An efficient dynamic bandwidth allocation method and system based on general SDN switches

Abstract

The application provides a high-efficiency dynamic bandwidth allocation method and system based on a general SDN switch, and the steps of the method include: taking each two logical groups as a candidate pair, determining the data throughput of the candidate pair based on historical traffic data, and constructing a traffic matrix; determining the link load rate of each candidate pair based on the traffic matrix and the topology bandwidth matrix of the current software-defined network, determining the hotspot link and the non-hotspot link from the candidate pair based on the link load rate; determining the number of physical links to be added based on the link load rate of the hotspot link and the preset target load rate; calculating the predicted load rate of the non-hotspot link after releasing the physical link, determining the releasable redundancy and the release safety margin of the non-hotspot link based on the predicted load rate; determining the candidate bandwidth migration unit based on the bandwidth gap, the releasable redundancy and the release safety margin, and applying the corresponding topology update of the candidate bandwidth migration unit.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and in particular to an efficient dynamic bandwidth allocation method and system based on a general-purpose SDN switch. Background Technology

[0002] With the development of cloud computing, big data, artificial intelligence training clusters, and high-performance computing services, east-west traffic in data center networks continues to increase, and a large amount of service traffic exhibits sudden and uneven characteristics. For software-defined networks that include optical switches and optical distribution frames, fiber optic links and port resources typically have high deployment costs. If physical bandwidth is configured statically for a long time, it will cause optical links to idle when some logical links are under low load, and it will be difficult to obtain available bandwidth in a timely manner when some logical links experience hot traffic.

[0003] Existing control planes based on software-defined networks can separate control from forwarding and collect link utilization, port status, and flow table status through controllers. However, in scenarios involving physical link reconfiguration, traditional solutions often employ static configuration, round-robin scheduling, or simple threshold triggering strategies. These tend to focus only on whether the logical topology is connected, failing to integrate real-time traffic matrices, the number of physical links, and link bandwidth capacity into the same decision-making process, making it difficult to guarantee the timeliness and effectiveness of decisions. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide an efficient dynamic bandwidth allocation method based on a general-purpose SDN switch to eliminate or improve one or more defects existing in the prior art.

[0005] One aspect of the present invention provides an efficient dynamic bandwidth allocation method based on a general-purpose SDN switch, the method being applied to a software-defined network comprising multiple logical packets, the method comprising the following steps: Each pair of logical groups is considered as a candidate pair. The data throughput of the candidate pair is determined based on the historical traffic data of each logical group. A traffic matrix is ​​constructed based on the data throughput of all the candidate pairs. The link load rate of each candidate pair is determined based on the traffic matrix and the topology bandwidth matrix of the current software-defined network, and hot links and non-hot links are determined from the candidate pairs based on the link load rate. Based on the link load rate and preset target load rate of the hot link, determine the number of new physical links to be added for the hot link; Calculate the predicted load rate of non-hotspot links after releasing physical links. The predicted load rate includes a first predicted load rate and a second predicted load rate. Determine the releaseable links among the non-hotspot links based on the first predicted load rate. Determine the maximum number of releaseable physical links based on the second predicted load rate. Based on the maximum number of releaseable physical links, physical links in the releaseable links will be migrated to hot links, so that the number of added physical links in the hot links is equal to the number of new physical links to be added.

[0006] The above scheme converts the cumulative number of received bytes in adjacent control cycles into instantaneous throughput, and then generates a traffic matrix for subsequent judgment through smoothing processing. This prevents sudden noise from directly driving physical link reconfiguration. Subsequently, the controller combines the traffic matrix with the topology bandwidth matrix, calculates the load rate according to the correspondence between traffic demand and the capacity of activated physical links, and identifies hot links. Furthermore, this scheme does not directly perform bandwidth replacement based on the single gain difference between hot links and several links. Instead, it first converts hot links into bandwidth gaps, and then converts non-hot links into releasable redundancy and release safety margins. This ensures that the candidates for bandwidth migration have a correspondence between demand-side gaps and supply-side redundancy. Finally, the topology update scheme is determined by comparing the predicted load rate matrix of the corresponding topology update method. This scheme integrates the real-time traffic matrix, the number of physical links, and the link bandwidth capacity into the same decision-making process, ensuring the real-time nature and effectiveness of the decision.

[0007] In some embodiments of the present invention, in the step of determining the data throughput of candidate pairs based on the historical traffic data of each logical group, and constructing a traffic matrix based on the data throughput of all candidate pairs, the data throughput of each candidate pair is combined to obtain the constructed traffic matrix, using the following formula for the data throughput of the candidate pairs: in, Indicates the first Within the first control cycle, the first The logical group and the first Data throughput between logical groups; Indicates the first The data collected during the first control cycle is the first... The logical group and the first The cumulative number of bytes exchanged between logical groups; Indicates the first The data collected during the first control cycle is the first... The logical group and the first The cumulative number of bytes exchanged between logical groups; Indicates the duration of a single control cycle.

[0008] In some embodiments of the present invention, in the step of combining the data throughput of each candidate pair to obtain the traffic matrix, the data throughput is smoothed by an exponentially weighted moving average method, and the smoothed data throughput is combined to construct the traffic matrix.

[0009] In some embodiments of the present invention, the link load rate of each candidate pair is determined based on the traffic matrix and the topology bandwidth matrix of the current software-defined network, and the link load rate is calculated using the following formula: in, Indicates the first The logical group and the first Load rate of logical links between logical groups; Represents the first in the topology bandwidth matrix The logical group and the first The number of active physical links between logical packets; This represents the baseline capacity of a single physical link. Represents the first in the flow matrix The logical group and the first Data throughput between logical groups.

[0010] In some embodiments of the present invention, in the step of determining hot links and non-hot links from the candidate pairs based on the link load rate, the candidate pair with the highest link load rate is regarded as a hot link, and the remaining candidate pairs are regarded as non-hot links.

[0011] In some embodiments of the present invention, in the step of determining the number of new physical links to be added for the hotspot links based on the link load rate and the preset target load rate, the number of new physical links to be added is calculated using the following formula: in, This indicates the bandwidth shortage of hotspot links; This indicates the number of activated physical links corresponding to hotspot links in the topology bandwidth matrix; Indicates the first link in the hotspot link The logical group and the first Load rate of logical links between logical groups; Indicates the preset target load rate; Indicates the first in the hot link The logical group and the first The number of new physical links to be added for each logical group. This indicates rounding up to the nearest integer.

[0012] In some embodiments of the present invention, in the step of calculating the predicted load rate of a non-hotspot link after the release of the physical link, wherein the predicted load rate includes a first predicted load rate and a second predicted load rate, the first predicted load rate and the second predicted load rate are calculated using the following formula: in, Indicates the first predicted load factor or the second predicted load factor; The traffic matrix representing non-hotspot links is shown in the figure. The logical group and the first Data throughput between logical groups; This indicates the number of activated physical links corresponding to non-hotspot links in the topology bandwidth matrix; This represents the number of physical links expected to be released. It is 1 when calculating the first predicted load rate and the number of physical links to be released when calculating the second predicted load rate.

[0013] In some embodiments of the present invention, in the step of determining the releasable link among the non-hotspot links based on the first predicted load rate, for any non-hotspot link, the first predicted load rate is calculated based on the link capacity after releasing a physical link; if the first predicted load rate is less than or equal to a preset safe load rate, then the non-hotspot link is determined as a releasable link. In the step of determining the maximum number of releaseable physical links based on the second predicted load rate, a release safety margin is calculated based on the second predicted load rate, and the maximum number of releaseable physical links for non-hotspot links is determined based on the release safety margin.

[0014] In some embodiments of the present invention, in the step of calculating the release safety margin based on the second predicted load rate and determining the maximum number of non-hotspot links to be released based on the release safety margin, the release safety margin is calculated using the following formula: in, Indicates non-hotspot logical links release The safety margin for releasing physical links; Indicates the preset safe load rate; Indicates non-hotspot logical links release The second predicted load rate after the physical link is released; This indicates the number of activated physical links corresponding to non-hotspot links in the topology bandwidth matrix; This represents the release intensity penalty coefficient. ≥0.

[0015] In some embodiments of the present invention, in the steps of calculating the release safety margin based on the second predicted load rate and determining the maximum number of physical links to be released for non-hotspot links based on the release safety margin, the following will be satisfied: ≥ The maximum number of physical links to be released under the given conditions is used as the maximum number of physical links to be released for non-hotspot links. This indicates the preset safety margin threshold.

[0016] In the specific implementation process The value range is 0.05 to 0.1, preferably 0.7.

[0017] In some embodiments of the present invention, the method further includes updating the topology bandwidth matrix of the software-defined network based on the migration scheme.

[0018] A second aspect of the present invention also provides a high-efficiency dynamic bandwidth allocation system based on a general-purpose SDN switch. The system includes a computer device, which includes a processor and a memory. The memory stores computer instructions, and the processor executes the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps of the method described above.

[0019] A third aspect of the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the aforementioned method for predicting multimodal anomalies in a logistics supply chain based on a hybrid expert model.

[0020] Additional advantages, objects, and features of the invention will be set forth in part in the description which follows, and will also become apparent in part to those skilled in the art upon studying the text, or may be learned by practice of the invention. The objects and other advantages of the invention will become apparent from the description and the accompanying drawings.

[0021] Those skilled in the art will understand that the objectives and advantages achievable with the present invention are not limited to those specifically described above, and that the above and other objectives achievable with the present invention will become clearer from the following detailed description. Attached Figure Description

[0022] The accompanying drawings, which are provided to further illustrate the invention and form part of this application, are not intended to limit the scope of the invention.

[0023] Figure 1 This is a schematic diagram illustrating one implementation of the present solution; Figure 2 This is a schematic diagram illustrating one implementation of the present solution; Figure 3 This is a schematic diagram of the overall processing architecture of this solution. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the embodiments and accompanying drawings. Here, the illustrative embodiments and descriptions of this invention are used to explain the invention, but are not intended to limit the invention.

[0025] It should also be noted that, in order to avoid obscuring the invention with unnecessary details, only the structures and / or processing steps closely related to the solution according to the invention are shown in the accompanying drawings, while other details that are not closely related to the invention are omitted.

[0026] Introduction to existing technologies: Technical solution of existing technology 1 In data center optical network architectures, physical layer scheduling technology based on OCS and reconfigurable optical distribution frames (OPFPS) is the mainstream dynamic bandwidth implementation solution. This technology relies on optical switch arrays at the physical layer to dynamically reconfigure the physical topology by arbitrarily connecting input fiber optic links to output fiber optic links. Its core control plane logic typically uses mathematical optimization models to calculate the optimal bandwidth allocation value for logical links, subsequently driving the OCS device to reconnect physical ports and adjust patch cords on the distribution frame, thereby expanding or shrinking the link bandwidth.

[0027] Disadvantages of existing technology 1 Although OCS-based solutions offer flexibility at the physical layer, they have the following significant drawbacks when applied to scenarios involving optical switches and patch panels: existing OCS control logic typically separates bandwidth allocation from routing planning, and the optimization model only focuses on logical bandwidth values, ignoring the actual load balancing of physical links, resulting in suboptimal global resource utilization; in large-scale topology environments, solving the mathematical model is time-consuming and may not be worthwhile.

[0028] Technical solution of existing technology 2 This solution collects network status information, including link utilization and port status, through an SDN controller. Then, based on preset policies or simple load balancing algorithms, it manages network traffic by issuing flow table rules. Common implementation methods include static QoS configuration based on the OpenFlow protocol and link aggregation technology.

[0029] Disadvantages of existing technology 2 Traditional bandwidth management solutions in SDN environments have the following shortcomings: First, static QoS configurations cannot dynamically adjust bandwidth allocation based on real-time traffic demands; second, while link aggregation technology provides some load balancing capabilities, it lacks a global resource optimization perspective and cannot accurately allocate bandwidth for specific traffic needs; third, these solutions typically do not consider the overall network bandwidth budget constraints, making it difficult to achieve optimal resource allocation under limited resource conditions; finally, traffic routing in traditional solutions is usually based on simple hash algorithms or shortest path algorithms, which cannot be optimized in conjunction with bandwidth allocation, resulting in low network resource utilization.

[0030] like Figure 1 and 3 As shown, this invention proposes an efficient dynamic bandwidth allocation method based on a general-purpose SDN switch. The method is applied to a software-defined network (SDN) comprising multiple logical packets. The steps of the method include: Step S100: Take every two logical groups as a candidate pair, determine the data throughput of the candidate pair based on the historical traffic data of each logical group, and construct a traffic matrix based on the data throughput of all the candidate pairs; In practice, each logical group can be a computing device or a network composed of multiple computing devices; the computing device can be a computer or a processor, etc.

[0031] Specifically, nodes within each logical group share the same set of optical switch ports. The physical link mapping table stores the physical network device pointers, interface information, and current rates between each pair of logical groups. The historical byte count is used to record the cumulative number of bytes received and transmitted between adjacent control cycles to obtain the data throughput of the candidate pair.

[0032] Step S200: Determine the link load rate of each candidate pair based on the traffic matrix and the topology bandwidth matrix of the current software-defined network, and determine hot links and non-hot links from the candidate pairs based on the link load rate. In the specific implementation process, the links between the candidate pairs with the highest link load rate are designated as hot links.

[0033] Step S300: Based on the link load rate of the hotspot link and the preset target load rate, determine the number of physical links to be added to the hotspot link; In practice, the number of new physical links is used to ensure the normal operation of hotspot links.

[0034] Step S400: Calculate the predicted load rate of non-hotspot links after releasing physical links. The predicted load rate includes a first predicted load rate and a second predicted load rate. Determine the releaseable links among the non-hotspot links based on the first predicted load rate. Determine the maximum number of releaseable physical links based on the second predicted load rate. In the specific implementation process, the step of calculating the predicted load rate of non-hotspot links after releasing physical links further includes pre-screening the non-hotspot links. The pre-screening step includes selecting candidate pairs in which one end of the candidate pair is a logical group corresponding to a hotspot link and participating in the calculation of the predicted load rate.

[0035] Using the above scheme, since one end of the final releasable link is the same as one end of the hot link, it is not necessary to occupy the interface of one end of the hot link separately during the link migration process.

[0036] Step S500: Based on the maximum number of releaseable physical links, migrate the physical links in the releaseable links to the hot links, so that the number of added physical links in the hot links is equal to the number of physical links to be added.

[0037] The above scheme converts the cumulative number of received bytes in adjacent control cycles into instantaneous throughput, and then generates a traffic matrix for subsequent judgment through smoothing processing. This prevents sudden noise from directly driving physical link reconfiguration. Subsequently, the controller combines the traffic matrix with the topology bandwidth matrix, calculates the load rate according to the correspondence between traffic demand and the capacity of activated physical links, and identifies hot links. Furthermore, this scheme does not directly perform bandwidth replacement based on the single gain difference between hot links and several links. Instead, it first converts hot links into bandwidth gaps, and then converts non-hot links into releasable redundancy and release safety margins. This ensures that the candidates for bandwidth migration have a correspondence between demand-side gaps and supply-side redundancy. Finally, the topology update scheme is determined by comparing the predicted load rate matrix of the corresponding topology update method. This scheme integrates the real-time traffic matrix, the number of physical links, and the link bandwidth capacity into the same decision-making process, ensuring the real-time nature and effectiveness of the decision.

[0038] In some embodiments of the present invention, in the step of determining the data throughput of candidate pairs based on the historical traffic data of each logical group, and constructing a traffic matrix based on the data throughput of all candidate pairs, the data throughput of each candidate pair is combined to obtain the constructed traffic matrix, using the following formula for the data throughput of the candidate pairs: in, Indicates the first Within the first control cycle, the first The logical group and the first Data throughput between logical groups; Indicates the first The data collected during the first control cycle is the first... The logical group and the first The cumulative number of bytes exchanged between logical groups; Indicates the first The data collected during the first control cycle is the first... The logical group and the first The cumulative number of bytes exchanged between logical groups; Indicates the duration of a single control cycle.

[0039] Specifically, the first Each control cycle can be the current control cycle.

[0040] In some embodiments of the present invention, in the step of combining the data throughput of each candidate pair to obtain the traffic matrix, the data throughput is smoothed by an exponentially weighted moving average method, and the smoothed data throughput is combined to construct the traffic matrix, wherein each value in the traffic matrix corresponds to a candidate pair.

[0041] In some embodiments of the present invention, the link load rate of each candidate pair is determined based on the traffic matrix and the topology bandwidth matrix of the current software-defined network, and the link load rate is calculated using the following formula: in, Indicates the first The logical group and the first Load rate of logical links between logical groups; Represents the first in the topology bandwidth matrix The logical group and the first The number of active physical links between logical packets; This represents the baseline capacity of a single physical link. Represents the first in the flow matrix The logical group and the first Data throughput between logical groups.

[0042] In some embodiments of the present invention, in the step of determining hot links and non-hot links from the candidate pairs based on the link load rate, the candidate pair with the highest link load rate is regarded as a hot link, and the remaining candidate pairs are regarded as non-hot links.

[0043] In some embodiments of the present invention, in the step of determining the number of new physical links to be added for the hotspot links based on the link load rate and the preset target load rate, the number of new physical links to be added is calculated using the following formula: in, This indicates the bandwidth shortage of hotspot links; This indicates the number of activated physical links corresponding to hotspot links in the topology bandwidth matrix; Indicates the first in the hot link The logical group and the first Load rate of logical links between logical groups; Indicates the preset target load rate; Indicates the first in the hot link The logical group and the first The number of new physical links to be added for each logical group. This indicates rounding up to the nearest integer.

[0044] In some embodiments of the present invention, in the step of calculating the predicted load rate of a non-hotspot link after the release of the physical link, wherein the predicted load rate includes a first predicted load rate and a second predicted load rate, the first predicted load rate and the second predicted load rate are calculated using the following formula: in, Indicates the first predicted load factor or the second predicted load factor; The traffic matrix representing non-hotspot links is shown in the figure. The logical group and the first Data throughput between logical groups; This indicates the number of activated physical links corresponding to non-hotspot links in the topology bandwidth matrix; This indicates the number of physical links expected to be released. The number of physical links expected to be released is 1 when calculating the first predicted load rate, and is the number of physical links to be released when calculating the second predicted load rate.

[0045] In some embodiments of the present invention, in the step of determining the releasable link among the non-hotspot links based on the first predicted load rate, for any non-hotspot link, the first predicted load rate is calculated based on the link capacity after releasing a physical link; if the first predicted load rate is less than or equal to a preset safe load rate, then the non-hotspot link is determined as a releasable link. In the step of determining the maximum number of releaseable physical links based on the second predicted load rate, a release safety margin is calculated based on the second predicted load rate, and the maximum number of releaseable physical links for non-hotspot links is determined based on the release safety margin.

[0046] In some embodiments of the present invention, in the step of calculating the release safety margin based on the second predicted load rate and determining the maximum number of non-hotspot links to be released based on the release safety margin, the release safety margin is calculated using the following formula: in, Indicates non-hotspot logical links release The safety margin for releasing physical links; Indicates the preset safe load rate; Indicates non-hotspot logical links release The second predicted load rate after the physical link is released; This indicates the number of activated physical links corresponding to non-hotspot links in the topology bandwidth matrix; This represents the release intensity penalty coefficient. ≥0.

[0047] The above-described scheme calculates the release safety margin of non-hotspot links under different candidate release numbers and determines the maximum number of candidate releases that meets the release safety margin threshold as the maximum number of physical links to be released. This scheme can predict and constrain the post-release carrying status of non-hotspot links before hotspot link expansion, so that physical link release no longer depends solely on the current low load state, but also considers the predicted load rate and remaining link carrying capacity after release. Therefore, on the one hand, it can recover redundant physical bandwidth resources in non-hotspot links to compensate for the bandwidth gap in hotspot links as much as possible; on the other hand, it can avoid the formation of new congested links due to over-release of non-hotspot links, thereby improving resource utilization, reconfiguration stability, and topology adjustment security during dynamic bandwidth allocation.

[0048] In some embodiments of the present invention, in the steps of calculating the release safety margin based on the second predicted load rate and determining the maximum number of physical links to be released for non-hotspot links based on the release safety margin, the following will be satisfied: ≥ The maximum number of physical links to be released under the given conditions is used as the maximum number of physical links to be released for non-hotspot links. This indicates the preset safety margin threshold.

[0049] In the specific implementation process The value range is 0.05 to 0.1, preferably 0.7.

[0050] like Figure 2As shown, in some embodiments of the present invention, the method further includes step S600, updating the topology bandwidth matrix of the software-defined network based on the migration scheme.

[0051] In the specific implementation process, the migration scheme is a randomly generated migration scheme that satisfies the maximum number of releaseable physical links and makes the number of added physical links in the hot links equal to the number of new physical links to be added.

[0052] The above approach can quickly meet the transmission requirements of hot links and ensure the rapid generation of migration solutions.

[0053] This invention also provides a high-efficiency dynamic bandwidth allocation system based on a general-purpose SDN switch. The system includes a computer device, which includes a processor and a memory. The memory stores computer instructions, and the processor executes the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps of the method described above.

[0054] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the aforementioned multimodal anomaly prediction method for logistics supply chains based on hybrid expert models. The computer-readable storage medium can be a tangible storage medium, such as random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, floppy disks, hard disks, removable storage disks, CD-ROMs, or any other form of storage medium known in the art.

[0055] Those skilled in the art will understand that the exemplary components, systems, and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software, or a combination of both. Whether implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this invention. When implemented in hardware, it can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this invention are programs or code segments used to perform the desired tasks. The programs or code segments can be stored in a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried in a carrier wave.

[0056] It should be clarified that the present invention is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of the present invention.

[0057] In this invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with or in place of features of other embodiments.

[0058] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, various modifications and variations of the embodiments of the present invention are possible. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A highly efficient dynamic bandwidth allocation method based on a general-purpose SDN switch, characterized in that, The method is applied to a software-defined network comprising multiple logical groups, and the steps of the method include: Each pair of logical groups is considered as a candidate pair. The data throughput of the candidate pair is determined based on the historical traffic data of each logical group. A traffic matrix is ​​constructed based on the data throughput of all the candidate pairs. The link load rate of each candidate pair is determined based on the traffic matrix and the topology bandwidth matrix of the current software-defined network, and hot links and non-hot links are determined from the candidate pairs based on the link load rate. Based on the link load rate and preset target load rate of the hot link, determine the number of new physical links to be added for the hot link; Calculate the predicted load rate of non-hotspot links after releasing physical links. The predicted load rate includes a first predicted load rate and a second predicted load rate. Determine the releaseable links among the non-hotspot links based on the first predicted load rate. Determine the maximum number of releaseable physical links based on the second predicted load rate. A migration scheme is determined based on the maximum number of releaseable physical links and the number of new physical links to be added for each releaseable link. Based on the migration scheme, physical links in the releaseable links are migrated to hot links so that the number of new physical links in the hot links is equal to the number of new physical links to be added.

2. The efficient dynamic bandwidth allocation method based on a general-purpose SDN switch according to claim 1, characterized in that, In the steps of determining the data throughput of candidate pairs based on historical traffic data of each logical group, and constructing a traffic matrix based on the data throughput of all candidate pairs, the data throughput of each candidate pair is combined to obtain the constructed traffic matrix, using the following formula for the data throughput of candidate pairs: in, Indicates the first Within the first control cycle, the first The logical group and the first Data throughput between logical groups; Indicates the first The data collected during the first control cycle is the first... The logical group and the first The cumulative number of bytes exchanged between logical groups; Indicates the first The data collected during the first control cycle is the first... The logical group and the first The cumulative number of bytes exchanged between logical groups; Indicates the duration of a single control cycle.

3. The efficient dynamic bandwidth allocation method based on a general-purpose SDN switch according to claim 1, characterized in that, In the step of determining the link load rate for each candidate pair based on the traffic matrix and the current software-defined network topology bandwidth matrix, the link load rate is calculated using the following formula: in, Indicates the first The logical group and the first Load rate of logical links between logical groups; Represents the first in the topology bandwidth matrix The logical group and the first The number of active physical links between logical packets; This represents the baseline capacity of a single physical link. Represents the first in the flow matrix The logical group and the first Data throughput between logical groups.

4. The efficient dynamic bandwidth allocation method based on a general-purpose SDN switch according to claim 1, characterized in that, In the step of determining the number of new physical links to be added based on the link load rate and the preset target load rate of the hotspot links, the number of new physical links to be added is calculated using the following formula: in, This indicates the bandwidth shortage of hotspot links; This indicates the number of activated physical links corresponding to hotspot links in the topology bandwidth matrix; Indicates the first link in the hotspot link The logical group and the first Load rate of logical links between logical groups; Indicates the preset target load rate; Indicates the first link in the hotspot link The logical group and the first The number of new physical links to be added for each logical group. This indicates rounding up to the nearest integer.

5. The efficient dynamic bandwidth allocation method based on a general-purpose SDN switch according to claim 1, characterized in that, In the step of calculating the predicted load rate of non-hotspot links after the release of physical links, wherein the predicted load rate includes a first predicted load rate and a second predicted load rate, the first predicted load rate and the second predicted load rate are calculated using the following formula: in, Indicates the first predicted load factor or the second predicted load factor; The traffic matrix representing non-hotspot links is shown in the figure. The logical group and the first Data throughput between logical groups; This indicates the number of activated physical links corresponding to non-hotspot links in the topology bandwidth matrix; This represents the number of physical links expected to be released. It is 1 when calculating the first predicted load rate and the number of physical links to be released when calculating the second predicted load rate.

6. The efficient dynamic bandwidth allocation method based on a general-purpose SDN switch according to any one of claims 1 to 5, characterized in that, In the step of determining the releasable links among the non-hotspot links based on the first predicted load rate, for any non-hotspot link, the first predicted load rate is calculated based on the link capacity after releasing a physical link; if the first predicted load rate is less than or equal to the preset safe load rate, then the non-hotspot link is determined as a releasable link. In the step of determining the maximum number of releaseable physical links based on the second predicted load rate, a release safety margin is calculated based on the second predicted load rate, and the maximum number of releaseable physical links for non-hotspot links is determined based on the release safety margin.

7. The efficient dynamic bandwidth allocation method based on a general-purpose SDN switch according to claim 6, characterized in that, In the steps of calculating the release safety margin based on the second predicted load rate and determining the maximum number of physical links to be released for non-hotspot links based on the release safety margin, the release safety margin is calculated using the following formula: in, Indicates non-hotspot logical links release The safety margin for releasing physical links; Indicates the preset safe load rate; Indicates non-hotspot logical links release The second predicted load rate after the physical link is released; This indicates the number of activated physical links corresponding to non-hotspot links in the topology bandwidth matrix; This represents the release intensity penalty coefficient. ≥0.

8. The efficient dynamic bandwidth allocation method based on a general-purpose SDN switch according to claim 7, characterized in that, In the steps of calculating the release safety margin based on the second predicted load rate and determining the maximum number of physical links to release for non-hotspot links based on the release safety margin, the following will be satisfied: ≥ The maximum number of physical links to be released under the given conditions is used as the maximum number of physical links to be released for non-hotspot links. This indicates the preset safety margin threshold.

9. The efficient dynamic bandwidth allocation method based on a general-purpose SDN switch according to claim 1, characterized in that, The method further includes updating the topology bandwidth matrix of the software-defined network based on the migration scheme.

10. A high-efficiency dynamic bandwidth allocation system based on a general-purpose SDN switch, characterized in that, The system includes a computer device, which includes a processor and a memory. The memory stores computer instructions, and the processor executes the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps of the method as described in any one of claims 1 to 9.

Images