CN114745343B - Network load balancing routing method, device and equipment based on SDN of QoS priority - Google Patents

Network load balancing routing method, device and equipment based on SDN of QoS priority Download PDF

Info

Publication number
CN114745343B
CN114745343B CN202210310244.4A CN202210310244A CN114745343B CN 114745343 B CN114745343 B CN 114745343B CN 202210310244 A CN202210310244 A CN 202210310244A CN 114745343 B CN114745343 B CN 114745343B
Authority
CN
China
Prior art keywords
flow
priority
sdn
service
traffic
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202210310244.4A
Other languages
Chinese (zh)
Other versions
CN114745343A (en
Inventor
刘旭
孟萍
杨龙祥
朱洪波
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing University of Posts and Telecommunications
Original Assignee
Nanjing University of Posts and Telecommunications
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nanjing University of Posts and Telecommunications filed Critical Nanjing University of Posts and Telecommunications
Priority to CN202210310244.4A priority Critical patent/CN114745343B/en
Publication of CN114745343A publication Critical patent/CN114745343A/en
Application granted granted Critical
Publication of CN114745343B publication Critical patent/CN114745343B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/70—Admission control; Resource allocation
    • H04L47/80—Actions related to the user profile or the type of traffic
    • H04L47/805—QOS or priority aware
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L49/00—Packet switching elements
    • H04L49/90—Buffering arrangements
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/50—Reducing energy consumption in communication networks in wire-line communication networks, e.g. low power modes or reduced link rate

Abstract

The application relates to a network load balancing routing method, system and equipment of SDN based on QoS priority. The method comprises the following steps: setting at least two stream buffer queues in at least one terminal exchanger, wherein the at least two stream buffer queues at least comprise a high-priority stream buffer queue and a low-priority stream buffer queue; determining a first-level QoS priority of a service flow according to the service type of the service flow received by at least one terminal exchanger, and distributing the service flow to one of at least two corresponding flow cache queues according to the first-level QoS priority; distributing a second-level QoS priority to the service flows distributed to one of the at least two flow cache queues according to the acquired real-time service flow state of the SDN, wherein the second-level QoS priority is used for indicating the scheduling sequence of the service flows in the one of the at least two flow cache queues; and indicating the scheduling sequence of the service flows in the at least one terminal switch according to the level of the second-level QoS priority of each service flow in one of the at least two flow cache queues.

Description

Network load balancing routing method, device and equipment based on SDN of QoS priority
Technical Field
The present application relates to the field of communications technologies, and in particular, to a network load balancing routing method, device and equipment for an SDN based on QoS priority.
Background
SDN (Software Define Network, software defined network) is a novel network architecture proposed by the university of Stenford, and SDN technology is a brand-new network architecture which breaks through the characteristics of logically centralizing, separating control and forwarding, opening interfaces, being programmable and the like of the traditional network organization application mode.
The control plane and the data plane of the SDN network have flexibility, which is suitable for solving the problem that the static path allocation cannot meet the specific requirements of the service due to the fast dynamic change of the traffic in the network. But SDN networks suffer from the problem of balancing network load while formulating personalized QoS (Quality of Service ) services for users.
Disclosure of Invention
Based on this, it is necessary to provide a network load balancing routing method, device and equipment based on the SDN of QoS priority in order to solve the above technical problem.
In one aspect of the present application, there is provided a network load balancing routing method of an SDN based on QoS priority, the SDN including an SDN network controller and at least one terminal switch communicatively connected to the SDN network controller, the method including:
Step S100, the SDN network controller sets at least two flow cache queues in the at least one terminal switch, where the at least two flow cache queues at least include a high priority flow cache queue and a low priority flow cache queue;
step 200, the SDN network controller determines a first-level QoS priority of a service flow according to a service type of the service flow received by the at least one terminal switch, and allocates the service flow to one of the at least two corresponding flow cache queues according to the first-level QoS priority;
step S300, the SDN network controller allocates a second level QoS priority to the traffic flows allocated to one of the at least two flow cache queues according to the acquired real-time traffic flow status of the SDN, and the second level QoS priorities of the traffic flows in the high priority flow cache queues are all higher than the second level QoS priorities of the traffic flows in the low priority flow cache queues, where the second level QoS priorities are used to indicate a scheduling order of the traffic flows in one of the at least two flow cache queues; and
step S400, the SDN network controller indicates the scheduling sequence of the traffic flows in the at least one terminal switch according to the second-level QoS priority of each traffic flow in one of the at least two flow cache queues.
In an embodiment, the step of allocating the traffic flow to a corresponding one of the at least two flow cache queues according to the first level QoS priority comprises:
step S240, when the first level QoS priority is greater than or equal to a preset threshold, distributing the service flow to the high priority flow cache queue; and
step S242, when the first-level QoS priority is smaller than the preset threshold, allocating the service flow to the low-priority flow cache queue.
In an embodiment, the step of the SDN network controller indicating the scheduling order of the traffic flows in the at least one terminal switch according to the size of the second-level QoS priority of each traffic flow in one of the at least two flow cache queues, further includes:
step S420, when there is no traffic flow in the high priority flow cache queue, the SDN network controller instructs the at least one terminal switch to schedule and forward the traffic flow in the low priority flow cache queue; and
step S440, when the second level QoS priority of the traffic flow received by the at least one terminating switch is higher than the second level QoS priority of the traffic flow being scheduled to be forwarded by the at least one terminating switch, interrupting the current scheduled forwarding and scheduling to forward the received traffic flow.
In an embodiment, the real-time traffic state includes factors that have an effect on a flow schedule, and the step of the SDN network controller allocating the second level QoS priority to the traffic flows allocated to one of the at least two flow cache queues according to the acquired real-time traffic state of the SDN includes:
step S320, establishing a hierarchical structure model in a fuzzy hierarchical analysis algorithm according to the real-time service flow state;
step S340, corresponding triangle fuzzy numbers are obtained according to the importance of each factor in the real-time service flow state of the factor layer of the hierarchical structure model, and a triangle fuzzy comparison matrix is established according to the triangle fuzzy numbers corresponding to each factor;
step S360, calculating a defuzzified normalized clear weight vector corresponding to each factor according to the triangular fuzzy comparison matrix; and
step S380, calculating the second-level QoS priority according to the normalized clear weight vector of each factor and the weight normalized by the quantized value vector of each factor.
In an embodiment, the triangular fuzzy comparison matrix is:
wherein the triangle fuzzy number is l ijk A lower limit value of the triangle blur number, u ijk An upper limit value, m, representing the triangle blur number ijk And expressing the value with the highest possibility of the triangle ambiguity number, i, j epsilon {1,2., n } and i not equal to j, k epsilon {1,2., r } and n represents the number of the factors, and r represents the number of service flows needing to be prioritized in a queue of the at least two flow cache queues to be calculated with the second-level QoS priority.
In one embodiment, the step of calculating the defuzzified clear weight corresponding to each factor according to the triangular fuzzy comparison matrix includes:
step S362, calculating the triangular fuzzy comparison matrix according to a logarithmic least square method calculation formula, to obtain the vector of the normalized fuzzy weight of each factor, where the logarithmic least square method calculation formula is:
s.t.：
the vector of normalized fuzzy weights
wherein ,said vector representing said normalized fuzzy weight of the ith factor, ++>A lower limit value of said normalized fuzzy weight representing said ith factor, ++>An upper limit value of the normalized weight representing the ith factor, and +.>A value representing the i-th factor having a highest likelihood of the normalized fuzzy weight;
Step S364, debluring the vector of the normalized fuzzy weights according to a centroid deblurring algorithm to obtain the clear weights, wherein the centroid deblurring algorithm is as follows:
in an embodiment, the factors of the real-time service flow state include a flow type, a user level, and a data request frequency of the service flow;
the normalized weights of the quantized value vectors of the factors are calculated by the following formulas:
wherein ,Vi Normalized weights for the quantized value vector of the ith factor of the traffic stream, u being the current quantized value of the ith factor, u min A minimum quantized value for the ith factor, and u max A maximum quantized value for the ith factor; and
the second level QoS priority is calculated by the following formula:
wherein ,ωi The clarity weight for the ith factor.
In an embodiment, the method further comprises:
step S520, substituting a first link cost function into Dijkstra algorithm to obtain a first routing path flow table of the service flow in the high priority flow buffer queue, and sending the first routing path flow table to a corresponding terminal switch, where the first link cost function is:
η ij ＝ω 1 *g ij +ω 2 *h ij +ω 3 *m ij ，
wherein ,ηij Representing the first link cost function g ij 、h ij 、m ij Representing the bandwidth, delay and packet loss rate, ω, of the traffic of link (i, j), respectively 1 、ω 2 、ω 3 To the extent of contribution to link cost and the need for service type of delay and packet loss rate according to the bandwidth used by a particular flowSolving a set scale factor, wherein omega 1 、ω 2 、ω 3 The value range of (2) is respectively more than or equal to 0 and less than or equal to 1; and
step S540, substituting a second link cost function into the Dijkstra algorithm to obtain a second routing flow table of the service flow in the low priority flow buffer queue, and sending the second routing flow table to a corresponding terminal switch, where the second link cost function is:
wherein ,Cij Representing the maximum available bandwidth on link (i, j), U ij Indicating the bandwidth that has been used.
In another aspect of the present application, there is also provided a network load balancing routing apparatus of an SDN based on QoS priority, where the SDN includes an SDN network controller and at least one terminal switch communicatively connected to the SDN network controller, the apparatus includes:
a buffer queue maintenance module, configured to set at least two flow buffer queues in the at least one terminal switch through the SDN network controller, where the at least two flow buffer queues at least include a high priority flow buffer queue and a low priority flow buffer queue, and second level QoS priorities of service flows in the high priority flow buffer queue are both higher than second level QoS priorities of service flows in the low priority flow buffer queue;
A first-level priority classification module, configured to determine, by using the SDN network controller, a first-level QoS priority of a service flow according to a service type of the service flow received by the at least one terminal switch, and allocate the service flow to one of the at least two corresponding flow cache queues according to the first-level QoS priority;
a fine-grained second-level prioritization module, configured to allocate, by the SDN network controller, a second-level QoS priority to the traffic flows allocated to one of the at least two flow cache queues according to the acquired real-time traffic flow state of the SDN, where the second-level QoS priority is used to indicate a scheduling order of the traffic flows in the one of the at least two flow cache queues; and
and the scheduling sequence control module is used for indicating the scheduling sequence of the service flows in the at least one terminal switch according to the second-level QoS priority of each service flow in one of the at least two flow cache queues through the SDN network controller.
In yet another aspect of the present application, there is also provided a network load balancing routing device of an SDN based on QoS priority, including a receiver, a transmitter, a memory and a processor, where the memory stores a computer program, and the processor implements the steps of the method in any of the foregoing embodiments when the computer program is executed.
The network load balancing routing method, device and equipment based on the SDN of the QoS priority realize the following technical effects:
by adopting a secondary priority classification method, namely by a first-level QoS priority and a second-level QoS priority, the service flows entering the user switch are subjected to priority ordering twice, so that ordered and reasonable scheduling of the service flows according to different service characteristics and network implementation conditions of the service flows is realized, and network congestion is avoided.
The fuzzy analytic hierarchy process is adopted to consider the real-time service flow state of the service flow, for example, consider factors such as flow type, user level, data request frequency and the like of the service flow to carry out two-layer fine-grained priority ordering, so that personalized QoS service is formulated for the service flow, and user experience and network service stability are improved.
Under the condition of dynamic change of network state, differentiated routing decisions are carried out on service flows with different requirements according to service characteristics so as to meet QoS requirements in order to ensure QoS requirements of different users and optimize network performance.
Drawings
Fig. 1 is a topology diagram of an application network architecture of an SDN network load balancing routing method based on QoS priorities according to an embodiment of the present application.
Fig. 2 is a flowchart of an SDN network load balancing routing method based on QoS priority according to an embodiment of the present application.
Fig. 3 is a flowchart of an SDN network load balancing routing method based on QoS priority in accordance with another embodiment of the present application.
Fig. 4 is a flowchart of an SDN network load balancing routing method based on QoS priority in accordance with a further embodiment of the present application.
Fig. 5 is a flowchart of an SDN network load balancing routing method based on QoS priority in accordance with a further embodiment of the present application.
Fig. 6 is a flowchart of an SDN network load balancing routing method based on QoS priority in accordance with a further embodiment of the present application.
FIG. 7 is a schematic diagram of a first level priority stream buffer queue according to the embodiment of FIG. 2.
Fig. 8 is a schematic diagram illustrating the secondary priority classification and ordering of traffic flows according to the embodiment of fig. 2.
FIG. 9 is a schematic diagram illustrating prioritization of traffic flows using fuzzy analytic hierarchy process according to one embodiment of the present application.
Fig. 10 is a graph comparing the average packet loss ratio versus network load for the embodiment of fig. 9 with an embodiment using a conventional routing algorithm.
Fig. 11 is a block diagram of an SDN network load balancing routing device based on QoS priority in an embodiment of the present application.
Fig. 12 is a schematic internal structure diagram of an SDN network load balancing routing device based on QoS priority in an embodiment of the present application.
Detailed Description
The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
In the present application, ordinal terms such as "first," "second," and the like are used to modify an element does not denote any priority, order, or sequence of acts in a method relative to another element. Unless specifically stated otherwise, such ordinal words are merely used as labels to distinguish one element having a particular name from another element having the same name (except for the ordinal words).
The SDN network load balancing routing method based on the QoS priority can be applied to an application environment shown in figure 1. Wherein different types of applications used by end users in the network have different network QoS requirements. At this time, traffic flows with different QoS requirements (or may be referred to as network flows, data packets, etc. corresponding to a specific service or a specific user's data traffic) enter a terminal switch (e.g., a user Open-Flow switch), and an SDN controller intelligently senses network parameters of the SDN, and thus performs secondary priority classification of the traffic flows. The SDN controller issues a forwarding strategy (for example, a flow table of a routing path) to the switch according to the collected global network parameter information, and the switch performs forwarding operation on the service flows with different priorities according to the forwarding strategy. Through the network architecture, network resources can be reasonably allocated to meet network load balancing and QoS requirements of users can be better met.
As shown in fig. 1, the SDN controller 100 and the terminal switch 200 constitute an SDN physical network. The end switch may be an Open-flow switch. The end switches 200 may be divided by domain, with the edge switches being the edge switches of the set of switches in the domain and the switches communicating with the switches at the edges of the other domains. The SDN controller is provided with a receiver, a transmitter, a memory and a processor. The receiver and the sender are respectively configured to receive the network awareness signal and send a control instruction to each switch 200, for example, send a routing path flow table to the switch, so as to provide a routing path for forwarding the traffic.
It will be appreciated that an undirected graph G (V, E) may also be used to describe an SDN physical network, where V represents a set of nodes in the network, each node representing a private branch exchange (e.g., an Open-flow switch), and E represents a set of links. (i, j) is used to represent the link between nodes i, j. b ij 、d ij 、l ij Respectively representing bandwidth capacity, transmission delay and loss of link (i, j)Packet rate. Binary variableIndicating whether traffic flow k passes through link (i, j), when +.>Indicating that traffic flow k passes through link (i, j), when +.>Indicating that traffic flow k does not pass through link (i, j). K represents the traffic flow set, K ε K. The node of the service flow k entering the network is used as a source node, s is used k A representation; the node of the traffic flow k leaving the network is taken as a destination node, and t is used k And (3) representing. R represents the reachable path from the source node to the destination node, R represents the set of all reachable paths from the source node to the destination node, R e R.
In the above embodiment, if the network is abstracted into the form of a completely undirected graph, where the switches act as nodes in the undirected graph, the weights between the two nodes are determined by the link cost function. And the SDN controller brings different link cost functions into the Dijkstra algorithm according to the priority of the service flow to obtain a corresponding routing path. And selecting an optimal path according to the size of the link weight, thereby realizing the planning of all the traffic flow paths.
In one embodiment, as shown in fig. 2, a method for balancing and routing an SDN network load based on QoS priority is provided, and the method is applied to the SDN network in fig. 1 for illustration, and includes the following steps:
step S100, the SDN network controller sets at least two flow cache queues in the at least one terminal switch, where the at least two flow cache queues at least include a high priority flow cache queue and a low priority flow cache queue.
In this embodiment, step S100 may be understood as a first level priority classification of the traffic flow. In connection with fig. 7, the terminating switch is configured to receive various traffic flows of a user, and may be, for example, an Open-flow switch. The topology of the connection of the terminating switch is not particularly limited herein. Each terminal exchanger is in communication connection with the SDN controller, so that the network state can be sent to the SDN controller, various control instructions can be received from the SDN controller, for example, an optimized route flow table which is obtained by the SDN controller through algorithm calculation and is based on QoS priority and considers the load balancing of the SDN network is received, and the service flows in the flow cache queues with different priorities are forwarded according to the route flow table.
In the present embodiment, the high-priority stream buffer queues and the low-priority stream buffer queues are provided, but it is to be understood that a plurality of stream buffer queues, for example, 3 or 4, may be provided according to the level of priority, as long as the number of stream buffer queues is greater than or equal to 2 so as to facilitate the formation of the first-level priority class, which is not particularly limited in the present application.
Step S200, the SDN network controller determines a first-level QoS priority of a service flow according to a service type of the service flow received by the at least one terminal switch, and allocates the service flow to a corresponding one of the at least two flow cache queues according to the first-level QoS priority.
In this embodiment, the service type of the service flow received by the at least one terminal switch may be determined by a header of an IPV4 message or a source IP address service type field, and those skilled in the art should be aware of various ways of obtaining the service type of the service flow, which is not particularly limited herein.
The service types may have different or the same QoS requirements, and the service types may include, for example, voIP, e-commerce seconds, video conferencing, live video, etc., and the SDN network controller sets corresponding first-level QoS priorities in advance according to the service flows of the different service types. After detecting a service flow and determining the service type of the service flow, the SDN network controller allocates a corresponding first-level QoS priority to the service flow, and allocates the service flow to a corresponding flow cache queue according to the first-level QoS priority level and the preset.
The terminating switch maintains two traffic buffer queues according to priority, and distributes traffic to different priority queues according to QoS requirements of users, which facilitates forwarding traffic in the queues based on first-level priority (first-level QoS priority) later, and then in order of second-level priority (second-level QoS priority), which will be described in detail below.
Step S300, the SDN network controller assigns a second-level QoS priority to the traffic flows assigned to one of the at least two flow cache queues according to the acquired real-time traffic flow status of the SDN, where the second-level QoS priority is higher than the second-level QoS priority of the traffic flows in the low-priority flow cache queue, and the second-level QoS priority is used to indicate a scheduling order of the traffic flows in one of the at least two flow cache queues.
Step S400, the SDN network controller indicates the scheduling sequence of the traffic flows in the at least one terminal switch according to the second-level QoS priority of each traffic flow in one of the at least two flow cache queues.
In this embodiment, the priority classification can be understood as fine-grained second-order prioritization in conjunction with fig. 8. The real-time traffic state of the SDN refers to parameters characterizing traffic that affect the importance of the traffic at the time, and may include, but is not limited to: stream type, user level, data request frequency. It will be appreciated that, for example, the higher the QoS priority of certain flow types, the higher the QoS priority of the traffic flows for users with higher user levels, the higher the QoS priority of the traffic flows with higher frequency of data requests, etc.
In this embodiment, these factors that affect the importance of the traffic flow are considered, so that the comprehensive second-level QoS priority of the traffic flow is obtained comprehensively, and the traffic flows in the single queue are prioritized according to the second-level QoS priority, so as to determine the scheduling order.
It should be noted that the second level QoS priorities of traffic flows in the high priority flow cache queues are all higher than the second level QoS priorities of traffic flows in the low priority flow cache pairs.
Therefore, in the SDN network load balancing routing method based on the QoS priority, the service flows entering the user switch are subjected to twice priority sorting by adopting a two-level priority sorting method, namely, the first-level QoS priority and the second-level QoS priority, so that ordered and reasonable scheduling of the service flows according to different service characteristics and network implementation conditions of the service flows is realized, and network congestion is avoided.
It is understood that in the above step S200, the following steps may be further included:
and step S240, when the first-level QoS priority is greater than or equal to a preset threshold, distributing the service flow to the high-priority flow cache queue.
Step S242, when the first-level QoS priority is smaller than the preset threshold, allocating the service flow to the low-priority flow cache queue.
That is, to allocate traffic flows of different QoS priorities to two flow cache queues, each traffic flow is compared with a preset value. The preset value can be set according to actual application conditions, so that different QoS screening requirements are met.
As shown in fig. 3, in an embodiment of the present application, the order in which the terminal switch schedules and forwards the traffic is adjusted in real time, that is, the method further includes:
step S420, when there is no traffic flow in the high priority flow cache queue, the SDN network controller instructs the at least one terminating switch to schedule and forward the traffic flow in the low priority flow cache queue.
In this embodiment, when there is no traffic in the high priority flow buffer queue (no traffic is allocated to the high priority flow buffer queue and all traffic in the high priority flow buffer queue has been forwarded), the scheduled forwarding task of the traffic in the low priority flow buffer queue is started to be executed.
Step S440, when the second level QoS priority of the traffic flow received by the at least one terminating switch is higher than the second level QoS priority of the traffic flow being scheduled to be forwarded by the at least one terminating switch, interrupting the current scheduled forwarding and scheduling to forward the received traffic flow.
The SDN network controller monitors new traffic received by each terminal switch, and when it is monitored that a new traffic flows into the terminal switch and the second QoS priority of the traffic is higher than the second QoS priority of the traffic being forwarded by the terminal switch, the SDN controller instructs the terminal switch to stop forwarding the current traffic and immediately start forwarding the new traffic.
It can be appreciated that when the second-level QoS priority of the new traffic flow monitored by the SDN network controller is lower than the second-level QoS priority of the traffic flow being forwarded by the terminal switch, forwarding of the current traffic flow is not affected, and the newly-incoming traffic flow is classified and ordered according to the first-level QoS priority and the second-level QoS priority.
Referring to fig. 9, a method for calculating a second level QoS according to a real-time traffic flow status according to an embodiment of the present application is shown in fig. 4. The method may further comprise:
step S320, establishing a hierarchical structure model in a fuzzy hierarchical analysis algorithm according to the real-time service flow state.
In connection with fig. 9, the built hierarchical model includes three layers, namely, a priority layer, a factor layer, and a target layer. In the present embodiment, a plurality of priorities, for example, priority 1, priority 2, and the like, are provided in the priority layer. In the factor layer, there are factors that have an influence on the importance of the traffic flow, in this embodiment, the flow type, the user level and the frequency of data requests. In the target layer is the resulting composite priority.
Step S340, obtaining corresponding triangle fuzzy numbers according to the importance of the factors in the real-time service flow state of the factor layer of the hierarchical structure model, and establishing a triangle fuzzy comparison matrix according to the triangle fuzzy numbers corresponding to the factors.
In this embodiment, the importance of three factors, namely, the stream type, the user level and the data request frequency of the factor layer, is analyzed, and the three factors are represented by triangle ambiguity numbers. And for any flow cache queue, according to the existing service flow situation, converting each factor evaluation of each service flow into a triangle fuzzy number according to a certain conversion scale. The preference of the user is expressed by using the triangular fuzzy number, and a triangular fuzzy comparison matrix is established.
For example, the triangular blur number M may be expressed as (l, M, u), where M is the median of membership of M to 1, and when x=m, x belongs to M safely. l and u are the lower and upper bounds, respectively, and do not belong to the fuzzy number M other than l and u.
Specifically, according to the triangle fuzzy data table, the following triangle fuzzy comparison matrix is adopted for calculation:
wherein the triangle fuzzy number isl ijk A lower limit value of the triangle blur number, u ijk An upper limit value, m, representing the triangle blur number ijk And expressing the value with the highest possibility of the triangle ambiguity number, i, j epsilon {1,2., n } and i not equal to j, k epsilon {1,2., r } and n represents the number of the factors, and r represents the number of service flows needing to be prioritized in a queue of the at least two flow cache queues to be calculated with the second-level QoS priority.
And step S360, calculating the defuzzified normalized clear weight vector corresponding to each factor according to the triangular fuzzy comparison matrix.
And sequentially performing defuzzification and normalization operation on each factor according to the triangular fuzzy comparison matrix, so as to obtain a defuzzification and normalization clear weight vector corresponding to each factor, wherein the clear weight vector contains weights corresponding to each factor.
In another embodiment, a logarithmic least square method calculation formula may be used to calculate the triangular fuzzy comparison matrix, so as to obtain fuzzy weight vectors corresponding to each factor, and perform normalization. The vector of normalized blur weights may then be deblurred according to a centroid deblurring algorithm to obtain the sharp weight vector.
Specifically, in connection with fig. 5, the following logarithmic least squares method (Logarithm Least Squares Method, LLSM) is used to calculate the fuzzy weight, which is better:
s.t.：
The models shown in the above formula are all linear constraint, are linear constraint optimization models, and can be directly solved. Thus, the vector of normalized fuzzy weights
wherein ,the normalized blurring representing the ith factorSaid vector of weights, +.>A lower limit value of said normalized fuzzy weight representing said ith factor, ++>An upper limit value of the normalized weight representing the ith factor, and +.>A value representing the i-th factor having the greatest likelihood of the normalized fuzzy weight.
In this embodiment, the centroid solution blur algorithm is:
therefore, in this embodiment, when the factors are the stream type, the user level and the data request frequency, the clear weight is calculated by normalization, and the clear weight vectors of the three influencing factors are:
ω＝(ω 1 ,ω 2 ,ω 3 )。
step S380, calculating the second-level QoS priority according to the normalized clear weight vector of each factor and the weight normalized by the quantized value vector of each factor.
In the present embodiment, the weight of the quantization value vector normalization of each factor can be obtained by calculation by the following formula:
wherein ,Vi Normalized weights for the quantized value vector of the ith factor of the traffic stream, u being the current quantized value of the ith factor, u min A minimum quantized value for the ith factor, and u max For the ith causeA maximum quantized value of the element; and
the second level QoS priority is calculated by the following formula:
wherein ,ωi The clarity weight for the ith factor.
After obtaining the second-level QoS priority Q corresponding to the service flow in each queue, the service flow priority of each queue may be ordered according to the size of the Q value of the data, where each value of Q corresponds to the priority of each service flow, and the SDN controller performs traffic scheduling according to the order of priority, where the greater the priority data value is, the higher the priority of the user (i.e., service flow).
In this embodiment, a fuzzy analytic hierarchy process is adopted to consider the real-time traffic flow state of the traffic flow, for example, consider factors such as the flow type, user level, data request frequency, etc. of the traffic flow to perform two-layer fine-grained priority ordering, so as to implement personalized QoS service formulation for the traffic flow, and improve user experience and network service stability.
As shown in fig. 6, a dynamic path selection method of an embodiment of the present application is shown.
Specifically, the method comprises the following steps:
step S520, substituting a first link cost function into Dijkstra algorithm to obtain a first routing path flow table of the service flow in the high priority flow buffer queue, and sending the first routing path flow table to a corresponding terminal switch, where the first link cost function is:
η ij ＝ω 1 *g ij +ω 2 *h ij +ω 3 *m ij ，
wherein ,ηij Representing the first link cost function g ij 、h ij 、m ij Representing the bandwidth, delay and packet loss rate, ω, of the traffic of link (i, j), respectively 1 、ω 2 、ω 3 For using bandwidth according to specific flow, delay timeAnd a scaling factor for setting a contribution degree of the packet loss rate to the link cost and a requirement of the service type, wherein ω 1 、ω 2 、ω 3 The range of the values of (2) is respectively more than or equal to 0 and less than or equal to 1.
Step S540, substituting a second link cost function into the Dijkstra algorithm to obtain a second routing flow table of the service flow in the low priority flow buffer queue, and sending the second routing flow table to a corresponding terminal switch, where the second link cost function is:
wherein ,Cij Representing the maximum available bandwidth on link (i, j), U ij Indicating the bandwidth that has been used.
For flows with high QoS requirements, the method selects the minimum cost path transmission. And for flows with low QoS requirements and no QoS requirements, load balancing of the network is considered, so that the shortest path is selected according to the remaining bandwidth situation. The method not only can ensure the transmission quality of the high-priority service flow, but also can improve the transmission performance of the service flow which is generally required.
When routing paths, the physical network can be abstracted into the form of a completely undirected graph, wherein a switch is used as a node in the undirected graph, and the weight between two nodes is determined by a corresponding link cost function. And the SDN controller brings different link cost functions into Dijkstra algorithm according to the priority of the service flow to obtain corresponding routing paths to obtain link weights. And selecting the optimal paths of the modes aiming at the service flows in the flow cache queues with different priorities according to the size of the link weight, thereby realizing the planning of all the service flow paths.
Under the condition of dynamic change of network state, differentiated routing decisions are carried out on service flows with different requirements according to service characteristics so as to meet QoS requirements in order to ensure QoS requirements of different users and optimize network performance.
As shown in fig. 10, a comparison of average packet loss rate versus network load for an embodiment versus an embodiment using a conventional routing algorithm is shown. The figure shows the difference in the efficiency of the network load and average packet loss rate of a conventional routing algorithm implemented by e.g. ECMP, hiQoS and the inventive method. As can be seen from the figure, with the increase of the network load, the average packet loss rate of the routing method in the embodiment of the application is lower than that of other methods.
Specifically, a ubuntu16.04 system is installed in VM Ware, and an Ryu controller and a Mininet are installed on the basis of the system. And constructing a network topology in the Mininet, connecting the Ryu controller and the switch through an OpenFlow1.3 protocol, and configuring an sFlow agent to realize sampling configuration of data. Various test streams required for the experiment were generated by the iperf tool.
The bandwidth of the main link was set to 100Mbps in the experiment. Considering the randomness of the link delay, the delay is set to a smooth random process with an expected value of more than 0 and less than 1, and the unit is (ms). The packet loss rate is a ratio of more than 0 and less than 1 and is generated by a random number generator. In the next experiments, the use of the iperf to generate different traffic flows tested the impact of the proposed algorithm on end-to-end QoS and network load in the SDN environment.
Fig. 6 simulates the packet loss rate for ECMP, hiQoS and the three algorithms herein. As network load increases, network congestion occurs, resulting in packet loss to some extent. The results show that higher network packet loss rates can make the Internet more heavily loaded. The simulation experiment compares the SDN network load balancing routing algorithm based on the QoS priority with the traditional algorithm, among the three routing algorithms, the algorithm provided with the network load increasing provides the lowest packet loss rate, and the algorithm can effectively avoid congestion and improve the network load balancing characteristic due to the use of a priority mechanism and a classified routing link cost function.
It should be understood that, although the steps in the flowcharts of fig. 2-6 are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in fig. 2-6 may include multiple sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor does the order in which the sub-steps or stages are performed necessarily occur in sequence, but may be performed alternately or alternately with at least a portion of the sub-steps or stages of other steps or other steps.
In an embodiment, as shown in fig. 11, there is provided a network load balancing routing apparatus of an SDN based on QoS priority, where the SDN includes an SDN network controller and at least one terminal switch communicatively connected to the SDN network controller, and the apparatus includes:
a buffer queue maintenance module, configured to set at least two flow buffer queues in the at least one terminal switch through the SDN network controller, where the at least two flow buffer queues at least include a high priority flow buffer queue and a low priority flow buffer queue, and second level QoS priorities of service flows in the high priority flow buffer queue are both higher than second level QoS priorities of service flows in the low priority flow buffer queue;
A first-level priority classification module, configured to determine, by using the SDN network controller, a first-level QoS priority of a service flow according to a service type of the service flow received by the at least one terminal switch, and allocate the service flow to one of the at least two corresponding flow cache queues according to the first-level QoS priority;
a fine-grained second-level prioritization module, configured to allocate, by the SDN network controller, a second-level QoS priority to the traffic flows allocated to one of the at least two flow cache queues according to the acquired real-time traffic flow state of the SDN, where the second-level QoS priority is used to indicate a scheduling order of the traffic flows in the one of the at least two flow cache queues; and
and the scheduling sequence control module is used for indicating the scheduling sequence of the service flows in the at least one terminal switch according to the second-level QoS priority of each service flow in one of the at least two flow cache queues through the SDN network controller.
The specific definition of the network load balancing routing device of the SDN based on QoS priority may be referred to as the definition of the network load balancing routing method of the SDN based on QoS priority, which is not described herein. The modules in the network load balancing routing device of the SDN based on QoS priority may be implemented in whole or in part by software, hardware, and a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In an embodiment, a routing device is provided, which may be an SDN controller, and an internal structure diagram thereof may be shown in fig. 12. The routing device includes a processor, memory, transmitter and receiver connected by a system bus, and may also include a display screen and/or input device in some embodiments. Wherein the processor of the routing device is configured to provide computing and control capabilities. The memory of the routing device includes a non-volatile storage medium, an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The transmitter and receiver of the routing device are used for communicating with external terminals via a network connection. The computer program, when executed by a processor, implements a network load balancing routing method for an SDN based on QoS priorities. The display screen of the routing equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the routing equipment can be a touch layer covered on the display screen, can also be a key, a track ball or a touch pad arranged on the routing equipment shell, and can also be an external keyboard, a touch pad or a mouse and the like.
It will be appreciated by those skilled in the art that the structure shown in fig. 12 is merely a block diagram of a portion of the structure associated with the present inventive arrangements and is not limiting of the routing device to which the present inventive arrangements are applied, and that a particular routing device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.
In summary, the exemplary embodiments of the present application provide a routing method for balancing network load based on QoS priority, which is a method for solving the problem of balancing network load while making personalized QoS service for users. In order to ensure QoS requirements of different users and optimize network performance, under the condition of dynamic change of network state, firstly, classifying the services of each application according to service characteristics and giving corresponding priority, and secondly, comprehensively analyzing information such as QoS requirements of the services, real-time state of links, priority of the services and the like, flexibly distributing paths, meeting the QoS requirements of the network services and relieving network load pressure. The exemplary embodiments of the application combine the two-stage classification method to carry out two-layer grain refining priority ordering on the business flow entering the terminal exchanger, optimize the link cost function according to the real-time parameters such as bandwidth, time delay, load and the like, and realize the effective resource allocation and network load balancing as targets to carry out path selection.
In an embodiment, a network load balancing routing device of an SDN based on QoS priority is provided, which includes a receiver, a transmitter, a memory and a processor, where the memory stores a computer program, and the processor executes the computer program to implement steps of a routing method according to embodiments of the present application.
The specific definition of the network load balancing routing device of the SDN based on QoS priority may be referred to the definition of the network load balancing routing method of the SDN based on QoS priority above, and will not be described herein.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples illustrate only a few embodiments of the application, which are described in detail and are not to be construed as limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of protection of the present application is to be determined by the appended claims.

Claims (8)

1. A network load balancing routing method of an SDN based on QoS priority, the SDN comprises an SDN network controller and at least one terminal switch communicatively connected with the SDN network controller,
characterized in that the method comprises:
step S100, the SDN network controller sets at least two flow cache queues in the at least one terminal switch, where the at least two flow cache queues at least include a high priority flow cache queue and a low priority flow cache queue;
Step 200, the SDN network controller determines a first-level QoS priority of a service flow according to a service type of the service flow received by the at least one terminal switch, and allocates the service flow to one of the at least two corresponding flow cache queues according to the first-level QoS priority;
step S300, the SDN network controller allocates a second level QoS priority to the traffic flows allocated to one of the at least two flow cache queues according to the acquired real-time traffic flow status of the SDN, and the second level QoS priorities of the traffic flows in the high priority flow cache queues are all higher than the second level QoS priorities of the traffic flows in the low priority flow cache queues, where the second level QoS priorities are used to indicate a scheduling order of the traffic flows in one of the at least two flow cache queues; and
step S400, the SDN controller indicates the scheduling sequence of the service flows in the at least one terminal exchanger according to the second-level QoS priority of each service flow in one of the at least two flow cache queues;
wherein the SDN network controller indicates the scheduling sequence of the traffic flows in the at least one terminal switch according to the second-level QoS priority of each traffic flow in one of the at least two flow cache queues, and the method further includes:
Step S420, when there is no traffic flow in the high priority flow cache queue, the SDN network controller instructs the at least one terminal switch to schedule and forward the traffic flow in the low priority flow cache queue; and
step S440, when the second level QoS priority of the traffic flow received by the at least one terminating switch is higher than the second level QoS priority of the traffic flow being scheduled to be forwarded by the at least one terminating switch, interrupting the current scheduled forwarding and scheduling to forward the received traffic flow;
the real-time traffic state includes factors influencing flow scheduling, and the step of the SDN network controller allocating the second-level QoS priority to the traffic flow allocated to one of the at least two flow cache queues according to the acquired real-time traffic state of the SDN includes:
step S320, establishing a hierarchical structure model in a fuzzy hierarchical analysis algorithm according to the real-time service flow state;
step S340, corresponding triangle fuzzy numbers are obtained according to the importance of each factor in the real-time service flow state of the factor layer of the hierarchical structure model, and a triangle fuzzy comparison matrix is established according to the triangle fuzzy numbers corresponding to each factor;
Step S360, calculating a defuzzified normalized clear weight vector corresponding to each factor according to the triangular fuzzy comparison matrix; and
step S380, calculating the second-level QoS priority according to the normalized clear weight vector of each factor and the weight normalized by the quantized value vector of each factor.
2. The network load balancing routing method of the QoS priority based SDN of claim 1,
the step of allocating the traffic flow to a corresponding one of the at least two flow cache queues according to the first level QoS priority comprises:
step S240, when the first level QoS priority is greater than or equal to a preset threshold, distributing the service flow to the high priority flow cache queue; and
step S242, when the first-level QoS priority is smaller than the preset threshold, allocating the service flow to the low-priority flow cache queue.
3. The network load balancing routing method of the QoS priority based SDN of claim 1,
the method is characterized in that the triangular fuzzy comparison matrix is as follows:
wherein the triangle fuzzy number isl ijk A lower limit value of the triangle blur number, u ijk An upper limit value, m, representing the triangle blur number ijk And expressing the value with the highest possibility of the triangle ambiguity number, i, j epsilon {1,2., n } and i not equal to j, k epsilon {1,2., r } and n represents the number of the factors, and r represents the number of service flows needing to be prioritized in a queue of the at least two flow cache queues to be calculated with the second-level QoS priority.
4. The network load balancing routing method of the SDN based on QoS priority as claimed in claim 1 or 3,
the method is characterized in that the step of calculating the defuzzified clear weight corresponding to each factor according to the triangular fuzzy comparison matrix comprises the following steps:
step S362, comparing the triangle with fuzzy according to the logarithmic least square method calculation formula
Calculating a matrix to obtain a vector of normalized fuzzy weights of the factors respectively, wherein the logarithmic least square method calculation formula is as follows:
s.t.：
the vector of normalized fuzzy weights
wherein ,said vector representing said normalized fuzzy weight of the ith factor, ++>A lower limit value of said normalized fuzzy weight representing said ith factor, ++>An upper limit value of the normalized weight representing the ith factor, and A value representing the i-th factor having a highest likelihood of the normalized fuzzy weight;
step S364, debluring the vector of the normalized fuzzy weights according to a centroid deblurring algorithm to obtain the clear weights, wherein the centroid deblurring algorithm is as follows:
5. the network load balancing routing method of the QoS priority based SDN of claim 4,
the real-time service flow state is characterized in that each factor of the real-time service flow state comprises the flow type, the user level and the data request frequency of the service flow;
the normalized weights of the quantized value vectors of the factors are calculated by the following formulas:
wherein ,Vi Normalized weights for the quantized value vector of the ith factor of the traffic stream, u being the current quantized value of the ith factor, u min A minimum quantized value for the ith factor, and u max A maximum quantized value for the ith factor; and
the second level QoS priority is calculated by the following formula:
wherein ,ωi The clarity weight for the ith factor.
6. The network load balancing routing method of the QoS priority based SDN of claim 1,
characterized in that the method further comprises:
Step S520, substituting the first link cost function into Dijkstra algorithm to obtain a first route flow table of the service flow in the high priority flow cache queue, and converting the first routeThe path flow table is sent to a corresponding terminal exchanger, wherein the first link cost function is as follows: η (eta) ij ＝ω 1 *g ij +ω 2 *h ij +ω 3 *m ij ，
wherein ,ηij Representing the first link cost function g ij 、h ij 、m ij Representing the bandwidth, delay and packet loss rate, ω, of the traffic of link (i, j), respectively 1 、ω 2 、ω 3 A scale factor set for the contribution degree of time delay and packet loss rate to link cost and the requirement of service type according to the use bandwidth of specific stream, wherein omega 1 、ω 2 、ω 3 The value range of (2) is respectively more than or equal to 0 and less than or equal to 1; and
step S540, substituting a second link cost function into the Dijkstra algorithm to obtain a second routing flow table of the service flow in the low priority flow buffer queue, and sending the second routing flow table to a corresponding terminal switch, where the second link cost function is:
wherein ,Cij Representing the maximum available bandwidth on link (i, j), U ij Indicating the bandwidth that has been used.
7. Network load balancing routing apparatus for an SDN based on QoS priority, for implementing the steps of the method of any one of claims 1 to 6, the SDN comprising an SDN network controller and at least one terminal switch communicatively connected to the SDN network controller, characterized in that the apparatus comprises:
A buffer queue maintenance module, configured to set at least two flow buffer queues in the at least one terminal switch through the SDN network controller, where the at least two flow buffer queues at least include a high priority flow buffer queue and a low priority flow buffer queue, and second level QoS priorities of service flows in the high priority flow buffer queue are both higher than second level QoS priorities of service flows in the low priority flow buffer queue;
a first-level priority classification module, configured to determine, by using the SDN network controller, a first-level QoS priority of a service flow according to a service type of the service flow received by the at least one terminal switch, and allocate the service flow to one of the at least two corresponding flow cache queues according to the first-level QoS priority;
a fine-grained second-level prioritization module, configured to allocate, by the SDN network controller, a second-level QoS priority to the traffic flows allocated to one of the at least two flow cache queues according to the acquired real-time traffic flow state of the SDN, where the second-level QoS priority is used to indicate a scheduling order of the traffic flows in the one of the at least two flow cache queues; and
And the scheduling sequence control module is used for indicating the scheduling sequence of the service flows in the at least one terminal switch according to the second-level QoS priority of each service flow in one of the at least two flow cache queues through the SDN network controller.
8. Network load balancing routing device of an SDN based on QoS priorities, comprising a receiver, a sender, a memory and a processor, said memory storing a computer program, characterized in that said processor implements the steps of the method according to any of claims 1 to 6 when executing said computer program.
CN202210310244.4A 2022-03-28 2022-03-28 Network load balancing routing method, device and equipment based on SDN of QoS priority Active CN114745343B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210310244.4A CN114745343B (en) 2022-03-28 2022-03-28 Network load balancing routing method, device and equipment based on SDN of QoS priority

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210310244.4A CN114745343B (en) 2022-03-28 2022-03-28 Network load balancing routing method, device and equipment based on SDN of QoS priority

Publications (2)

Publication Number Publication Date
CN114745343A CN114745343A (en) 2022-07-12
CN114745343B true CN114745343B (en) 2023-08-11

Family

ID=82276982

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210310244.4A Active CN114745343B (en) 2022-03-28 2022-03-28 Network load balancing routing method, device and equipment based on SDN of QoS priority

Country Status (1)

Country Link
CN (1) CN114745343B (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107896192A (en) * 2017-11-20 2018-04-10 电子科技大学 The QoS control method of differentiated service priority in a kind of SDN
CN110417656A (en) * 2019-08-07 2019-11-05 广东电网有限责任公司 A kind of QoS guarantee system and method based on SDN network

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2015187946A1 (en) * 2014-06-05 2015-12-10 KEMP Technologies Inc. Adaptive load balancer and methods for intelligent data traffic steering
US9917781B2 (en) * 2014-06-05 2018-03-13 KEMP Technologies Inc. Methods for intelligent data traffic steering

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107896192A (en) * 2017-11-20 2018-04-10 电子科技大学 The QoS control method of differentiated service priority in a kind of SDN
CN110417656A (en) * 2019-08-07 2019-11-05 广东电网有限责任公司 A kind of QoS guarantee system and method based on SDN network

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于多队列切换的SDN拥塞控制;张莉敏;田小路;杨廷来;;软件(03);全文 *

Also Published As

Publication number Publication date
CN114745343A (en) 2022-07-12

Similar Documents

Publication Publication Date Title
EP3318008B1 (en) Negotiating quality of service for data flows
Deng et al. An application-aware QoS routing algorithm for SDN-based IoT networking
US9806987B2 (en) Method and apparatus for reducing response time in information-centric networks
Tajiki et al. CECT: computationally efficient congestion-avoidance and traffic engineering in software-defined cloud data centers
Kim et al. Dynamic bandwidth provisioning using ARIMA-based traffic forecasting for Mobile WiMAX
Banaie et al. Load-balancing algorithm for multiple gateways in Fog-based Internet of Things
Daradkeh et al. Development of QoS methods in the information networks with fractal traffic
CN114827021B (en) Multimedia service flow acceleration system based on SDN and machine learning
US20150100694A1 (en) Use of iterative learning for resolving scalability issues of bandwidth broker
Tajiki et al. QRTP: QoS-aware resource reallocation based on traffic prediction in software defined cloud networks
JP7067170B2 (en) Route control system, route control method, and program
Cai et al. Optimal cloud network control with strict latency constraints
CN113518039B (en) Deep reinforcement learning-based resource optimization method and system under SDN architecture
Craveirinha et al. A meta-model for multiobjective routing in MPLS networks
CN114745343B (en) Network load balancing routing method, device and equipment based on SDN of QoS priority
Araujo et al. Qos in forwarding strategies for icn: New algorithm and experimental evaluation
Bensalem et al. Towards optimal serverless function scaling in edge computing network
Schmidt et al. Scalable bandwidth optimization in advance reservation networks
Rezaei et al. QoS‐based routing scheme in software‐defined networks using fuzzy analytic hierarchy process
Biyar et al. Flow-based network tomography agent for software defined data center
Maity et al. Resq: Reinforcement learning-based queue allocation in software-defined queuing framework
Boussada et al. A Flow-Level Performance Evaluation of Elastic Traffic Under Low Latency Queuing System
Chilukuri et al. NimbleCache-Low Cost, Dynamic cache allocation in constrained edge environments
Świątek et al. Adaptive packet scheduling for requests delay guaranties in packet-switched computer communication network
Abramova Analysis of the Impact of Priority Traffic Control Mechanisms on Network Quality of Service

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant