CN106850859A - A kind of user's request distribution method of the CDN based on SDN - Google Patents

A kind of user's request distribution method of the CDN based on SDN Download PDF

Info

Publication number
CN106850859A
CN106850859A CN201710192747.5A CN201710192747A CN106850859A CN 106850859 A CN106850859 A CN 106850859A CN 201710192747 A CN201710192747 A CN 201710192747A CN 106850859 A CN106850859 A CN 106850859A
Authority
CN
China
Prior art keywords
proxy server
user
cdn network
cdn
time
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.)
Pending
Application number
CN201710192747.5A
Other languages
Chinese (zh)
Inventor
秦枫
赵志峰
张宏纲
李荣鹏
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.)
Zhejiang University ZJU
Original Assignee
Zhejiang University ZJU
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 Zhejiang University ZJU filed Critical Zhejiang University ZJU
Priority to CN201710192747.5A priority Critical patent/CN106850859A/en
Publication of CN106850859A publication Critical patent/CN106850859A/en
Pending legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
    • H04L67/63—Routing a service request depending on the request content or context
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/12—Shortest path evaluation
    • H04L45/124—Shortest path evaluation using a combination of metrics

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

The invention discloses a kind of user's request distribution method of the CDN based on SDN.SDN carries out centralized management by controller to bottom CDN equipment, while can be with real-time collecting network global information, for forming the intelligent decision of user's request redirection in CDN.Intelligent centre is used as global brain, do not constrained by single SDN controllers disposal ability is limited, the decision-making module that the decision-making that some amounts of calculation are larger, computing is complicated is transferred to the intelligent centre being deployed in is carried out, the pressure of SDN controllers can be mitigated, with efficient intelligent algorithm, cross-domain intelligent decision is formed, while making network function that there is stronger autgmentability.The decision-making module of intelligent centre can effectively reduce user's average response time further using the user's request allocation algorithm based on MPC, improve user bandwidth satisfaction, while ensureing the stability of system.

Description

SDN-based CDN (content delivery network) user request distribution method
Technical Field
The invention relates to a user request distribution method of a CDN (content delivery network) based on an SDN (software defined network), belonging to the field of network communication.
Background
As research progresses, the conventional CDN network has the following drawbacks:
1. the network global topology and the perception network state information cannot be known, and the control [ means and capability of routing and user QoE cannot be guaranteed;
2. DNS redirection is generally used, which makes it difficult to manage long-time, bulky flows, and how to distribute user requests to reasonable proxy servers to improve quality of service remains a problem.
Based on this, application of Software Defined Network (SDN) technology in CDN networks is beginning to attract attention. The SDN is a novel network architecture based on software programmable thought, a control plane of network equipment is separated from a data forwarding plane, centralized unified scheduling can be achieved by the control plane, and a network provider can dynamically allocate resources for users in real time according to requirements to achieve flexible intelligent control of the network. Meanwhile, the complex network management and control problem can be more effectively solved by utilizing the programmability of the SDN. The core advantages of combining SDN and CDN networks are:
1. SDN enables CDN providers to better manage and upgrade underlying network devices.
2. SDN provides interfaces to collect real-time network information, including network global topology information, latency of links, traffic monitoring information, etc., which can be used to form decisions for request distribution.
3. The SDN controller redirects the user request to a proper server in real time by directly rewriting flow table items through an OpenFlow protocol.
In an existing system architecture combining an SDN and a CDN, an SDN controller selects a proper proxy server and a proper path by adopting a dynamic allocation algorithm according to collected network information, but the processing capacity of a single SDN controller is limited and cross-control-domain decision cannot be supported.
Now, as the quality requirement of the user for obtaining resources is higher and higher, server selection and path selection are the key for determining the performance of the CDN network. In the existing CDN research, joint optimization is hardly performed for server selection and path selection.
Disclosure of Invention
The invention aims to provide a user request distribution method of a CDN (content delivery network) based on an SDN (software defined network), which can support cross-control-domain decision of the CDN.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
the invention discloses a user request distribution method of a CDN (content delivery network) based on an SDN (software defined network), which comprises the following steps of:
step one, each SDN controller collects CDN network information in real time through an OpenFlow interface, wherein the CDN network information comprises network global topology, link time delay and flow, and uploads the collected CDN network information to a database of an intelligent center;
step two, a decision module of the intelligent center calculates the proportion of content requests distributed to users by each proxy server of the CDN by adopting a dynamic distribution algorithm according to the CDN network information received by the database, calculates the proportion of the content requests distributed to the users by paths from an inlet switch of the CDN to each proxy server, and sends the IP addresses of the proxy servers, the proportion of the content requests distributed to the users by each proxy server of the CDN network and the proportion of the content requests distributed to the users by paths from the inlet switch of the CDN network to each proxy server to the SDN controller;
modifying a flow table entry of an ingress switch by the SDN controller according to the received IP address of the proxy server, the proportion of content requests distributed to users by each proxy server of the CDN network and the proportion of content requests distributed to users by paths from the ingress switch of the CDN network to each proxy server;
step four, the user sends a content request to the entrance switch, the entrance switch matches the content request with the flow table item and forwards the matched content request to the target proxy server, the user obtains the requested content from the target proxy server, and the proportion of the content request distributed to the user by the target proxy server is not 0.
Further, the present invention calculates the proportion of content requests distributed to users by each proxy server of the CDN network and the proportion of content requests distributed to users by the path from the ingress switch of the CDN network to each proxy server using formula (1):
wherein,
in the formulas (1) - (6), JdRepresents a user average response time optimization parameter, JbRepresents a user bandwidth satisfaction degree offset optimization parameter, JpRepresenting a stability optimization parameter, ωbDeviation optimization parameter J for representing user bandwidth satisfaction degreebOptimizing parameter J with respect to user average response timedWeight value of, ωpRepresents the stability optimization parameter JpOptimizing parameter J with respect to user average response timedWeighted value of ps(t) represents the proportion of content requests assigned to users by each proxy server of the CDN network during t, i represents the path from the ingress switch of the CDN network to the target proxy server s, pi,s(t) represents the proportion of content requests for users whose path i is assigned to target proxy server s during time t, s representing a single target proxy server; s denotes a set of target proxy servers, IsRepresenting a set of paths from an inlet switch of the CDN network to each target proxy server s, d (T) representing the average user response time in T time, H representing a prediction time domain, T representing the current time, s representing a single target proxy server, and e representing a link in a path i from the inlet switch of the CDN network to the target proxy server s; BLe(t) represents the user bandwidth satisfaction degree deviation degree of a link e in a path i from an inlet switch of the CDN network to a target proxy server s in t time; v represents a mathematical average of the user's content request; rtRepresenting the total number of content requests of users in the CDN network during time t; ceIndicating the capacity, BP, of link eiAnd (t) represents the user bandwidth satisfaction degree deviation degree of the path from the inlet switch of the CDN network to each target proxy server in the time t.
Further, the average response time of the user in the time t of the present invention is:
wherein,
in equations (7) and (8), d (t) represents the average response time of the user during t time, ds(t) represents the processing time of each proxy server in the CDN networkM, deRepresents the link delay, lambda, of the link e in the path i from the ingress switch of the CDN network to the target proxy server s collected by the SDN controllertRepresenting the average rate, λ, at which a user's content request reaches the target proxy server s during time tsRepresenting the processing rate, p, of the target proxy server ss(t) represents the proportion of content requests that are distributed to users by each proxy server of the CDN network over time.
Compared with the prior art, the invention has the beneficial effects that:
(1) the intelligent center is used as a global brain and is not limited by the limited processing capacity of a single SDN controller, and some decisions with large calculation amount and complex operation are transferred to a decision module of the intelligent center, so that the pressure of the SDN controller is reduced.
(2) The SDN controller collects the information of the CDN in real time through the OpenFlow interface and uploads the collected information to the database of the intelligent center, so that the decision module of the intelligent center makes a decision according to the information received by the database, and the decision of the CDN across control domains can be supported.
(3) The decision module of the intelligent center further adopts a user request allocation algorithm based on the MPC, so that the average response time of the user can be effectively reduced, the bandwidth satisfaction degree of the user is improved, and the stability of the system is ensured.
Drawings
Fig. 1 is an architectural diagram of an SDN-based CDN network of the present invention.
Detailed Description
The following describes a user request allocation method for a CDN network based on an SDN in detail with reference to the accompanying drawings.
In the present invention, a CDN network architecture based on SDN may be as shown in fig. 1. In fig. 1, the SDN-based CDN network architecture mainly includes three parts, an underlying network forwarding device, an SDN controller, and an intelligent center.
1. Underlying network forwarding device
The underlying network forwarding device is typically an OpenFlow switch. The OpenFlow switch interacts with the SDN controller through an OpenFlow protocol.
2. SDN controller
The SDN controller is mainly responsible for collecting CDN network information in real time, uploading the collected CDN network information to a database of an intelligent center, and controlling bottom layer forwarding equipment to forward by modifying flow table items in the OpenFlow switch. The CDN network information comprises network global topology, link delay, flow and the like.
3. Intelligent center
The intelligent center comprises a decision module and a database. The database is responsible for storing CDN network information, and the decision module is responsible for forming an intelligent decision according to the network information.
The working process of the SDN-based CDN network user request allocation method is as follows:
the method comprises the following steps: each SDN controller collects CDN network information in real time through an OpenFlow interface, the CDN network information comprises network global topology, link time delay and flow, and the collected CDN network information is uploaded to a database of an intelligent center.
Step two: a decision module of the intelligent center calculates the proportion p of content requests distributed to users by each proxy server of the CDN network by adopting a dynamic distribution algorithm according to the CDN network information received by the databases(t) calculating the proportion of the content requests distributed to the users by the paths from the entry switch of the CDN network to the proxy servers, and distributing the IP addresses of the proxy servers and the proportion p of the content requests distributed to the users by the proxy servers of the CDN networki,s(t) and ingress switch for CDN networksThe content request proportion of the paths to the proxy servers is distributed to the users and is sent to the SDN controller.
Step three: the SDN controller distributes the proportion p of the content request of the user to each proxy server of the CDN network according to the received IP address of the proxy servers(t) and the proportion p of content requests for which the path from the ingress switch of the CDN network to each proxy server is assigned to a useri,s(t) modifying a flow entry of the ingress switch.
Step four: the user sends a content request to the entrance switch, the entrance switch matches the content request with the flow table item and forwards the matched content request to a target proxy server, the user obtains the requested content from the target proxy server, and the target proxy server is distributed to the proportion p of the content request of the users(t) is not 0.
Further, the present invention may adopt a method of dynamically allocating user requests sent by a single ingress switching device to perform optimization, considering two core indicators of user-perceived QoE (Quality of Experience): the average user response time and the user bandwidth satisfaction degree deviation degree are considered, meanwhile, stability is considered, an MPC (Model predictive control) algorithm is applied to perform content request distribution (MPC-based user request distribution algorithm), and distribution results are presented in a proportional mode. The specific method comprises the following steps:
solving the nonlinear programming problem by using the following formula (1), calculating the proportion of the content requests distributed to the users by each proxy server of the CDN network and the proportion of the content requests distributed to the users by the path from the inlet switch of the CDN network to each proxy server:
wherein,
in the formula (1), JdRepresenting the user average corresponding time optimization parameter, JbRepresents a user bandwidth satisfaction degree offset optimization parameter, JpDenotes the stability optimization parameter, ps(t) represents the proportion of content requests distributed to users by each proxy server of the CDN network within t time; p is a radical ofi,s(t) represents the proportion of content requests for which the path i from the ingress switch of the CDN network to the target proxy server s is allocated to the user of the target proxy server s during time t; s represents a single target proxy server; s denotes a set of target proxy servers, IsRepresenting the set of paths from the ingress switches of the CDN network to each target proxy server s. OmegabDeviation optimization parameter J for representing user bandwidth satisfaction degreebOptimizing parameter J with respect to user average response timedWeight value of, ωpRepresents the stability optimization parameter JpOptimizing parameter J with respect to user average response timedThe weight value of (2).
Formula (2) defines the user average response time optimization parameter, d (T) represents the user average response time in T time, H represents the prediction time domain, and T represents the current time. Wherein, the average user response time comprises the link delay of the round trip and the processing time of the server. The average response time d (t) of the user in the time t in the formula (2) satisfies the relationship in the formula (7):
in the formula (7), deAnd representing the link delay of the link e in the path i from the inlet switch of the CDN network to the target proxy server s collected by the SDN controller.
Wherein, all proxy servers can be abstracted into M/M/1 queuing model, so that each proxy server in CDN processes time ds(t) satisfies the relationship as described in equation (8):
in the formula (8), λtRepresenting the average rate, λ, at which a user's content request reaches the target proxy server s during time tsRepresenting the processing rate, p, of the target proxy server ss(t) represents the proportion of content requests that each proxy server of the CDN network is assigned to a user during time t.
The user bandwidth satisfaction degree refers to the ratio of the bandwidth actually provided by the network to the bandwidth requested by the user, and the user bandwidth satisfaction degree deviation degree BL on the link ee(t) satisfies the relationship as described in the formula (3):
in formula (3), BLe(t) represents the user bandwidth satisfaction degree deviation degree of a link e in a path i from an inlet switch of the CDN network to a target proxy server within t time, pi,s(t) represents the proportion of user requests from the selected path i to the server s in the time t, V represents the average user request size, RtRepresents the total number of user requests in the CDN network within t time, CeRepresenting the capacity of said link e. [ x ] of]+It means that when x > 0, the value is x itself, and when x ≦ 0, the value is 0.
Considering that a path includes multiple links, the bandwidth satisfaction degree offset of the path i takes the value of the largest link among the multiple links due to the "barrel effect", that is:
in the formula (4), BPi(t) denotes C during t timeAnd the user bandwidth satisfaction degree of the path i from the entry switch of the DN network to the target proxy server is offset.
Thereby optimizing parameter J of user bandwidth satisfaction degree offset degreebSatisfies the relationship as described in equation (5):
finally, since the adjacent input varies too much, which tends to cause system instability, it is considered to reduce the system input pi,s(t) to maintain the stability of the system, where the stability optimization parameter J is definedpComprises the following steps:
in the formula (6), JpDenotes the stability optimization parameter, IsRepresenting the set of paths from the ingress switches of the CDN network to each target proxy server s.
Solving the nonlinear programming problem by using the formula (1) can calculate the proportion p of content requests distributed to users by each proxy server of the CDN network within t times(t) and the proportion p of content requests for users for which a path i is allocated to a target proxy server s during time ti,sAnd (t), thereby effectively reducing the average response time of the user, improving the bandwidth satisfaction degree of the user and ensuring the stability of the system.
Meanwhile, the parameter J can be optimized by adjusting the degree of deviation of the user bandwidth satisfaction degree according to the actual requirement of the networkbOptimizing parameter J with respect to user average response timedWeight value of omegabAnd stability optimization parameter JpOptimizing parameter J with respect to user average response timedWeight value of omegapTo balance the weight among three optimization parameters of the average response time of the user, the degree of deviation of the satisfaction degree of the user bandwidth and the stability of the systemNature is important.
In the SDN-based CDN network user request allocation method, the intelligent center is used as a global brain, the constraint of limited processing capacity of a single SDN controller is avoided, and some decisions with large calculation amount and complex calculation are transferred to the decision module of the intelligent center, so that the pressure of the SDN controller is reduced. Meanwhile, the SDN controller collects the information of the CDN in real time through the OpenFlow interface and uploads the collected information to the database of the intelligent center, so that a decision module of the intelligent center makes a decision according to the information received by the database, and the decision of the CDN across control domains can be supported.
In addition, the decision module of the intelligent center further adopts a user request allocation algorithm based on the MPC, so that the average response time of the user can be effectively reduced, the bandwidth satisfaction degree of the user is improved, and the stability of the system is ensured.

Claims (3)

1. A user request distribution method of a CDN network based on an SDN is characterized by comprising the following steps:
step one, each SDN controller collects CDN network information in real time through an OpenFlow interface, wherein the CDN network information comprises network global topology, link time delay and flow, and uploads the collected CDN network information to a database of an intelligent center;
step two, a decision module of the intelligent center calculates the proportion of content requests distributed to users by each proxy server of the CDN by adopting a dynamic distribution algorithm according to the CDN network information received by the database, calculates the proportion of the content requests distributed to the users by paths from an inlet switch of the CDN to each proxy server, and sends the IP addresses of the proxy servers, the proportion of the content requests distributed to the users by each proxy server of the CDN network and the proportion of the content requests distributed to the users by paths from the inlet switch of the CDN network to each proxy server to the SDN controller;
modifying a flow table entry of an ingress switch by the SDN controller according to the received IP address of the proxy server, the proportion of content requests distributed to users by each proxy server of the CDN network and the proportion of content requests distributed to users by paths from the ingress switch of the CDN network to each proxy server;
step four, the user sends a content request to the entrance switch, the entrance switch matches the content request with the flow table item and forwards the matched content request to the target proxy server, the user obtains the requested content from the target proxy server, and the proportion of the content request distributed to the user by the target proxy server is not 0.
2. The SDN-based CDN network user request distribution method of claim 1, wherein: calculating the proportion of content requests distributed to users by each proxy server of the CDN network and the proportion of content requests distributed to users by the path from an entrance switch of the CDN network to each proxy server by using a formula (1):
m i n J d + ω b J b + ω p J p s . t . p s ( t ) = Σ i ∈ I s p i , s ( t ) 0 ≤ p i , s ( t ) ≤ 1 , ∀ i , ∀ s 0 ≤ p s ( t ) ≤ 1 , ∀ s Σ i ∈ I s Σ s ∈ S p i , s ( t ) = 1 - - - ( 1 )
wherein,
J d = Σ t = T + 1 T + H d ( t ) - - - ( 2 )
BL e ( t ) = [ Σ i ∋ e p i , s ( t ) VR t - C e Σ i ∋ e p i , s ( t ) VR t ] + - - - ( 3 )
BP i ( t ) = m a x e ∈ i BL e ( t ) - - - ( 4 )
J b = Σ t = T + 1 T + H p i , s ( t ) BP i ( t ) - - - ( 5 )
J p = Σ t = T + 1 T + H Σ s ∈ S Σ i ∈ I s | p i , s ( t ) - p i , s ( t - 1 ) | - - - ( 6 )
in the formulas (1) - (6), JdRepresents a user average response time optimization parameter, JbRepresents a user bandwidth satisfaction degree offset optimization parameter, JpRepresenting a stability optimization parameter, ωbDeviation optimization parameter J for representing user bandwidth satisfaction degreebOptimizing parameter J with respect to user average response timedWeight value of, ωpRepresents the stability optimization parameter JpOptimizing parameter J with respect to user average response timedWeighted value of ps(t) represents the proportion of content requests assigned to users by each proxy server of the CDN network during t, i represents the path from the ingress switch of the CDN network to the target proxy server s, pi,s(t) represents the proportion of content requests for users whose path i is assigned to target proxy server s during time t, s representing a single target proxy server; s denotes a set of target proxy servers, IsRepresenting a set of paths from an inlet switch of the CDN network to each target proxy server s, d (T) representing the average user response time in T time, H representing a prediction time domain, T representing the current time, s representing a single target proxy server, and e representing a link in a path i from the inlet switch of the CDN network to the target proxy server s; BLe(t) represents the user bandwidth satisfaction degree deviation degree of a link e in a path i from an inlet switch of the CDN network to a target proxy server s in t time; v represents a mathematical average of the user's content request; rtRepresenting the total number of content requests of users in the CDN network during time t; ceIndicating the capacity, BP, of link eiAnd (t) represents the user bandwidth satisfaction degree deviation degree of the path from the inlet switch of the CDN network to each target proxy server in the time t.
3. The SDN-based CDN network user request distribution method of claim 2, wherein: the average response time of the user in the time t is as follows:
d ( t ) = Σ s ∈ S d s ( t ) p s ( t ) + Σ e ∈ i , i ∈ I d e p i , s ( t ) - - - ( 7 )
wherein,
d s ( t ) = Σ s ∈ S λ t p s ( t ) λ s - λ t p s ( t ) - - - ( 8 )
in equations (7) and (8), d (t) represents the average response time of the user during t time, ds(t) represents the processing time of each proxy server in the CDN network, deRepresents the link delay, lambda, of the link e in the path i from the ingress switch of the CDN network to the target proxy server s collected by the SDN controllertRepresenting the average rate, λ, at which a user's content request reaches the target proxy server s during time tsRepresenting the processing rate, p, of the target proxy server ss(t) represents the proportion of content requests that each proxy server of the CDN network is assigned to a user during time t.
CN201710192747.5A 2017-03-28 2017-03-28 A kind of user's request distribution method of the CDN based on SDN Pending CN106850859A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710192747.5A CN106850859A (en) 2017-03-28 2017-03-28 A kind of user's request distribution method of the CDN based on SDN

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710192747.5A CN106850859A (en) 2017-03-28 2017-03-28 A kind of user's request distribution method of the CDN based on SDN

Publications (1)

Publication Number Publication Date
CN106850859A true CN106850859A (en) 2017-06-13

Family

ID=59141531

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710192747.5A Pending CN106850859A (en) 2017-03-28 2017-03-28 A kind of user's request distribution method of the CDN based on SDN

Country Status (1)

Country Link
CN (1) CN106850859A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108769097A (en) * 2018-03-30 2018-11-06 中国科学院信息工程研究所 Support the content distribution network system of network-control
CN109561057A (en) * 2017-09-27 2019-04-02 中兴通讯股份有限公司 A kind of content distributing network entities business processing method, apparatus and system
CN110022373A (en) * 2019-04-17 2019-07-16 北京达佳互联信息技术有限公司 Method for distributing business, device, server and storage medium
CN112769590A (en) * 2020-11-16 2021-05-07 国网河北省电力有限公司信息通信分公司 Software defined content distribution network server selection mode based on block chain

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103795805A (en) * 2014-02-27 2014-05-14 中国科学技术大学苏州研究院 Distributed server load balancing method based on SDN
CN105634992A (en) * 2015-12-29 2016-06-01 网宿科技股份有限公司 CDN platform self-adaptive bandwidth control method and system

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103795805A (en) * 2014-02-27 2014-05-14 中国科学技术大学苏州研究院 Distributed server load balancing method based on SDN
CN105634992A (en) * 2015-12-29 2016-06-01 网宿科技股份有限公司 CDN platform self-adaptive bandwidth control method and system

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
FENG QIN等: "《Optimizing Routing and Server Selection in Intelligent SDN-based CDN》", 《IEEE》 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109561057A (en) * 2017-09-27 2019-04-02 中兴通讯股份有限公司 A kind of content distributing network entities business processing method, apparatus and system
CN108769097A (en) * 2018-03-30 2018-11-06 中国科学院信息工程研究所 Support the content distribution network system of network-control
CN110022373A (en) * 2019-04-17 2019-07-16 北京达佳互联信息技术有限公司 Method for distributing business, device, server and storage medium
CN112769590A (en) * 2020-11-16 2021-05-07 国网河北省电力有限公司信息通信分公司 Software defined content distribution network server selection mode based on block chain
CN112769590B (en) * 2020-11-16 2023-09-08 国网河北省电力有限公司信息通信分公司 Software defined content distribution network server selection method based on blockchain

Similar Documents

Publication Publication Date Title
EP3382963B1 (en) Method and system for self-adaptive bandwidth control for cdn platform
Huang et al. Buffer state is enough: Simplifying the design of QoE-aware HTTP adaptive video streaming
CN108540384B (en) Intelligent rerouting method and device based on congestion awareness in software defined network
CN109714275B (en) SDN controller for access service transmission and control method thereof
CN106850859A (en) A kind of user's request distribution method of the CDN based on SDN
CN108449265B (en) Path calculation method and device in SD-WAN environment
CN112486690B (en) Edge computing resource allocation method suitable for industrial Internet of things
CN113114582B (en) Link congestion fault prediction and network autonomous control method based on machine learning
JP2010538581A (en) Method and system for aggregate bandwidth control
de Morais et al. Application of active queue management for real-time adaptive video streaming
WO2016132933A1 (en) Design device, design method, and recording medium
CN116599904A (en) Parallel transmission load balancing device and method
Hoßfeld et al. Can context monitoring improve QoE? A case study of video flash crowds in the internet of services
Tuncer et al. Scalable cache management for ISP-operated content delivery services
CN110535770B (en) QoS-aware-based intelligent routing method for video stream in SDN environment
Ma et al. Learning-based joint QoE optimization for adaptive video streaming based on smart edge
Ahmad et al. Timber: An SDN based emulation platform for QoE management experimental research
De Cicco et al. A resource allocation controller for cloud-based adaptive video streaming
JP7067170B2 (en) Route control system, route control method, and program
Bhattacharyya et al. QFlow: A learning approach to high QoE video streaming at the wireless edge
CN115550233A (en) Distributed route determination method, electronic device, and storage medium
Sina et al. CaR-PLive: Cloud-assisted reinforcement learning based P2P live video streaming: a hybrid approach
CN110191362B (en) Data transmission method and device, storage medium and electronic equipment
Sayit et al. DASH-QoS: A scalable network layer service differentiation architecture for DASH over SDN
US20090116495A1 (en) Method and Device for Dynamic Management of 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
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20170613

WD01 Invention patent application deemed withdrawn after publication