Data flow scheduling method, system, electronic device and storage medium for aggregated link

By using controllers in the SDN architecture to detect link load and available bandwidth in real time, calculate the probability of being selected, and automatically allocate data flow, the problem of aggregated link selection is solved, and rapid deployment and cross-vendor-compatible data is achieved Stream schedule.

CN115604190BActive Publication Date: 2024-11-15CHINA CONSTRUCTION BANK
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202211214109.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-09-30
Publication Date
2024-11-15
Estimated Expiration
2042-09-30

AI Technical Summary

Technical Problem

The data flow scheduling method of existing aggregation links does not take into account link load, and may choose to links with higher loads, resulting in link blockage, and requires manual configuration and difficult to compatible with equipment from different manufacturers.

Method used

By using controllers in the SDN architecture, the remaining available bandwidth and link load of the link are detected in real time, the probability of being selected for each link is calculated, the target link is determined and the data flow is allocated, and the data flow on the faulty link is automatically transferred. , avoid choosing high-load links and realize automated scheduling.

Benefits of technology

It effectively avoids link blockage, achieves rapid deployment and cross-vendor equipment compatibility, and improves the efficiency and automation of data flow scheduling.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115604190B_ABST
    Figure CN115604190B_ABST
Patent Text Reader

Abstract

The present application provides a data flow scheduling method, system, electronic device and storage medium for an aggregated link, which are applied to a controller in an SDN architecture. When a data flow to be allocated is detected, the remaining available bandwidth of each link in the aggregated link is obtained, and the selection probability of each link is calculated according to the remaining available bandwidth of each link; for each link, the cumulative selection probability of the link is calculated according to the selection probability of the link and the selection probabilities of each other link; according to the cumulative selection probability of each link, a target link to be allocated is determined from each link, and the data flow to be allocated is allocated to the target link to be allocated; whether there is a faulty link in each link of the aggregated link is detected in real time; when a faulty link is detected in each link of the aggregated link, the target available link is screened out from each available link, and the data flow on the faulty link is forwarded to the target available link.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and more specifically, to a data flow scheduling method, system, electronic device and storage medium for an aggregated link. Background Art

[0002] With the continuous development of computer technology, multiple links of the same pair of network devices can be aggregated into one logical link to generate an aggregated link. When there is a new data flow, the data flow can be assigned to a link in the aggregated link, or when one of the links fails, the data flow on the failed link can be transferred to other links for transmission.

[0003] The existing data flow scheduling method of the aggregated link selects a suitable link for the data flow by adopting strategies such as hashing, polling, and randomization; wherein the data flow may be a new data flow or a data flow on a failed link.

[0004] However, the existing data flow scheduling method of the aggregated link does not take into account the link load of the currently selected link, and may select a link with a higher load, which will cause link congestion; it also requires manual configuration, cannot be deployed quickly, and there is a problem of incompatibility between devices from different manufacturers. Summary of the invention

[0005] In view of this, the present invention provides a data flow scheduling method, system, electronic device and storage medium for an aggregated link to solve the problem that the prior art does not take the link load of the currently selected link into consideration, and may select a link with a higher load, which will cause link congestion; and also requires manual configuration, cannot be deployed quickly, and has the problem of incompatibility between devices from different manufacturers.

[0006] A first aspect of the present invention discloses a data flow scheduling method for an aggregated link, which is applied to a controller in an SDN architecture. The method includes:

[0007] When a data stream to be allocated is detected, the remaining available bandwidth of each link in the aggregated link is obtained, and the probability of each link being selected is calculated according to the remaining available bandwidth of each link;

[0008] For each of the links, calculating the cumulative probability of being selected of the link according to the probability of being selected of the link and the probabilities of being selected of each of the other links;

[0009] Determine a target link to be allocated from each of the links according to the accumulated probability of being selected of each of the links, and allocate the data flow to be allocated to the target link to be allocated;

[0010] Detect in real time whether there is a faulty link in each of the links of the aggregated link;

[0011] When a faulty link is detected in each link of the aggregated link, a target available link is selected from each available link, and the data stream on the faulty link is forwarded to the target available link; wherein the available link is a link in the aggregated link that has not failed.

[0012] Optionally, if the faulty link is a congested link, when it is detected that a faulty link exists in each of the links of the aggregated link, a target available link is selected from each of the available links, and a data flow on the faulty link is forwarded to the target available link, including:

[0013] When it is detected that there is a faulty link in each of the links of the aggregated link, the proportion of sampled data packets of each data flow on the faulty link is obtained;

[0014] According to the proportion of sampled data packets of each of the faulty links, at least one candidate elephant flow is selected from each of the faulty links;

[0015] For each of the candidate elephant flows, obtain the preset flow table entry information corresponding to the candidate elephant flow, and determine the transmission rate of the candidate elephant flow according to the flow table entry information;

[0016] For each of the candidate elephant flows, if the transmission rate of the candidate elephant flow is greater than a preset transmission rate, determining the candidate elephant flow as an elephant flow;

[0017] A target available link is selected from each available link, and the data flow on the faulty link is forwarded to the target available link; wherein the target available link is the available link with the largest remaining available bandwidth among each available link on the aggregated link.

[0018] Optionally, the real-time detection of whether there is a faulty link in each link of the aggregated link includes:

[0019] For each link on the aggregated link, detecting the link load of the link in real time;

[0020] Determining whether the link load of the link is greater than a preset link load;

[0021] If the link load of the link is greater than the preset link load, determining that the link is a faulty link, and the faulty link is a congested link;

[0022] If the link load of the link is not greater than the preset link load, it is determined that no congestion occurs on the link.

[0023] Optionally, the real-time detection of whether there is a faulty link in each of the links of the aggregated link includes:

[0024] For each of the links on the aggregated link, obtaining a current occupied bandwidth of the link;

[0025] Determining whether the currently occupied bandwidth of the link is greater than a preset link congestion threshold bandwidth;

[0026] If the currently occupied bandwidth of the link is greater than the preset link congestion threshold bandwidth, the link is determined to be a faulty link, and the faulty link is a congested link;

[0027] If the currently occupied bandwidth of the link is not greater than the preset link congestion threshold bandwidth, it is determined that no congestion occurs on the link.

[0028] Optionally, if the faulty link is an invalid link, when detecting that a faulty link exists in each of the links of the aggregated link, screening out a target available link from each available link, and forwarding the data flow on the faulty link to the target available link, includes:

[0029] When it is detected that there is a faulty link in each of the links of the aggregated link, an OFPMP_FLOW request is sent to switches corresponding to both ends of the faulty link to obtain each data flow on the faulty link;

[0030] Determine a target available link from each available link of the aggregated link, and send each of the data streams to the target available link;

[0031] After each of the data flows is successfully sent to the target available link, the invalid link is deleted from the aggregated link.

[0032] A second aspect of the present invention discloses a data flow scheduling system for an aggregated link, which is applied to a controller in an SDN architecture. The system includes:

[0033] A selection probability calculation unit, configured to obtain the remaining available bandwidth of each link in the aggregated link when a data stream to be allocated is detected, and calculate the selection probability of each link according to the remaining available bandwidth of each link;

[0034] a cumulative selection probability calculation unit, configured to calculate, for each of the links, a cumulative selection probability of the link according to the selection probability of the link and the selection probabilities of each of the other links;

[0035] An allocating unit, configured to determine a target link to be allocated from each of the links according to the cumulative probability of being selected of each of the links, and allocate the data flow to be allocated to the target link to be allocated;

[0036] A real-time detection unit, used for detecting in real time whether there is a faulty link in each of the links of the aggregated link;

[0037] A forwarding unit is used to select a target available link from each available link when a faulty link is detected in each link of the aggregated link, and forward the data flow on the faulty link to the target available link; wherein the available link is a link in the aggregated link that has not failed.

[0038] Optionally, if the faulty link is a congested link, the forwarding unit includes:

[0039] A sampling data packet ratio acquisition unit, configured to, when detecting that a faulty link exists in each of the links of the aggregated link, acquire the sampling data packet ratio of each data flow on the faulty link;

[0040] A candidate elephant flow screening unit, configured to screen out at least one candidate elephant flow from each of the faulty links according to the proportion of sampled data packets of each of the faulty links;

[0041] The transmission rate determination unit of the candidate elephant flow is used to obtain, for each of the candidate elephant flows, the preset flow table entry information corresponding to the candidate elephant flow, and determine the transmission rate of the candidate elephant flow according to the flow table entry information;

[0042] An elephant flow determination unit, configured to determine, for each of the candidate elephant flows, if a transmission rate of the candidate elephant flow is greater than a preset transmission rate, that the candidate elephant flow is an elephant flow;

[0043] The first forwarding subunit is used to select a target available link from each available link and forward the data flow on the faulty link to the target available link; wherein the target available link is the available link with the largest remaining available bandwidth among the available links on the aggregated link.

[0044] Optionally, the real-time detection unit includes:

[0045] A link load detection unit, configured to detect the link load of each link on the aggregated link in real time;

[0046] A first judging unit, configured to judge whether the link load of the link is greater than a preset link load;

[0047] A first determining unit, configured to determine that the link is a faulty link if the link load of the link is greater than the preset link load, and the faulty link is a congested link;

[0048] The second determining unit is configured to determine that congestion does not occur on the link if the link load of the link is not greater than the preset link load.

[0049] The third aspect of the present invention discloses an electronic device, comprising: a processor and a memory, wherein the processor and the memory are connected via a communication bus; wherein the processor is used to call and execute a program stored in the memory; and the memory is used to store a program, wherein the program is used to implement the data flow scheduling method of the aggregated link disclosed in the first aspect of the present invention.

[0050] A fourth aspect of the present invention discloses a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions for executing the data flow scheduling method for an aggregated link disclosed in the first aspect of the present invention.

[0051] The present invention provides a data flow scheduling method, system, electronic device and storage medium of an aggregated link, which are applied to a controller in an SDN architecture. The controller detects in real time whether there is a data flow to be allocated at present. When the data flow to be allocated is detected, the remaining available bandwidth of each link in the aggregated link is obtained, and the selection probability of each link is calculated according to the remaining available bandwidth of each link; for each link, the cumulative selection probability of the link is calculated according to the selection probability of the link and the selection probabilities of each other link; according to the cumulative selection probabilities of each link, a target link to be allocated is determined from each link, and the data flow to be allocated is allocated to the target link to be allocated; and when it is detected that there is a faulty link in each link of the aggregated link, a target available link is screened out from each available link, and the data flow on the faulty link is forwarded to the target available link. The present invention selects a suitable link from each link according to the remaining available bandwidth of each link, so as to avoid selecting a link with a higher load, thereby avoiding link congestion, and automatically implements data flow scheduling through the controller in the SDN architecture, without manual configuration, and can be quickly deployed. At the same time, it also solves the problem of compatibility between devices from different manufacturers. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0053] Figure 1 A schematic diagram of a flow chart of a data flow scheduling method for an aggregated link provided in an embodiment of the present invention;

[0054] Figure 2 A structural diagram of a controller of an SDN architecture provided by an embodiment of the present invention;

[0055] Figure 3 A link determination diagram in an aggregated link provided by an embodiment of the present invention;

[0056] Figure 4 A schematic diagram of the structure of a data flow scheduling system for an aggregation link provided by an embodiment of the present invention;

[0057] Figure 5 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0058] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0059] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0060] It should be noted that the concepts such as "first" and "second" mentioned in the disclosure of the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0061] It should be noted that the modifications of "one" and "plurality" mentioned in the disclosure of the present invention are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0062] In order to better understand the technical terms involved in this application, the following explanations are given:

[0063] Aggregate link: A logical link formed by aggregating multiple data links connecting the same pair of network devices.

[0064] SDN architecture: (Software Defined Network) is a new type of network architecture that separates the control function and forwarding function of network devices to achieve more precise network control and high-speed data forwarding. Its typical architecture includes three layers: application layer, control layer and infrastructure layer. The application layer contains network applications that implement various business functions; the control layer usually refers to the controller of the SDN architecture, which can control the behavior of network devices and monitor the network status of network devices; the infrastructure layer consists of data forwarding devices in the network, which forward data streams to the destination.

[0065] OpenFlow protocol: One of the commonly used network protocols in SDN, which centralizes all control functions on the controller, and the OpenFlow switch is only responsible for forwarding data.

[0066] Data flow: A collection of data packets with the same characteristics in the network. For example, data packets with the same source IP and destination IP can be regarded as a data flow.

[0067] Flow table: A set of rules that an OpenFlow switch follows when processing packets, one of which is called a flow entry. A flow entry specifies the characteristics of the matching packets and the corresponding processing methods, and counts the number of packets and bytes matched during its existence.

[0068] Elephant flow: generally refers to data flow that occupies a large bandwidth, with a small proportion, but carries most of the data in the network, and has high requirements for network throughput and bandwidth. In contrast, rat flow generally refers to data flow that occupies a small bandwidth, with a large proportion, but only occupies a small part of the network traffic.

[0069] See also Figure 1 , shows a flow chart of a data flow scheduling method for an aggregated link provided by an embodiment of the present invention, the data flow scheduling method for the aggregated link is applied to a controller of an SDN architecture, the controller includes a network status monitoring module and a traffic scheduling module, wherein the network status monitoring module includes a topology perception unit, a link load monitoring unit and an elephant flow monitoring unit, and the traffic scheduling module includes forwarding link decision and data flow rescheduling, such as Figure 2 As shown, the data flow scheduling method of the aggregated link specifically includes the following steps:

[0070] S101: When a data flow to be allocated is detected, the remaining available bandwidth of each link in the aggregated link is obtained, and the selection probability of each link is calculated according to the remaining available bandwidth of each link.

[0071] In the examples of this application, see Figure 3, the LLDP data packet is encapsulated in a preset Packet-out message through the topology perception unit in the controller, and the Packet-out message is sent to each OpenFlow switch respectively; for each OpenFlow switch, according to the specified action in the received Packet-out message, the LLDP data packet in the received Packet-out message is sent to the adjacent OpenFlow switch through the corresponding port, so that the adjacent OpenFlow switch encapsulates the received LLDP message and the switch name and corresponding port information of the OpenFlow switch and the adjacent OpenFlow switch in the corresponding Packet-in message to obtain the target Packet-in message, and sends the target Packet-in message to the controller, so that the controller determines the link between the OpenFlow switch and the adjacent OpenFlow switch according to the target Packet-in message, and then can generate the corresponding aggregate link according to the determined links.

[0072] The OpenFlow switches include switch s1 and switch s2.

[0073] In the specific process of executing step S101, the forwarding decision link decision unit in the traffic scheduling module of the controller detects in real time whether there is a data stream to be allocated. When it is detected that there is a data stream to be allocated, the remaining available bandwidth of each link in the aggregated link can be obtained; for each link, the probability of the link being selected can be calculated according to the remaining available bandwidth of the link.

[0074] For example, assuming that there are m member links in the aggregated link, the member link set can be , is the remaining available bandwidth of the ith link, which is the total remaining available bandwidth of the aggregated link. The normalized value of , that is, the probability of the i-th link being selected is shown in formula (1).

[0075] (1)

[0076] S102: For each link, the cumulative selection probability of the link is calculated according to the selection probability of the link and the selection probabilities of each other link.

[0077] In the specific process of executing step S102, for each link, after the probability of selection of the link is calculated, the forwarding decision link decision unit in the traffic scheduling module of the controller can calculate the cumulative probability of selection of the link based on the selection probabilities of the link and each other link.

[0078] The specific method of calculating the cumulative selection probability of the link according to the selection probabilities of the link and other links is shown in formula (2).

[0079] (2)

[0080] Among them, the probability is the probability of the i-th link being selected, is the cumulative probability of being selected for the i-th link, is the probability of the jth link being selected.

[0081] S103: Determine a target link to be allocated from each link according to the cumulative probability of being selected of each link, and allocate the data flow to be allocated to the target link to be allocated.

[0082] In the specific process of executing step S103, for each link, the forwarding decision link decision unit in the traffic scheduling module of the controller determines whether the cumulative selection probability of the link and the cumulative selection probability of the previous link of the link meet the preset formula. If the link meets the preset formula, the data flow to be allocated can be allocated to the target link to be allocated.

[0083] It should be noted that the preset formula can be , where R can be a random number in the range [0,1). is equal to 0.

[0084] S104: Detect in real time whether there is a faulty link in each link of the aggregated link.

[0085] As a preferred method of the present application, the data flow rescheduling unit in the traffic scheduling module of the controller can detect in real time whether there are invalid links in each link of the aggregated link; if there are invalid links in each link of the aggregated link, the invalid link can be determined as a faulty link.

[0086] As another preferred method of an embodiment of the present application, for each link on the aggregated link, the link load of the link is detected in real time by the link load monitoring unit of the controller; it is determined whether the link load of the link is greater than the preset link load; if the link load of the link is greater than the preset link load, it is determined that the link is a faulty link, and the faulty link is a congested link; if the link load of the link is not greater than the preset link load, it is determined that the link is not congested.

[0087] For example, for each link on the aggregated link, the link load monitoring unit of the controller can send an OFPPortStatsRequest request to the switch corresponding to the link based on the OpenFlow protocol, so that the switch can inform the controller of the number of bytes sent and the port activity time of the corresponding port according to the received OFPPortStatsRequest request, and calculate the link load of the link according to the number of bytes sent and the port activity time, as shown in formula (3); determine whether the link load of the link is greater than the preset link load; if the link load of the link is greater than the preset link load, determine that the link is a faulty link, and the faulty link is a congested link; if the link load of the link is not greater than the preset link load, determine that the link is not congested.

[0088] (3)

[0089] Among them, B t is the number of bytes sent at the tth response, B t+1 is the number of bytes sent at the t+1th response, D t is the port activity time at the tth response, D t+1 It is the port activity time during the tth response.

[0090] As another preferred method of an embodiment of the present application, for each link on the aggregated link, the current occupied bandwidth of the link is obtained through the data flow rescheduling unit of the controller; it is determined whether the current occupied bandwidth of the link is greater than the preset link congestion threshold bandwidth; if the current occupied bandwidth of the link is greater than the preset link congestion threshold bandwidth, it is determined that the link is a faulty link and the faulty link is a congested link; if the current occupied bandwidth of the link is not greater than the preset link congestion threshold bandwidth, it is determined that the link is not congested.

[0091] S105: When it is detected that there is a faulty link among the links of the aggregated link, a target available link is selected from the available links, and the data flow on the faulty link is forwarded to the target available link.

[0092] As a preferred method of an embodiment of the present application, if the faulty link is a congested link, the flow table entry information determines the transmission rate of the candidate elephant flow; for each candidate elephant flow, if the transmission rate of the candidate elephant flow is greater than the preset transmission rate, the candidate elephant flow is determined to be an elephant flow; the target available link is selected from each available link, and the data flow on the faulty link is forwarded to the target available link; wherein the target available link is the available link with the largest remaining available bandwidth among the available links on the aggregated link.

[0093] For example, suppose there are m member links in the link aggregation group, where l cong is a congested link, and the used bandwidth is Bcurr ; The rest are idle links, and the idle link set is L={l1, l2, l3, …, l m}, if idle link l i The congestion threshold bandwidth is , the bandwidth used is b i . Congested Link cong There are n elephant flows on the , then the elephant flow set is EF={f1, f2, …,f n}, the elephant flows in the set have been arranged in descending order according to the transmission rate, and the corresponding rates are v1, v2, …, v n , starting from k=1, if , then it can be transferred to the idle link, and the target available link is the link with the largest remaining available bandwidth among the current idle links. Then, the remaining available bandwidth value of the target available link and the used bandwidth on the congested link are updated. , then f k Does not transfer.

[0094] When the updated used bandwidth B on the congested link curr <B th , or k=n, the transfer process ends. Otherwise, the value of k increases by 1 and continues to judge f k Can it be transferred?

[0095] As another preferred method of an embodiment of the present application, if the faulty link is an invalid link, when it is detected that there is a faulty link in each link of the aggregated link, an OFPMP_FLOW request is sent to the switches corresponding to both ends of the faulty link to obtain each data stream on the faulty link; the target available link is determined from each available link of the aggregated link, and each data stream is sent to the target available link; after each data stream is successfully sent to the target available link, the invalid link is deleted from the aggregated link.

[0096] The present invention provides a data flow scheduling method for an aggregated link, which is applied to a controller in an SDN architecture. The controller detects in real time whether there is a data flow to be allocated at present. When the data flow to be allocated is detected, the remaining available bandwidth of each link in the aggregated link is obtained, and the selection probability of each link is calculated according to the remaining available bandwidth of each link; for each link, the cumulative selection probability of the link is calculated according to the selection probability of the link and the selection probability of each other link; according to the cumulative selection probability of each link, a target link to be allocated is determined from each link, and the data flow to be allocated is allocated to the target link to be allocated; and when a faulty link is detected in each link of the aggregated link, the target available link is screened out from each available link, and the data flow on the faulty link is forwarded to the target available link. The present invention selects a suitable link from each link according to the remaining available bandwidth of each link, so that a link with a higher load can be avoided, thereby avoiding link congestion, and automatically realizes data flow scheduling through the controller in the SDN architecture, without manual configuration, and can be quickly deployed, and also solves the problem of incompatibility between devices of different manufacturers.

[0097] Based on the data flow scheduling method of the aggregated link disclosed in the embodiment of the present invention, the embodiment of the present invention also discloses a data flow scheduling system of the aggregated link, such as Figure 4 As shown, the data flow scheduling system of the aggregation link is applied to the controller in the SDN architecture, and the data flow scheduling of the aggregation link includes:

[0098] A selected probability calculation unit 41 is used to obtain the remaining available bandwidth of each link in the aggregated link when a data stream to be allocated is detected, and calculate the selected probability of each link according to the remaining available bandwidth of each link;

[0099] The cumulative selected probability calculation unit 42 is used to calculate the cumulative selected probability of each link according to the selected probability of the link and the selected probabilities of each other link;

[0100] The allocation unit 43 is used to determine the target link to be allocated from each link according to the cumulative probability of being selected of each link, and allocate the data flow to be allocated to the target link to be allocated;

[0101] A real-time detection unit 44, used for detecting in real time whether there is a faulty link in each link of the aggregated link;

[0102] The forwarding unit 45 is used to select a target available link from the available links when a faulty link is detected in the links of the aggregated link, and forward the data flow on the faulty link to the target available link; wherein the available link is a link in the aggregated link that has not failed.

[0103] The specific principles and execution processes of each unit in the data flow scheduling system of the aggregated link disclosed in the above embodiment of the present invention are the same as the data flow scheduling method of the aggregated link disclosed in the above embodiment of the present invention. Please refer to the corresponding parts of the data flow scheduling method of the aggregated link disclosed in the above embodiment of the present invention, and will not be repeated here.

[0104] The present invention provides a data flow scheduling system for an aggregated link, which is applied to a controller in an SDN architecture. The controller detects in real time whether there is a data flow to be allocated at present. When the data flow to be allocated is detected, the remaining available bandwidth of each link in the aggregated link is obtained, and the selection probability of each link is calculated according to the remaining available bandwidth of each link; for each link, the cumulative selection probability of the link is calculated according to the selection probability of the link and the selection probability of each other link; according to the cumulative selection probability of each link, a target link to be allocated is determined from each link, and the data flow to be allocated is allocated to the target link to be allocated; and when a faulty link is detected in each link of the aggregated link, the target available link is screened out from each available link, and the data flow on the faulty link is forwarded to the target available link. The present invention selects a suitable link from each link according to the remaining available bandwidth of each link, so that a link with a higher load can be avoided, thereby avoiding link congestion, and automatically realizes data flow scheduling through the controller in the SDN architecture, without manual configuration, and can be quickly deployed, and also solves the problem of incompatibility between devices of different manufacturers.

[0105] Optionally, if the faulty link is a congested link, the forwarding unit includes:

[0106] A sampling data packet ratio acquisition unit, used for acquiring the sampling data packet ratio of each data flow on the faulty link when a faulty link is detected in each link of the aggregated link;

[0107] A candidate elephant flow screening unit is used to screen at least one candidate elephant flow from each faulty link according to the proportion of sampled data packets of each faulty link;

[0108] The transmission rate determination unit of the candidate elephant flow is used to obtain the preset flow table entry information corresponding to each candidate elephant flow, and determine the transmission rate of the candidate elephant flow according to the flow table entry information;

[0109] An elephant flow determination unit is used for determining, for each candidate elephant flow, if the transmission rate of the candidate elephant flow is greater than a preset transmission rate, the candidate elephant flow as an elephant flow;

[0110] The first forwarding subunit is used to select a target available link from each available link and forward the data flow on the faulty link to the target available link; wherein the target available link is the available link with the largest remaining available bandwidth among each available link on the aggregated link.

[0111] Optional, real-time detection unit, including:

[0112] A link load detection unit, used for detecting the link load of each link on the aggregated link in real time;

[0113] A first judging unit, configured to judge whether a link load of a link is greater than a preset link load;

[0114] A first determining unit, configured to determine that a link is a faulty link if a link load of the link is greater than a preset link load, and the faulty link is a congested link;

[0115] The second determining unit is configured to determine that no congestion occurs on the link if the link load of the link is not greater than a preset link load.

[0116] Optional, real-time detection unit, including:

[0117] A link currently occupied bandwidth acquisition unit, used for acquiring the link currently occupied bandwidth for each link on the aggregated link;

[0118] A second judgment unit, used to judge whether the current occupied bandwidth of the link is greater than a preset link congestion threshold bandwidth;

[0119] A third determining unit, configured to determine that a link is a faulty link and that the faulty link is a congested link if the current occupied bandwidth of the link is greater than a preset link congestion threshold bandwidth;

[0120] The fourth determining unit is configured to determine that the link is not congested if the currently occupied bandwidth of the link is not greater than a preset link congestion threshold bandwidth.

[0121] Optionally, a forwarding unit includes:

[0122] A data flow acquisition unit, configured to, when detecting that a faulty link exists in each link of the aggregated link, send an OFPMP_FLOW request to switches corresponding to both ends of the faulty link to acquire each data flow on the faulty link;

[0123] A second forwarding subunit, configured to determine a target available link from each available link of the aggregated link, and send each data stream to the target available link;

[0124] The deleting unit is used to delete the invalid link from the aggregated link after each data stream is successfully sent to the target available link.

[0125] The present application embodiment provides an electronic device, such as Figure 5 As shown, the electronic device includes a processor 501 and a memory 502, the memory 502 is used to store program codes and data for data flow scheduling of the modular aggregation link, and the processor 501 is used to call the program instructions in the memory to execute the steps shown in the data flow scheduling method of the aggregation link in the above embodiment.

[0126] An embodiment of the present application provides a storage medium, which includes a storage program, wherein when the program is running, a device where the storage medium is located is controlled to execute the data flow scheduling method for the aggregated link shown in the above embodiment.

[0127] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system or system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. The system and system embodiments described above are merely schematic, in which the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0128] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0129] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

[0130] The above are only preferred embodiments of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A data flow scheduling method for an aggregated link, characterized in that: A controller applied to an SDN architecture, the method comprising: When a data stream to be allocated is detected, the remaining available bandwidth of each link in the aggregated link is obtained, and the probability of each link being selected is calculated according to the remaining available bandwidth of each link; For each of the links, calculating the cumulative probability of being selected of the link according to the probability of being selected of the link and the probabilities of being selected of each of the other links; Determine a target link to be allocated from each of the links according to the accumulated probability of being selected of each of the links, and allocate the data flow to be allocated to the target link to be allocated; Detect in real time whether there is a faulty link in each of the links of the aggregated link; When it is detected that there is a faulty link in each of the links of the aggregated link, a target available link is selected from each available link, and the data flow on the faulty link is forwarded to the target available link; wherein the available link is a link in the aggregated link that has not failed; if the faulty link is a congested link, when it is detected that there is a faulty link in each of the links of the aggregated link, a target available link is selected from each available link, and the data flow on the faulty link is forwarded to the target available link, including: When it is detected that there is a faulty link in each of the links of the aggregated link, the proportion of sampled data packets of each data flow on the faulty link is obtained; According to the proportion of sampled data packets of each of the faulty links, at least one candidate elephant flow is selected from each of the faulty links; For each of the candidate elephant flows, obtain the preset flow table entry information corresponding to the candidate elephant flow, and determine the transmission rate of the candidate elephant flow according to the flow table entry information; For each of the candidate elephant flows, if the transmission rate of the candidate elephant flow is greater than a preset transmission rate, determining the candidate elephant flow as an elephant flow; A target available link is selected from each available link, and the data flow on the faulty link is forwarded to the target available link; wherein the target available link is the available link with the largest remaining available bandwidth among each available link on the aggregated link.

2. The method according to claim 1, characterized in that The real-time detection of whether there is a faulty link in each link of the aggregated link includes: For each link on the aggregated link, detecting the link load of the link in real time; Determining whether the link load of the link is greater than a preset link load; If the link load of the link is greater than the preset link load, determining that the link is a faulty link, and the faulty link is a congested link; If the link load of the link is not greater than the preset link load, it is determined that no congestion occurs on the link.

3. The method according to claim 1, characterized in that The real-time detection of whether there is a faulty link in each link of the aggregated link includes: For each of the links on the aggregated link, obtaining a current occupied bandwidth of the link; Determining whether the currently occupied bandwidth of the link is greater than a preset link congestion threshold bandwidth; If the currently occupied bandwidth of the link is greater than the preset link congestion threshold bandwidth, the link is determined to be a faulty link, and the faulty link is a congested link; If the currently occupied bandwidth of the link is not greater than the preset link congestion threshold bandwidth, it is determined that no congestion occurs on the link.

4. The method according to claim 1, characterized in that: If the faulty link is an invalid link, when detecting that a faulty link exists in each of the links of the aggregated link, selecting a target available link from each of the available links, and forwarding the data flow on the faulty link to the target available link, comprises: When it is detected that there is a faulty link in each of the links of the aggregated link, an OFPMP_FLOW request is sent to switches corresponding to both ends of the faulty link to obtain each data flow on the faulty link; Determine a target available link from each available link of the aggregated link, and send each of the data streams to the target available link; After each of the data flows is successfully sent to the target available link, the invalid link is deleted from the aggregated link.

5. A data flow scheduling system for an aggregated link, characterized in that: A controller applied to an SDN architecture, the system comprising: A selection probability calculation unit, configured to obtain the remaining available bandwidth of each link in the aggregated link when a data stream to be allocated is detected, and calculate the selection probability of each link according to the remaining available bandwidth of each link; a cumulative selection probability calculation unit, configured to calculate, for each of the links, a cumulative selection probability of the link according to the selection probability of the link and the selection probabilities of each of the other links; An allocating unit, configured to determine a target link to be allocated from each of the links according to the cumulative probability of being selected of each of the links, and allocate the data flow to be allocated to the target link to be allocated; A real-time detection unit, used for detecting in real time whether there is a faulty link in each of the links of the aggregated link; A forwarding unit, configured to, when detecting that a faulty link exists in each of the links of the aggregated link, select a target available link from each of the available links, and forward the data stream on the faulty link to the target available link; wherein the available link is a link in the aggregated link that has not failed; if the faulty link is a congested link, the forwarding unit comprises: A sampling data packet ratio acquisition unit, configured to, when detecting that a faulty link exists in each of the links of the aggregated link, acquire the sampling data packet ratio of each data flow on the faulty link; A candidate elephant flow screening unit, configured to screen out at least one candidate elephant flow from each of the faulty links according to the proportion of sampled data packets of each of the faulty links; The transmission rate determination unit of the candidate elephant flow is used to obtain, for each of the candidate elephant flows, the preset flow table entry information corresponding to the candidate elephant flow, and determine the transmission rate of the candidate elephant flow according to the flow table entry information; An elephant flow determination unit, configured to determine, for each of the candidate elephant flows, if a transmission rate of the candidate elephant flow is greater than a preset transmission rate, that the candidate elephant flow is an elephant flow; The first forwarding subunit is used to select a target available link from each available link and forward the data flow on the faulty link to the target available link; wherein the target available link is the available link with the largest remaining available bandwidth among the available links on the aggregated link.

6. The system according to claim 5, characterized in that The real-time detection unit comprises: A link load detection unit, configured to detect the link load of each link on the aggregated link in real time; A first judging unit, configured to judge whether the link load of the link is greater than a preset link load; A first determining unit, configured to determine that the link is a faulty link if the link load of the link is greater than the preset link load, and the faulty link is a congested link; The second determining unit is configured to determine that congestion does not occur on the link if the link load of the link is not greater than the preset link load.

7. An electronic device, characterized in that: include: A processor and a memory, wherein the processor and the memory are connected via a communication bus; wherein the processor is used to call and execute a program stored in the memory; The memory is used to store a program, and the program is used to implement the data flow scheduling method for the aggregation link as described in any one of claims 1-4.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to execute the data flow scheduling method for the aggregated link according to any one of claims 1 to 4.

Citation Information

Patent Citations

  • Software-defined network congestion control algorithm based on stream segmentation

    CN106357547A

  • Elephant flow path monitoring and scheduling method based on SDN data center network

    CN110932989A

  • Method for multilink aggregation transmission of data and sending equipment

    CN111277502A

  • Method And Apparatus For Datacenter Congestion Control Based On Software Defined Network

    US20190394129A1