 to W units of bandwidth; Introducing the profit matrix w ijk , which is represented by the j∈{1,2,…,K i The benefits brought by allocating k∈{1,2,…,W} unit bandwidth to each path; Introduce binary decision variable x ijk ∈{0,1}, indicating whether k units of bandwidth are allocated to the jth path of the i-th SD pair, satisfying the allocation uniqueness constraint: The continuous variable z is introduced to represent the minimum benefit threshold that can be obtained among all SD pairs in the network. The optimization goal is to maximize this minimum benefit. The formula is: maxz Assume that the total benefit of the i-th SD pair is represented by the weighted sum of each path, and is required to be no less than the minimum benefit threshold: According to the network node resource constraints, let the node set be node The available resources are Q n , define the variable q ijn represents the resource consumption on node n when one unit of bandwidth is allocated to the jth path of the i-th SD pair. The node resource constraint is: For link resource constraints, let the link set be link The available resources are C m , define the variable c ijm represents the resource consumption on link m when one unit of bandwidth is allocated to the jth path of the i-th SD pair. The link resource constraint is: The final resource allocation optimization model based on maximum-minimum fairness is as follows: max z st x ijk ∈{0,1} The resource allocation optimization model is an integer programming problem with strong NP-hardness, and is applicable to the fair scheduling of discrete resources under multi-dimensional constraints.

3. The method according to claim 2, characterized in that The initialization of the resource allocation scheme sets the bandwidth of each path to the maximum allocatable value, calculates the benefits of each source-destination pair and the resource utilization of nodes and links according to the allocation scheme, and identifies the nodes or links with the most serious resource constraint violations, including: For each SD pair i∈{1,2,…,N} and each corresponding path j∈{1,2,…,K i }, initialize its resource allocation variable to x ijW =1, the rest x ijk =0(k≠W); Under the initial resource allocation scheme, calculate the total profit value of each SD pair i , the formula is: Calculate the resource utilization of each node and link, where each node With each link Resource utilization and They are: By comparing the resource utilization of all nodes and links, we can identify the location where the constraint violation is most serious in the current resource allocation scheme, which is defined as: like This indicates that at least one resource constraint has been violated. This node or link is the location with the most severe violation in the current resource allocation scheme. In subsequent iterations, the path bandwidth passing through this node or link will be adjusted first to reduce its resource consumption.

4. The method according to claim 3, characterized in that The selecting of a path from a set of paths passing through the node or link and reducing the bandwidth allocated to the path to reduce the amount of resources consumed by the path, wherein the path selection is based on a strategy of minimizing the minimum benefit reduction for users and maximizing the amount of resource violation mitigation, includes: After identifying the node or link that violates the most serious resource constraint, determine the set of all paths passing through the node or link L = {l ij }, where l ij represents the jth candidate path for the i-th source-destination pair; Calculate the minimum benefit z for all SD pairs i∈{1,2,…,N} current =min i {profit i }; For each candidate path l in the set L ij , simulate reducing its bandwidth from k to k-1, and calculate the profit newprofit of the i-th SD pair i , the decrease in the minimum benefit of all SDs is Δz ij , resource violation mitigation amount The calculation formulas are: newprofit i =profit i -(In ijk -In ij(k-1) ) Δz ij =z current -min{{profit p |p≠i}∪{newprofit i }} Based on the simulation results, path l is selected from the set L by the first strategy or the second strategy. i*j* , and reduce its bandwidth by 1, that is, if the selected path l i*j* is the i * The jth SD pair * paths, and k are assigned * Unit bandwidth (x i*j*k* =1), then assign k instead * -1 unit bandwidth, that is, let x i*j*(k*-1) =1,x i*j*k* =0.

5. The method according to claim 4, characterized in that Based on the simulation results, the path l is selected by the first strategy. i*j* ,include: Choose Δz ij The minimum path is l i*j* If there are multiple parallel candidate paths, the corresponding user simulation profit newprofit is selected first i The largest path is l i*j* .

6. The method according to claim 4, characterized in that Based on the simulation results, the path l is selected by the second strategy. i*j* ,include: Calculate the ratio Select The minimum path is l i*j* If there are multiple parallel candidate paths, the corresponding user simulation profit newprofit is selected first i The largest path is l i*j* .

7. The method according to claim 5 or 6, characterized in that When all resource constraints are satisfied, the minimum benefits and final resource allocation solutions for all source-destination pairs are output, including: When judging If it does not hold, stop the resource adjustment process and calculate the final total profit of all SD pairs i∈{1,2,…,N} i , and take the minimum value as the final minimum benefit value z = min i {profit i }; Output each decision variable x ijk The value of is used as the allocation scheme for discrete resources, and the minimum benefit value z after maximization is output.

8. A device for fair allocation of discrete resources in a multi-dimensional constrained network environment, characterized in that: include: A modeling module is used to construct a resource allocation optimization model based on maximum-minimum fairness. It introduces binary decision variables and a minimum benefit threshold variable to represent the path resource allocation plan for each source-destination pair. With the goal of maximizing the minimum benefit threshold, it sets the benefit calculation rules for each source-destination pair and establishes node resource constraints and link resource constraints. an initialization and identification module, configured to initialize a resource allocation scheme, set the bandwidth of each path to the maximum allocatable value, calculate the benefits of each source-destination pair and the resource utilization of nodes and links according to the allocation scheme, and identify the nodes or links with the most severe resource constraint violations; a path adjustment module, configured to select a path from a set of paths passing through the node or link, and reduce the bandwidth allocated to the path to reduce the amount of resources consumed by the path, wherein the path selection is based on a strategy of minimizing the minimum user benefit reduction and maximizing the amount of resource violation mitigation; The output module is used to update the resource usage and repeatedly execute the path selection and resource adjustment process until all resource constraints are satisfied, and then output the minimum benefits of all source-destination pairs and the final resource allocation plan.

9. An electronic device, characterized in that: include: a processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 7 when executed by a processor.

Images