Elastic optical network virtual optical network mapping method based on carbon emission efficiency factor

Abstract

The invention discloses an elastic optical network and virtual optical network mapping method based on a carbon emission efficiency factor, and belongs to the technical field of elastic optical network and virtual optical network mapping. The method comprises the following steps: acquiring physical nodes and physical link parameters of an elastic optical network, and virtual nodes and virtual link parameters of a virtual optical network; calculating a carbon emission efficiency factor CEEF of the physical node, and constructing a subgraph candidate set according to the sequence of the CEEF and the adjacency relation; the virtual nodes are arranged in a descending order according to computing resource requirements and sequentially mapped with the sub-graph physical nodes arranged in a CEEF descending order, and when resources are insufficient, a candidate set is expanded for retry; after node mapping succeeds, calculating a shortest path for each virtual link, selecting a path meeting the requirement of a continuous spectrum slot to complete link mapping, and expanding a candidate set and restarting a complete process when the link mapping fails until the link mapping succeeds or resources are used up; according to the invention, efficient low-carbon mapping of the virtual optical network is realized.

Description

Technical Field

[0001] This invention relates to the field of flexible optical network and virtual optical network mapping technology, and particularly to a method for mapping flexible optical networks and virtual optical networks based on carbon emission efficiency factors. Background Technology

[0002] With the rapid penetration of Internet of Things (IoT) technology into fields such as healthcare, smart homes, and intelligent transportation, the data traffic generated by various applications is growing exponentially. A large amount of data needs to be stored and processed using geographically distributed cloud data centers (DCs). However, the high energy consumption of data centers is becoming increasingly prominent. Related research indicates that their electricity consumption already accounts for 2%-3% of global total electricity consumption, and due to regional differences in the energy structure upon which power generation depends, the carbon footprint of data centers in different geographical locations varies significantly. Against the backdrop of increasingly urgent needs for global warming and greenhouse gas (GHG) emission reduction, reducing the carbon emissions of cloud data centers and related networks has become a core challenge in the field of green communications.

[0003] Elastic Optical Network (EON), a key supporting technology for connecting geographically distributed data centers, can perform fine-grained allocation of spectrum resources (e.g., 12.5 GHz, 6.25 GHz, or smaller) to dynamically match the resource requirements of different services. Virtual Optical Network Embedding (VONE), as a core component of network virtualization, is responsible for mapping the nodes and links of the VON to the physical resources of the underlying Elastic Optical Network, directly impacting network resource utilization and carbon emissions. Existing carbon emission-aware VONE methods frequently change the underlying nodes allocated to virtual nodes when remaining resources change dynamically, leading to frequent device start-ups and shutdowns, increased energy consumption, and ultimately increased carbon emissions. Furthermore, in dynamic network scenarios, traditional methods need to search for feasible mapping schemes throughout the entire underlying network, easily activating a large number of redundant underlying nodes, further exacerbating energy consumption and carbon emission problems, making it difficult to balance network performance and green emission reduction goals.

[0004] Furthermore, existing research on virtual optical network mapping in flexible optical networks (FONs) sometimes focuses solely on energy consumption optimization without adequately considering the carbon intensity differences of energy sources, leading to a mismatch between energy consumption reduction and carbon emission reduction. Other methods, while addressing carbon emissions, lack effective control over the mapping region, resulting in high computational complexity and long runtimes in large-scale networks, failing to meet the real-time and efficiency requirements of practical communication networks. Therefore, there is an urgent need for a flexible optical network virtual optical network mapping scheme that can establish a correlation between carbon emissions and energy consumption, precisely control the mapping region, and balance network performance with carbon emission reduction. Summary of the Invention

[0005] The purpose of this invention is to provide a flexible optical network virtual optical network mapping method based on carbon emission efficiency factor, which can realize efficient and low-carbon mapping of virtual optical networks.

[0006] To address the aforementioned technical problems, embodiments of the present invention provide a method for mapping a virtual optical network in a flexible optical network based on a carbon emission efficiency factor, comprising the following steps: Obtain the carbon emission factor and computing resource capacity of each physical node in the elastic optical network, as well as the number of continuous spectrum slots in each physical link; obtain the computing resource capacity of each virtual node in the virtual optical network, as well as the bandwidth requirements of each virtual link. The carbon emission efficiency factor is determined by the relationship between the carbon emission factor of each physical node and the computational resource capacity and the carbon emission amount. A candidate set of subgraphs for the elastic optical network is constructed by sorting the carbon emission efficiency factors of each physical node and iterating the adjacency relationship; the candidate set of subgraphs includes subgraph physical nodes and subgraph physical links. The virtual nodes are sorted in descending order of their computing resource capacity, and the subgraph physical nodes are sorted in descending order of their carbon emission efficiency factor. The virtual nodes are mapped to the corresponding subgraph physical nodes in descending order. If the computing resource capacity of the current virtual node exceeds that of the subgraph physical node, the virtual node mapping in this round is considered a failure. The subgraph candidate set is expanded by one physical node and then reconstructed. The above sorting and node mapping process is repeated until all virtual nodes are successfully mapped or the subgraph candidate set covers the entire elastic optical network. For all virtual links between mapped virtual nodes, a preset number of shortest physical links are calculated in the elastic optical network. The number of consecutive spectrum slots required for a virtual link is determined according to bandwidth requirements. If the number of consecutive spectrum slots of a shortest physical link is not less than the number of consecutive spectrum slots required for the virtual link, then the virtual link is mapped to that physical link. If any virtual link mapping fails, the subgraph candidate set is expanded by one physical node and then reconstructed. The complete process starting from node mapping is re-executed until all virtual links are successfully mapped or the subgraph candidate set covers the entire elastic optical network.

[0007] In some optional embodiments, the method of constructing a candidate set of subgraphs for the elastic optical network by sorting the carbon emission efficiency factors of each physical node and iterating the adjacency relationship includes the following steps: The subgraph candidate set starts with the physical node with the highest carbon emission efficiency factor. Through an iterative process, the physical node with the highest carbon emission efficiency factor is selected from the unselected physical nodes that are directly connected to the current subgraph candidate set and added to the subgraph candidate set, until the number of physical nodes in the subgraph candidate set reaches the number of virtual nodes to be mapped.

[0008] In some optional embodiments, determining the carbon emission efficiency factor by relating the carbon emission factor of each physical node and the computational resource capacity to carbon emissions includes the following steps: The formula for calculating the carbon emission factor of a physical node is as follows: ; In the formula, Carbon emission factors at physical nodes Let i be the set of power sources for physical node i. For power source a, a pair of physical nodes The percentage of contribution, For physical nodes Geographical location of power source The carbon emission coefficient; The formula for calculating the carbon emissions of a physical node is as follows: ; In the formula, Let i be the fixed energy consumption of physical node i. , Let i be the maximum energy consumption of physical node i. , The computing resources allocated to physical node i for the virtual node. The total computing resources of physical node i; make and By normalizing parameters to eliminate differences between different physical nodes, the carbon emission efficiency factor (CEEF) is increased. i The formula is as follows: ; In the formula, Carbon emission factors at physical nodes Calculate the resource capacity for physical nodes.

[0009] In some optional embodiments, the calculation formula for determining the number of consecutive spectrum slots required for the virtual link based on bandwidth requirements is as follows: ; In the formula, The required number of consecutive spectrum slots; For virtual link bandwidth requirements; represents the modulation format coefficients, where BPSK is 1, QPSK is 2, 8QAM is 3, 16QAM is 4, 64QAM is 6, and 256QAM is 8; G represents the number of guard spectrum slots, which is set to 1.

[0010] Embodiments of the present invention also provide a flexible optical network virtual optical network mapping system based on a carbon emission efficiency factor, comprising: The data acquisition module acquires the carbon emission factor, computing resource capacity, and carbon emissions of each physical node in the elastic optical network, as well as the number of continuous spectrum slots in each physical link; it also acquires the computing resource capacity of each virtual node in the virtual optical network and the bandwidth requirements of each virtual link. The carbon emission efficiency factor calculation module is used to determine the carbon emission efficiency factor by the relationship between the carbon emission factor of each physical node and the computing resource capacity and the carbon emission amount. The subgraph candidate set construction module is used to construct a subgraph candidate set of the elastic optical network by sorting the carbon emission efficiency factors of each physical node and iterating the adjacency relationship; the subgraph candidate set includes subgraph physical nodes and subgraph physical links; The virtual node mapping module sorts virtual nodes in descending order of their computing resource capacity and subgraph physical nodes in descending order of their carbon emission efficiency factor. It then maps virtual nodes sequentially to subgraph physical nodes of the same order. If the computing resource capacity of a current virtual node exceeds that of a subgraph physical node, the mapping process fails. The subgraph candidate set is then expanded by one physical node and reconstructed. This sorting and node mapping process is repeated until all virtual nodes are successfully mapped or the subgraph candidate set covers the entire elastic optical network. The virtual link mapping module is used to calculate a preset number of shortest physical links in the elastic optical network for all virtual links between mapped virtual nodes. It determines the number of consecutive spectrum slots required for the virtual link based on bandwidth requirements. If the number of consecutive spectrum slots of a shortest physical link is not less than the number of consecutive spectrum slots required for the virtual link, then the virtual link is mapped to the physical link. If any virtual link mapping fails, the subgraph candidate set is expanded by one physical node and then reconstructed. The complete process starting from node mapping is re-executed until all virtual links are successfully mapped or the subgraph candidate set covers the entire elastic optical network.

[0011] Embodiments of the present invention also provide a computer device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to execute the above-described flexible optical network virtual optical network mapping method based on carbon emission efficiency factor.

[0012] Embodiments of the present invention also provide a computer-readable storage medium storing a computer program, which, when run by a processor, is capable of executing the above-described method for mapping a flexible optical network based on a carbon emission efficiency factor.

[0013] The flexible optical network virtual optical network mapping method based on carbon emission efficiency factor provided by this invention has at least the following beneficial effects: This invention introduces a carbon emission efficiency factor as a core evaluation index, prioritizing physical nodes with low carbon emission intensity per unit of computing resources during node mapping. This optimizes the node selection strategy from the source and effectively reduces overall carbon emissions. Simultaneously, by constructing and iteratively expanding a connected subgraph composed of nodes with high carbon emission efficiency as the mapping range, the search space is significantly reduced, computational complexity is decreased, and computational efficiency is greatly improved while ensuring a feasible mapping solution. This achieves low-carbon and efficient virtual optical network mapping. Attached Figure Description

[0014] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings:

[0015] Figure 1 This is a flowchart of a flexible optical network virtual optical network mapping method based on a carbon emission efficiency factor according to an embodiment of the present invention; Figure 2 This is a schematic diagram of a UNET multi-subdomain network topology provided according to an embodiment of the present invention; Figure 3 Figure (a) shows two virtual optical networks (VON1, VON2) and Figure (b) shows an elastic optical network (EON) based on an embodiment of the present invention. Figure 4 This is a diagram of a carbon emission sensing virtual optical network architecture based on an embodiment of the present invention.

[0016] Figure 5 This is a comparison diagram of VON mapping methods provided according to an embodiment of the present invention; Figure 6 This is a performance comparison chart of the algorithm of the present invention with other heuristic algorithms (Lin-VONE, ESE-VONE, etc.) under EUNET according to an embodiment of the present invention; Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0018] One embodiment of the present invention relates to a method for mapping a virtual optical network based on a carbon emission efficiency factor for an elastic optical network. The implementation details of the method for mapping a virtual optical network based on a carbon emission efficiency factor for an elastic optical network in this embodiment are described in detail below. The following implementation details are provided for ease of understanding and are not necessary for implementing this solution.

[0019] The specific process of the elastic optical network virtual optical network mapping method based on the carbon emission efficiency factor in this embodiment can be described as follows: Figure 1 As shown, it includes: Step 101: Obtain the carbon emission factor and computing resource capacity of each physical node in the elastic optical network, as well as the number of continuous spectrum slots in each physical link; obtain the computing resource capacity of each virtual node in the virtual optical network, as well as the bandwidth requirements of each virtual link. The physical network nodes include bandwidth variable transponders (BVT), bandwidth variable wavelength cross-connectors (BV-WXC), erbium-doped fiber amplifiers (EDFA), IP routers (IPR), and servers. The power consumption parameters of each device strictly refer to Table 1. Table 1 Equipment Power Consumption Reference Table Among them, BV-WXC's Represents the underlying node The degree (number of connection links). Representative node Number of add / drop ports; maximum server power consumption Fixed energy consumption Dynamic energy consumption and computing resource utilization It exhibits a linear relationship, that is This model is a commonly used model for computing server energy consumption in this field, and it can accurately reflect the relationship between computing resource usage and energy consumption.

[0020] The simulation program runs on a 64-bit computer, and the hardware parameters adopt the conventional configuration of simulation experiments in this field, specifically: processor: Intel i5-1035G1 (1.19GHz); memory: 16GB; storage capacity: sufficient to meet the storage requirements of 5000 virtual optical networks (VON) generated data and 10 independent running results.

[0021] Programming language: Java. As the mainstream cross-platform programming language in this field, its object-oriented features and high execution efficiency can ensure the stability and computing speed of the simulation program, which meets the technical requirements of elastic optical network simulation. Virtual network generation tool: GT-ITM (Georgia Tech Internetwork Topology Models). This tool is commonly used in the field for generating network topologies. It can generate random or structured topologies (such as the European topology EUNET) that conform to the characteristics of real networks and supports custom parameters such as the number of nodes and links. Data statistical tools: 95% confidence interval analysis based on Poisson t distribution. This statistical method is a routine means in this field to verify the reliability of experimental results. It can effectively reduce the impact of single-run errors on the results and ensure the credibility of performance indicators.

[0022] This invention sets typical underlying network topologies for different application scenarios of elastic optical networks. The parameters are all based on network simulation scenarios disclosed in the art, as detailed below: The scenario is the European Network topology (EUNET), with 28 nodes and 41 links. This topology is based on the node distribution and link connection relationship of the actual European optical network, and is a common scenario for verifying the performance of medium-to-large-scale network algorithms in this field. carbon emission factors The value ranges from 88.56 to 421.66 gCO2eq / kWh, and is set based on the carbon intensity data of power generation in European countries published by the International Energy Agency (IEA). For example, countries with a high proportion of hydropower / wind power, such as Norway and Sweden, have lower carbon emission factors at corresponding nodes, while countries with a high proportion of thermal power, such as Germany and Poland, have higher carbon emission factors at corresponding nodes. VON generation parameters: 5000 VONs, 2-10 virtual nodes (covering the scale of small to medium-large VONs), computing resource requirements (0, 5] units, bandwidth requirements (0, 25] Gbps, each pair of virtual nodes is randomly connected with a 50% probability. These parameters can simulate the random generation characteristics of VONs in a real network.

[0023] The spectrum slots are 12.5GHz / slot, which is the mainstream configuration for flexible optical networks. It can flexibly match services with different bandwidth requirements (such as 25Gbps bandwidth requires 2 spectrum slots) and complies with the relevant recommendations of ITU-T (International Telecommunication Union Telecommunication Standardization Sector). The workload is measured in Erlang units, ranging from 50 to 300 Erlang, covering light (50 Erlang), medium (100, 200 Erlang), and heavy (300 Erlang) scenarios, which can comprehensively verify the performance of the algorithm under different loads. The simulation window is 100 time units / window. The performance index is calculated once in each window, and the result is the average of 10 independent runs. This setting can effectively reduce the impact of random factors on the results and conforms to the conventional design of simulation experiments in this field.

[0024] A schematic diagram of the UNET multi-subdomain network topology is shown below. Figure 2 As shown in the figure, the EUNET network topology with nodes, link parameters and sub-region division is presented. It shows the nodes (numbered 0-27) and links (labeled with distance and parameters) of the EUNET network, and divides it into three sub-regions: Sub1, Sub2 and Sub3, which intuitively reflects the network structure, resource distribution and sub-region division.

[0025] Elastic Optical Network (EON) Model Build: Based on a selected scenario (such as EUNET), the GT-ITM tool generates a topology: first, node coordinates are set (simulating real geographical locations), and then links are generated based on the distances between nodes and the connection probabilities, forming a node set. With link set For example, in the EUNET scenario, node 9 and node 5 are adjacent (based on the geographical distribution of Europe), and the link... The length is set at 160 kilometers based on the actual intercity distance; erbium-doped fiber amplifiers (EDFAs) are deployed in the link at 80-kilometer intervals (the conventional amplification distance for EDFAs in this field), and the link... Two EDFAs are deployed on top to ensure the transmission quality of optical signals.

[0026] Allocate total computing resources to each underlying node Referring to the computing resource configurations of servers in this field (such as standardized computing units for converting CPU cores and memory capacity), the computing resources for node 9 in the EUNET scenario are set to 83 units, and the remaining nodes are allocated 50-100 units; the remaining computing resources The initial value is equal to the total computing resources. When a virtual node is mapped to the underlying node, Updated to ( (This represents the computing resource requirements of virtual nodes) and dynamically reflects resource usage.

[0027] Total number of spectrum slots per link =100 (typical number of spectrum slots in large-scale flexible optical networks in this field), each spectrum slot with a bandwidth of 12.5 GHz; spectrum slot status (t is the spectrum slot index, ranging from 1 to 100), initial state A value of 1 indicates that the spectrum slot is available. This value is used when a spectrum slot is allocated to a virtual link. (Indicates spectrum slot occupancy), ensuring conflict-free allocation of spectrum resources.

[0028] bottom-level nodes carbon emission factors Calculate using the following formula: ,in: For nodes Collection of electricity sources; For source Percentage of contribution; For example, the electricity source for EUNET node 9 is "60% hydropower + 30% wind power + 10% thermal power". Hydropower: Wind power: Thermal power: ,but

[0029] This value is consistent with the carbon intensity characteristics of areas dominated by hydropower / wind power.

[0030] Virtual Optical Network (VON) Model Build: Generate virtual nodes using the GT-ITM tool: Set the number of virtual nodes (2-10) according to the scenario, and the computing resource requirements of each virtual node. Generated using a random number generator, with values ​​ranging from 0 to 5 units (simulating the computing resource requirements of different business operations, such as 1-2 units for small data processing operations and 3-5 units for large AI training operations). For example, the computing resource requirements of virtual node 0. Units need to be mapped to remaining computing resources later. The underlying nodes.

[0031] Virtual link generation rules: A link is generated between each pair of virtual nodes with a 50% probability (simulating the randomness of VON link connections in a real network), bandwidth requirements. Generated using a random number generator, with values ​​ranging from (0, 25] Gbps; The architecture diagram of a carbon emission sensing virtual optical network based on elastic optical networks is shown below. Figure 3As shown in the figure, the mapping process from Virtual Optical Network (VON) to Elastic Optical Network (EON) is illustrated: Figure (a) shows two virtual optical networks (VON1, VON2), including virtual nodes, links, and resource requirements; Figure (b) shows the Elastic Optical Network (EON), presenting the distribution of physical nodes, links, and resources; the right side shows the hardware architecture of the network nodes, including components such as optical switching, transmission, and servers; the whole figure embodies the Virtual Optical Network Embedded (VONE) technology, which maps virtual network resources to physical elastic optical networks.

[0032] Step 102: Determine the carbon emission efficiency factor by considering the relationship between the carbon emission factor of each physical node and the computational resource capacity and the carbon emission amount; Obtain the carbon emission efficiency factor (CEEF) for each physical node. i The steps of the formula are as follows: The formula for calculating the carbon emission factor of a physical node is as follows: ; In the formula, Carbon emission factors at physical nodes Let i be the set of power sources for physical node i. For power source a, a pair of physical nodes The percentage of contribution, For physical nodes Geographical location of power source The carbon emission coefficient; Example: Carbon emission factor at node 9 initial state Then its carbon emissions This value only reflects the carbon emissions generated by the server's fixed energy consumption.

[0033] The formula for calculating the carbon emissions of a physical node is as follows: ; In the formula, Let i be the fixed energy consumption of physical node i. , Let i be the maximum energy consumption of physical node i. , The computing resources allocated to physical node i for the virtual node. The total computing resources of physical node i; make and By normalizing parameters to eliminate differences between different physical nodes, the carbon emission efficiency factor (CEEF) is increased. i The formula is as follows: ; In the formula, Carbon emission factors at physical nodes Calculate the resource capacity for physical nodes. That is, the higher the CEEF value, the lower the carbon emission intensity.

[0034] Example: Total computing resources of node 9 Unit, carbon emission factor Then its CEEF value Sort the CEEF values ​​of all bottom-level nodes in descending order to form a set W (which will be used in the Sub-VONE algorithm to prioritize nodes with high CEEF).

[0035] Step 103: Construct a candidate set of subgraphs for the elastic optical network by sorting the carbon emission efficiency factors of each physical node and iterating the adjacency relationship; the candidate set of subgraphs includes physical nodes and physical links of the subgraph. The subgraph candidate set starts with the physical node with the highest carbon emission efficiency factor. Through an iterative process, the physical node with the highest carbon emission efficiency factor is selected from the unselected physical nodes that are directly connected to the current subgraph candidate set and added to the subgraph candidate set, until the number of physical nodes in the subgraph candidate set reaches the number of virtual nodes to be mapped.

[0036] Preset the number of subgraph nodes n (by default, it equals the number of virtual nodes in the VON to be mapped; for example, if the VON contains 4 virtual nodes, then n=4). Input the EON model. Initial empty set graph And the CEEF values ​​of each node combined with W.

[0037] The first step is to traverse all the bottom-level nodes if D is an empty set, and select the node with the largest CEEF value to add to D (prioritize nodes with low carbon emission intensity and sufficient computing resources, which meet the carbon emission reduction target). For example, in the EUNET scenario, node 9 has the largest CEEF value (0.0113), so it is added to D, and D = {9}.

[0038] Step 2: Iteratively expand nodes: From W, select the nodes that are adjacent to nodes in D (i.e., the bottom-level nodes that are directly connected to nodes in D but have not yet joined D), and add the node with the largest CEEF value to D. Repeat this step until |D|=n. For example, when D={9}, the adjacent nodes of node 9 include node 5 (CEEF=0.0112) and node 11 (CEEF=0.0044). Select node 5 to join D, and D={9, 5}. Continue iteratively selecting the node with the largest CEEF among the adjacent nodes of node 5 until |D|=4 (assuming the initial n=4), and finally D={9, 5, x, y} (x, y are the adjacent nodes with higher CEEF selected later).

[0039] Traverse all node pairs in the subgraph node set D If a link (i, j) exists in the underlying EON, then that link is added to the subgraph link set L. For example... If there are links (9,5), (5,x), and (x,y) at the bottom layer, then L={(9,5),(5,x),(x,y)}, ensuring the connectivity between nodes within the subgraph and satisfying the mapping requirements of the virtual links.

[0040] If in the current subgraph If no feasible solution for VON mapping is found (e.g., the computational resource requirements of the virtual node cannot be met, or the spectrum slot requirements of the virtual link cannot be met), then subgraph expansion is initiated: the preset number of subgraph nodes n is increased by 1, and steps 2-3 are repeated (selecting the node with the largest CEEF among the nodes adjacent to the nodes in D from C and adding it to D, and supplementing the corresponding link to L), until one of the following two situations occurs: Case 1: A feasible solution is found in the subgraph, stop the expansion and execute the subsequent VON mapping steps; Case 2: The subgraph expands to include all nodes of the underlying EON. If no feasible solution is found, the VON mapping is deemed to have failed, the failure result is recorded, and the next VON is processed.

[0041] This expansion mechanism balances mapping success rate and computational efficiency: incremental expansion with a step size of 1 avoids a surge in computational complexity caused by excessive expansion of the subgraph range, while ensuring coverage of all potentially feasible mapping regions. The specific algorithm is shown in Table 2.

[0042] Table 2 Specific Algorithm Step 104: Sort the virtual nodes in descending order of their computing resource capacity, and sort the subgraph physical nodes in descending order of their carbon emission efficiency factor; map the virtual nodes to the corresponding subgraph physical nodes in sequence. If the computing resource capacity of the current virtual node exceeds the computing resource capacity of the subgraph physical node, the virtual node mapping in this round is deemed to have failed; expand the subgraph candidate set by one physical node and reconstruct the subgraph candidate set, repeating the above sorting and node mapping process until all virtual nodes are successfully mapped or the subgraph candidate set covers the entire elastic optical network; Sorting of bottom-level nodes: Sort all bottom-level nodes in descending order of their CEEF values ​​to form a set. (e.g., in the EUNET scenario) To ensure that nodes with high carbon emission efficiency are prioritized; Virtual Node Sorting: Sort all virtual nodes in descending order of computing resource requirements to form a set. (e.g., V={0,1,2,3}, where virtual node 0 has the highest resource requirements), ensuring that virtual nodes with high resource requirements are matched first, thereby improving the overall mapping success rate.

[0043] Initialize the mapping flags: (Underlying node j is not mapped). (Virtual node i is not mapped). (Virtual link l is not mapped); Virtual node mapping: the sorted set of virtual nodes after traversal For each virtual node i, traverse the sorted underlying node set W and select the node that satisfies the following conditions: , The first node Virtual nodes Mapped to Update the tag , ; Virtual link mapping: Traverse all virtual links ,like, and (Both ends are mapped), call the K-shortest path algorithm (K=3) to calculate in the subgraph and Check and allocate spectrum slots between the paths, and update the tags. (Already mapped); details are as follows: 1. Virtual Link Mapping (based on K-shortest path algorithm) The goal of virtual link mapping is to find the shortest path that satisfies the spectrum slot constraint. The steps are as follows: The K-shortest path algorithm (K=3, a commonly used path number setting in this field, balancing path diversity and computational efficiency) is called to calculate the three shortest paths between the mapped nodes at both ends of the virtual link (such as virtual node 0→9, virtual node 1→5) (the path length is calculated by summing the physical length of the link). Check spectrum continuity: For each candidate path, traverse all underlying links on the path and check if there is a continuous number of spectrum slots greater than or equal to the number of spectrum slots required by the virtual link. If so, select the path. Spectrum slot allocation: Select the shortest path from the paths that meet the conditions, allocate consecutive spectrum slots (e.g., virtual link (0,1) requires 2 spectrum slots, select time slots 1-2 of link (9,5), and update the spectrum slot status of the link. ; Handling cases where no feasible path exists: If none of the three paths have a spectrum slot that meets the conditions, increase the value of K (e.g., K=5) and recalculate the path. If no feasible path still exists, the VON mapping is considered to have failed.

[0044] Mapping result judgment: If all virtual nodes and links are successfully mapped, the mapping is considered successful; otherwise, the mapping is considered unsuccessful, and this step is re-executed after subgraph expansion is initiated; the specific algorithm is shown in Table 3: Table 3 Specific Algorithm Step 105: For all virtual links between mapped virtual nodes, calculate a preset number of shortest physical links in the elastic optical network. Determine the number of consecutive spectrum slots required for the virtual link based on bandwidth requirements. If the number of consecutive spectrum slots of a shortest physical link is not less than the number of consecutive spectrum slots required for the virtual link, then map the virtual link to the physical link. If any virtual link mapping fails, expand the subgraph candidate set by one physical node and reconstruct the subgraph candidate set. Re-execute the complete process starting from node mapping until all virtual links are successfully mapped or the subgraph candidate set covers the entire elastic optical network.

[0045] The formula for calculating the required number of consecutive spectrum slots based on virtual link bandwidth requirements is as follows: ; In the formula, The required number of consecutive spectrum slots; For virtual link bandwidth requirements; represents the modulation format coefficients, where BPSK is 1, QPSK is 2, 8QAM is 3, 16QAM is 4, 64QAM is 6, and 256QAM is 8; G represents the number of guard spectrum slots, which is set to 1.

[0046] Example: Virtual Link bandwidth Select QPSK modulation format The required number of spectrum slots is: This means that the virtual link needs to occupy two consecutive spectrum slots.

[0047] If the candidate set of the subgraph has been expanded to cover the entire elastic optical network, but still cannot meet the mapping conditions of the virtual node or virtual link, the process is terminated and the mapping of the virtual optical network is determined to be a failure.

[0048] The goal of performance verification is to evaluate the effectiveness and applicability of the method of this invention. Based on commonly used performance metrics in the field, the steps are as follows: Six core indicators were selected, including acceptance rate, system benefits, energy consumption, and carbon emissions. The calculation formulas and definitions are as follows: Acceptance rate: ,in The number of VONs successfully mapped. The total number of input VONs reflects the algorithm's capacity to handle VONs; System benefits: ,in For the set of virtual nodes that have accepted VON, For virtual link sets that have accepted VON, For virtual nodes The resource yield coefficient, For virtual links The bandwidth benefit coefficient, where T is the number of time windows, reflects the overall benefit of the algorithm; Energy consumption: ,in Energy consumption of EDFA , For the fixed energy consumption of each EDFA, For link (number of EDFAs on) Energy consumption of BVT ( ), For BVT fixed energy consumption, This represents the energy consumption per unit flow rate through the BVT. For physical nodes The traffic demand of BVT A binary variable representing a physical node. If the BVT is in an active state, then =1, otherwise, =0, Represents physical nodes carbon emission coefficient, For IPR energy consumption, For BV-WXC's energy consumption, This represents the server's energy consumption, reflecting the energy consumption of the algorithm. Carbon emissions: ,in For the bottom layer node Total energy consumption (including the energy consumption of all components). For nodes The carbon emission factor reflects the environmental impact of the algorithm; Running time: The total time from VON input to mapping completion (including subgraph construction, algorithm execution, and result output), reflecting the real-time performance of the algorithm.

[0049] The architecture diagram of a carbon emission sensing virtual optical network based on elastic optical networks is shown below. Figure 4 As shown in the figure, the process of Virtual Optical Network Embedding (VONE) is illustrated. It starts with the input of the physical network and virtual request, goes through node mapping (including subgraph construction and expansion), link mapping, and finally determines whether the mapping is successful or not.

[0050] Another embodiment of the present invention relates to a flexible optical network virtual optical network mapping system based on carbon emission efficiency factor. The implementation details of the flexible optical network virtual optical network mapping system based on carbon emission efficiency factor in this embodiment are described in detail below. The following implementation details are provided for ease of understanding and are not necessary for implementing this solution.

[0051] Specifically, the data acquisition module acquires the carbon emission factor, computing resource capacity, and carbon emissions of each physical node in the elastic optical network, as well as the number of continuous spectrum slots in each physical link; and acquires the computing resource capacity of each virtual node in the virtual optical network, as well as the bandwidth requirements of each virtual link. The carbon emission efficiency factor calculation module is used to determine the carbon emission efficiency factor by the relationship between the carbon emission factor of each physical node and the computing resource capacity and the carbon emission amount. The subgraph candidate set construction module is used to construct a subgraph candidate set of the elastic optical network by sorting the carbon emission efficiency factors of each physical node and iterating the adjacency relationship; the subgraph candidate set includes subgraph physical nodes and subgraph physical links; The virtual node mapping module sorts virtual nodes in descending order of their computing resource capacity and subgraph physical nodes in descending order of their carbon emission efficiency factor. It then maps virtual nodes sequentially to subgraph physical nodes of the same order. If the computing resource capacity of a current virtual node exceeds that of a subgraph physical node, the mapping process fails. The subgraph candidate set is then expanded by one physical node and reconstructed. This sorting and node mapping process is repeated until all virtual nodes are successfully mapped or the subgraph candidate set covers the entire elastic optical network. The virtual link mapping module is used to calculate a preset number of shortest physical links in the elastic optical network for all virtual links between mapped virtual nodes. It determines the number of consecutive spectrum slots required for the virtual link based on bandwidth requirements. If the number of consecutive spectrum slots of a shortest physical link is not less than the number of consecutive spectrum slots required for the virtual link, then the virtual link is mapped to the physical link. If any virtual link mapping fails, the subgraph candidate set is expanded by one physical node and then reconstructed. The complete process starting from node mapping is re-executed until all virtual links are successfully mapped or the subgraph candidate set covers the entire elastic optical network.

[0052] A specific example: This invention underwent comprehensive experimental verification in an EUNET topology with 28 nodes and 41 links, covering various scenarios with service loads of 50 Erlang (light to medium load), 100 Erlang (medium load), 200 Erlang (medium to heavy load), and 300 Erlang (heavy load). To ensure the fairness of the comparison and the reliability of the results, the representative existing technologies NCR-VONE (a penalty-based heuristic algorithm) and Lin-VONE (a degree-based heuristic algorithm) were selected as comparison algorithms. Each experimental scenario was run independently 10 times, and the impact of random errors on the results was effectively reduced by calculating the average value and the 95% confidence interval based on the Poisson t distribution.

[0053] Experimental results demonstrate that the method of this invention exhibits superior performance under varying network sizes and workloads. Regarding network size, in elastic optical networks with 6 to 100 nodes, the virtual network request acceptance rate of the method consistently remains above 0.94, while carbon emissions are reduced by at least 12.3% and energy consumption by at least 19.3% compared to the NCR-VONE algorithm, showcasing excellent scalability and environmental friendliness. In terms of workload, within a load range of 50-300 Erlang, the running time of this method is controlled within 12,000 seconds, a reduction of over 40% compared to the Lin-VONE algorithm. This fully verifies its efficiency and real-time performance in handling large-scale network requests, meeting the dual requirements of green energy saving and high-efficiency operation in practical communication networks.

[0054] A comparison diagram of VON mapping methods is shown below. Figure 5 As shown in the figure, the topology mapping and carbon emission comparison of two schemes of Virtual Optical Network Embedding (VONE) (CEEF-based and Typical scheme) are illustrated. The figure includes the virtual optical network VON1, the two mapping topologies (CEEF-based VONE and Typical VONE), and the corresponding carbon emission calculation and comparison. The carbon reduction advantage of the CEEF method is verified by GHG1-GHG2≈-76.

[0055] The performance comparison chart of the algorithm of this invention with other heuristic algorithms (Lin-VONE, ESE-VONE, etc.) under EUNET is shown in the figure below. Figure 6 As shown in the bar chart, different network schemes are compared in terms of acceptance rate, system benefits, uptime, energy consumption, carbon emissions, spectrum cost, and resource fragmentation. The bar chart covers a comprehensive analysis of indicators from network resource utilization to cost, energy consumption, and carbon emissions.

[0056] The steps of the various methods described above are only for clarity. In practice, they can be combined into one step or some steps can be split into multiple steps. As long as they include the same logical relationship, they are all within the protection scope of this invention. Adding insignificant modifications or introducing insignificant designs to the algorithm or process, without changing the core design of the algorithm and process, are also within the protection scope of this invention.

[0057] Another embodiment of the present invention relates to a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the method embodiments described above.

[0058] That is, those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. This program is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0059] Those skilled in the art will understand that the above embodiments are specific embodiments for implementing the present invention, and in practical applications, various changes can be made to them in form and detail without departing from the spirit and scope of the present invention.

Claims

1. A method for mapping a virtual optical network in an elastic optical network based on a carbon emission efficiency factor, characterized in that, The method includes: Obtain the carbon emission factor, computing resource capacity, and carbon emission of each physical node in the elastic optical network, as well as the number of continuous spectrum slots in each physical link; obtain the computing resource capacity of each virtual node in the virtual optical network, as well as the bandwidth requirements of each virtual link. The carbon emission efficiency factor is determined by the relationship between the carbon emission factor of each physical node and the computational resource capacity and the carbon emission amount. A candidate set of subgraphs for the elastic optical network is constructed by sorting the carbon emission efficiency factors of each physical node and iterating the adjacency relationship; the candidate set of subgraphs includes subgraph physical nodes and subgraph physical links. The virtual nodes are sorted in descending order of their computing resource capacity, and the subgraph physical nodes are sorted in descending order of their carbon emission efficiency factor. The virtual nodes are mapped to the corresponding subgraph physical nodes in descending order. If the computing resource capacity of the current virtual node exceeds that of the subgraph physical node, the virtual node mapping in this round is considered a failure. The subgraph candidate set is expanded by one physical node and then reconstructed. The above sorting and node mapping process is repeated until all virtual nodes are successfully mapped or the subgraph candidate set covers the entire elastic optical network. For all virtual links between mapped virtual nodes, a preset number of shortest physical links are calculated in the elastic optical network. The number of consecutive spectrum slots required for a virtual link is determined according to bandwidth requirements. If the number of consecutive spectrum slots of a shortest physical link is not less than the number of consecutive spectrum slots required for the virtual link, then the virtual link is mapped to that physical link. If any virtual link mapping fails, the subgraph candidate set is expanded by one physical node and then reconstructed. The complete process starting from node mapping is re-executed until all virtual links are successfully mapped or the subgraph candidate set covers the entire elastic optical network.

2. The method for mapping a virtual optical network based on a carbon emission efficiency factor in an elastic optical network as described in claim 1, characterized in that, The method of constructing a candidate set of subgraphs for the elastic optical network by sorting the carbon emission efficiency factors of each physical node and iterating the adjacency relationship includes the following steps: The subgraph candidate set starts with the physical node with the highest carbon emission efficiency factor. Through an iterative process, the physical node with the highest carbon emission efficiency factor is selected from the unselected physical nodes that are directly connected to the current subgraph candidate set and added to the subgraph candidate set, until the number of physical nodes in the subgraph candidate set reaches the number of virtual nodes to be mapped.

3. The method for mapping a virtual optical network based on a carbon emission efficiency factor in an elastic optical network as described in claim 1, characterized in that, The method for determining the carbon emission efficiency factor by considering the relationship between the carbon emission factors of each physical node and the computational resource capacity and carbon emissions includes the following steps: The formula for calculating the carbon emission factor of a physical node is as follows: ； In the formula, Carbon emission factors at physical nodes Let i be the set of power sources for physical node i. For power source a, a pair of physical nodes The percentage of contribution, For physical nodes Geographical location of power source The carbon emission coefficient; The formula for calculating the carbon emissions of a physical node is as follows: ； In the formula, Let i be the fixed energy consumption of physical node i. , Let i be the maximum energy consumption of physical node i. , The computing resources allocated to physical node i for the virtual node. The total computing resources of physical node i; make and By normalizing parameters to eliminate differences between different physical nodes, the carbon emission efficiency factor (CEEF) is increased. i The formula is as follows: ； In the formula, Carbon emission factors at physical nodes Calculate the resource capacity for physical nodes.

4. The method for mapping a virtual optical network based on a carbon emission efficiency factor in an elastic optical network as described in claim 1, characterized in that, The formula for calculating the number of consecutive spectrum slots required for a virtual link based on bandwidth demand is as follows: ； In the formula, The required number of consecutive spectrum slots; For virtual link bandwidth requirements; represents the modulation format coefficients, where BPSK is 1, QPSK is 2, 8QAM is 3, 16QAM is 4, 64QAM is 6, and 256QAM is 8; G represents the number of guard spectrum slots, which is set to 1.

5. A flexible optical network virtual optical network mapping system based on carbon emission efficiency factor, characterized in that, The system includes: The data acquisition module acquires the carbon emission factor, computing resource capacity, and carbon emissions of each physical node in the elastic optical network, as well as the number of continuous spectrum slots in each physical link; it also acquires the computing resource capacity of each virtual node in the virtual optical network and the bandwidth requirements of each virtual link. The carbon emission efficiency factor calculation module is used to determine the carbon emission efficiency factor by the relationship between the carbon emission factor of each physical node and the computing resource capacity and the carbon emission amount. The subgraph candidate set construction module is used to construct a subgraph candidate set of the elastic optical network by sorting the carbon emission efficiency factors of each physical node and iterating the adjacency relationship; the subgraph candidate set includes subgraph physical nodes and subgraph physical links; The virtual node mapping module sorts virtual nodes in descending order of their computing resource capacity and subgraph physical nodes in descending order of their carbon emission efficiency factor. It then maps virtual nodes sequentially to subgraph physical nodes of the same order. If the computing resource capacity of a current virtual node exceeds that of a subgraph physical node, the mapping process fails. The subgraph candidate set is then expanded by one physical node and reconstructed. This sorting and node mapping process is repeated until all virtual nodes are successfully mapped or the subgraph candidate set covers the entire elastic optical network. The virtual link mapping module is used to calculate a preset number of shortest physical links in the elastic optical network for all virtual links between mapped virtual nodes. It determines the number of consecutive spectrum slots required for the virtual link based on bandwidth requirements. If the number of consecutive spectrum slots of a shortest physical link is not less than the number of consecutive spectrum slots required for the virtual link, then the virtual link is mapped to the physical link. If any virtual link mapping fails, the subgraph candidate set is expanded by one physical node and then reconstructed. The complete process starting from node mapping is re-executed until all virtual links are successfully mapped or the subgraph candidate set covers the entire elastic optical network.

6. A computer system, characterized in that, include: At least one processor; And a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the flexible optical network virtual optical network mapping method based on carbon emission efficiency factor as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, is capable of performing the flexible optical network virtual optical network mapping method based on carbon emission efficiency factor as defined in any one of claims 1 to 4.

Images