CN107124306A - Content delivery network server optimization dispositions method under network function virtualized environment - Google Patents
Content delivery network server optimization dispositions method under network function virtualized environment Download PDFInfo
- Publication number
- CN107124306A CN107124306A CN201710270020.4A CN201710270020A CN107124306A CN 107124306 A CN107124306 A CN 107124306A CN 201710270020 A CN201710270020 A CN 201710270020A CN 107124306 A CN107124306 A CN 107124306A
- Authority
- CN
- China
- Prior art keywords
- node
- vector
- matrix
- network
- cost
- 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.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/142—Network analysis or design using statistical or mathematical methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/45595—Network integration; Enabling network access in virtual machine instances
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Probability & Statistics with Applications (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Mathematical Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Algebra (AREA)
- Evolutionary Biology (AREA)
- Mathematical Optimization (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The present invention is directed to the defect of existing CDN Replica placements scheme, it is proposed that the content delivery network server optimization dispositions method under a kind of network function virtualized environment.The present invention utilizes spectral clustering model, has taken into account the data and topological property of physical network.Adjacency matrix is set up using the topological relation of bottom-layer network and the traffic demand of node, physical network nodes are divided into several set using spectral clustering.Then the node in each set is traveled through, it is calculated as copy Centroid and carrys out the cost for needed for the node of remaining in set provides content service.The node of cost minimization is chosen, makes it as the placement node of replica server, cost is reduced with this.
Description
Technical field
The invention belongs to network technique field, and in particular to the content delivery network service under network function virtualized environment
Device Optimization deployment method.
Background technology
With the fast development of the development of network technology, especially development of Mobile Internet technology so that Web content is from word
The propagation of image content gradually becomes visual form propagation.This causes network traffics to show explosive growth, increases network
The possibility got congestion.The congestion of network causes the experience of user to decline, so that content supplier have lost substantial amounts of visitor
Family, profit is reduced.
And CDN (content delivery network) development so that copy of content server can be deployed in the side nearer from user
On hoddy network so that user can obtain Web content in the short period of time, add Consumer's Experience.Shown in such as figure one (a)
For legacy network, three parts are broadly divided into：Server, communication link and terminal user.In traditional network, terminal user
Perhaps server provides the user service in request server, and the communication link to be passed through is longer.When terminal user is more
Wait, the flow on network is larger, easily causes congestion, while only one of which server is user service, be likely to cause
Server load is excessive to cause the machine of delaying.Shown in such as figure one (b) in CDN, main part has four：Server, communication
Link, replica server and terminal user.Wherein server is to provide the source server that content or content update are deleted, and owns
Web content all enter network from this server flows.And be directly connected with this server is replica server, and Copy Service
The position of device is closer to from terminal user.And the content needed for user directly provides clothes by nearer replica server
Business, content is asked for without reaching original server across the substantial amounts of network equipment.Therefore it can improve service quality, reduce user
Etc. it is to be delayed.And a key issue is exactly the Placement Problems of CDN content replica server in CDN technologies, existing copy
Placement is all based on special CDN physical servers to be disposed.And special server needs to expend substantial amounts of people
Power is disposed.The maintenance and renewal of equipment need substantial amounts of manpower and materials simultaneously.If old equipment processing is bad, it can also make
Into environmental pollution.Existing CDN replica servers Placement has obvious defect simultaneously, it is impossible to take into account the topology of bottom-layer network
Disposed with the request data on flows of user.Therefore it and can not effectively reduce core network flow, improve Consumer's Experience.
In the research of existing CDN Replica placements algorithm, a kind of is to be abstracted into the Placement Problems of CDN replica servers
Facility locating problem.Then placed using random algorithm.Concretely comprise the following steps：The thing of CDN is undertaken the construction of in determination first
Topological structure is managed, the bandwidth between the traffic demand and different nodes of each node is determined；Then in all physical nodes with
Machine chooses appropriate node to place replica server, calculates each node to optimal (closest or load balancing isometry)
Copy center cost summation.Repeat the above steps 10 times, take that minimum time of cost summation as final solution.
Theoretical foundation is proved, is being repeated 10 times left and right, the result of random algorithm is relatively stable, and effect is preferable.But still exist following
It is not enough：(1) random algorithm has larger fluctuation, and produced solution has significant limitation.(2) random algorithm
It is insensitive to the topological structure of physical network, it is impossible to be adjusted according to different network topology structures, the topology aware of algorithm
Property is poor.(3) random algorithm is insensitive to the node flow demand of physical network, and bandwidth resources, it is impossible to according to actual
Bottom-layer network demand is adjusted, and the data sensitive of algorithm is poor.
Another mode to CDN Replica placements is：CDN Replica placement schemes based on greedy algorithm, utilize iteration
Thought reduces the complexity of Placement.The detailed process of algorithm is as follows：All physical nodes are traveled through, each physics section is calculated
The cost summation that point is consumed as copy Centroid for other node serves.Choose the minimum node of cost summation and be used as pair
This Centroid.Iteration said process, until all nodes place completion.The shortcoming of the placement schemes is embodied in：(1) covet
Center algorithm is easily ensnared into locally optimal solution.Because the copy center first placed has influence, institute to the copy center of rear placement
It is of overall importance optimal to hardly result in.(2) greedy algorithm is sensitive to the data of bottom-layer network, but to the topology of bottom-layer network
Sensitiveness is relatively low.
The content of the invention
The goal of the invention of the present invention is：For the defect of existing CDN Replica placements scheme, it is proposed that a kind of NFV (networks
Virtualization of function, Network Function Virtualization) CDN replica server laying method under environment.
The present invention utilizes spectral clustering model, has taken into account the data and topological property of physical network.Using the topological relation of bottom-layer network with
And the traffic demand of node sets up adjacency matrix, physical network nodes are divided into several set using spectral clustering.Then
Node in each set of traversal, calculates it and comes as copy Centroid for needed for the node of remaining in set provides content service
Cost.The node of cost minimization is chosen, makes it as the placement node of replica server, cost is reduced with this.
Content delivery network server optimization dispositions method under the network function virtualized environment of the present invention, including it is following
Step：Step 1：Based on the physical topological structure for undertaking the construction of content delivery network, the server in physical node is regard as virtual section
Point, it is determined that virtual network topology to be disposed；
Step 2：Similarity matrix W is calculated according to the node flow demand of virtual network topology and dummy node：
Calculate the similarity w between any two dummy nodeij：If there is link connection (i.e. empty between two dummy nodes
Intend depositing direct-connected relation between node), then similarity wij=(| di-dj|)2.5,i≠j；Otherwise, similarity wij=0；Wherein i, j are
Virtual node identifiers, di、djRepresent the node flow demand of different dummy nodes；
By similarity wijObtain similarity matrixN represents dummy node number；
Step 3：According to formula D=dia (dr1,dr2,...,drn) degree matrix D is obtained, La Pula is obtained by L=D-W
This matrix L, wherein dia are diagonal symbol, diagonal element
And Laplacian Matrix L is normalized, obtain normalized Laplacian Matrix Lsym：Lysm=D-1/ 2LD-1/2；
Step 4：Calculate LysmN characteristic value and characteristic vector, by the corresponding characteristic vector of preceding k minimal eigenvalue
u1,...,ukBy row arrangement form matrix Un×k=(u1,...,uk), wherein k represents replica server number to be disposed；
To matrix Un×kMatrix T=(t are obtained by row normalizationij)n×k, wherein matrix element
Vector y is obtained by row to matrix T1,y2,…,yn, wherein vector yiSubscript i homographies T the i-th row, and i=
1,…,n；
Step 5：To y1,...,ynK mean cluster processing is carried out, k cluster result is obtained for C1,C2,...,CkIt is (wherein every
Element in individual set is the subscript value for representing y vectors), the virtual node identifiers included by each cluster result obtain k
Individual node clustering set Am, m=1 ..., k, i.e. Am={ i:i∈Ci, m=1 ..., k；
Such as：Vector y in cluster result1,y4,y7∈C1, then set A is clustered1By dummy node v1,v4,v7Composition.
Step 6：For each node clustering set Am, travel through AmIn each dummy node vr, calculate dummy node vrMake
Centered on node cost summationWherein cirRepresent from node viTo node vrLinkage length,
That is node viTo node vrHop count, p be unit bandwidth cost；The minimum dummy nodes of cost summation cost are chosen as each
AmCopy Centroid.
By adopting the above-described technical solution, the beneficial effects of the invention are as follows：
(1) lower deployment cost is reduced.The present invention has taken into full account the topological sum data message of network.Can opening up according to network
Flutter effectively select rational node to be used as copy Centroid with data message take into account topological sum traffic demand information from
And reduce cost.
(2) service delay is reduced.The present invention is split network topology using spectral clustering, more can sufficiently be examined
Consider the lag characteristic of network, therefore the selection at copy center can be distributed to node closer to the user and prolonged with reducing network
Late.
(3) core network flow is reduced.Due to the present invention more can effectively disperse copy center choose position to from
Family closer to node on, so traffic demand between CDN and user subtracts only at copy center to being transmitted back and forth between user
The small flow of core network, so that the robustness of extra-high network reduces the possibility of congestion generation.
Brief description of the drawings
Fig. 1 is legacy network, CDN illustraton of model.
Fig. 2 is the virtualization process schematic diagram of the legend network of the present invention.
Fig. 3 is the clustering processing schematic flow sheet of the present invention.
Embodiment
To make the object, technical solutions and advantages of the present invention clearer, with reference to embodiment and accompanying drawing, to this hair
It is bright to be described in further detail.
The implementation bag of content delivery network server optimization dispositions method under the network function virtualized environment of the present invention
Include the CDN modelling under NFV environment, spectral clustering and divide subgraph and the selection of copy center, be specially：
(1) the CDN model under NFV environment
Fig. 2 (a) show actual physical network topology, includes general server, large-scale data center and cloud data
The functional node such as the calculate nodes such as center and router switch.In network virtualization (NV) environment, the one of physical network
Computing resource and bandwidth resources can virtually be turned to by cutting node and bandwidth, shown in such as Fig. 2 (b).And NFV technologies can be in difference
Calculate node deployment identical network function software or dispose different network function softwares in same calculate node.
In this patent, we mainly dispose identical network function in different calculate nodes.Therefore former problem is in the thing of bottom
Deployment replica server is changed into abstract upper deployment replica server on reason network.Bottom can be masked not by so doing
Same network entity, and it is absorbed in the selection of replica server.
(2) spectral clustering divides subgraph
G=(V, E) is made to represent the network topological diagram in Fig. 1 (b).Wherein V represents node clustering set, and E represents link knot
Close.d1,d2,...,dnThe traffic demand of each node in node clustering set is represented, subscript n represents total node number.If
Have link connection between node and node, using formula (1) calculate similarity, otherwise similarity be set to 0.
wij=(| di-dj|)2.5,(i≠j) (1)
The adjacency matrix for the similarity asked using formula (2)
Matrix D=dia (dr1,dr2,...,drn) it is degree matrix, wherein dri(i=1 ..., n) obtained by formula (3)
Dia is diagonal symbol, represents diagonal matrix, remaining matrix entries is zero.
Laplacian Matrix L is tried to achieve using formula (4)：
L=D-W (4)
Laplacian Matrix L is tried to achieve using formula (5)sym
Lysm=D-1/2LD-1/2 (5)
L is tried to achieve using matlabysmAll eigenvalue λs1,λ2,...,λnWherein λ1≤λ2≤...λnAnd its corresponding spy
Levy vectorial u1,u2,...,un(characteristic vector is column vector ui=(u1,i,u2,i,...,un,i)T).Take eigenvalue λ1,...,λk(k
For the replica server number to be placed, this value should be less than n) corresponding characteristic vector u1,...,uk.This k characteristic vector is pressed
Row arrangement form matrix Un×k=(u1,...,uk).Matrix U is sought by the normalized matrix T=(t of row gainedij)n×k, wherein matrix
Element tijTried to achieve according to formula (6)
Vector y is obtained by row to matrix T1,y2,...,yi,...,yn, wherein vector subscript i homographies T the i-th row.
Using k-mean algorithms to y1,...,ynClustered.
Wherein k-mean algorithms the step of it is following in n vector, randomly select first k it is vectorial as center.
2. other non-central vectors are traveled through, non-central vector is calculated to the distance of k center vector according to formula (7)
3. to each non-central vector, by this non-central vector with dividing a class into apart from minimum center vector.Gathered
Class result C1',...,C'k, wherein the element in each set is the subscript value for representing y vectors.
To each cluster result, recalculate a center vector, this vector into cluster result other nodes away from
It is minimum from sum.Therefore k new center vectors have been obtained.
5. repeat step 2.-, the centre-to-centre spacing deviation until k center newly updating and before updating is less than.
6. k-means clustering procedures are obtained to vectorial y1,...,ynCluster result be C1,C2,...,Ck(wherein each set
In element be represent y vector subscript value)
After k-means algorithms, we have obtained cluster result C1,...,Ck, wherein element in each set also with
The subscript of network topology G=(V, E) interior joint is corresponding.The cluster set of network topology, i.e. node is tried to achieve using formula (8) to gather
Class set A1,A2,...,Ak：
Am={ i:i∈Ci, m=1 ..., k (8)
So far, the cluster of figure terminates, and we have obtained different cluster set, and A1,...,Ak.Next will be
A suitable node is chosen in each cluster set as copy Centroid to come for every other demand nodes service.
(3) copy center is chosen
For each cluster set, each node of traversal cluster set, it is other node serves in set to calculate it
Cost summation cost, specific calculate see formula (9)：
diRepresent cluster set AmInterior joint viTraffic demand, vrRepresent, cirRepresent from node viTo node vjLink
Length (represents) that p is unit bandwidth cost with hop count.
The minimum nodes of cost summation cost are chosen as the copy Centroid of each cluster set.
To sum up, of the invention be combined the Placement Problems of virtual CDN replica server with clustering problem more can be effective
The virtual CDN of solution placement.
Claims (3)
1. the content delivery network server optimization dispositions method under network function virtualized environment, it is characterised in that including under
Row step：Step 1：Based on the physical topological structure for undertaking the construction of content delivery network, using the server in physical node as virtual
Node, it is determined that virtual network topology to be disposed；
Step 2：Similarity matrix W is calculated according to the node flow demand of virtual network topology and dummy node：
Calculate the similarity w between any two dummy nodeij：If having link connection, similarity between two dummy nodes
wij=(| di-dj|)2.5,i≠j；Otherwise, similarity wij=0；Wherein i, j are virtual node identifiers, di、djRepresent different void
Intend the node flow demand of node；
By similarity wijObtain similarity matrixN represents dummy node number；
Step 3：According to formula D=dia (dr1,dr2,...,drn) degree matrix D is obtained, Laplce's square is obtained by L=D-W
Battle array L, wherein dia are diagonal symbol, diagonal element
And Laplacian Matrix L is normalized, obtain normalized Laplacian Matrix Lsym：Lysm=D-1/2LD-1/2；
Step 4：Calculate LysmN characteristic value and characteristic vector, by the corresponding characteristic vector u of preceding k minimal eigenvalue1,...,
ukBy row arrangement form matrix Un×k=(u1,...,uk), wherein k represents replica server number to be disposed；
To matrix Un×kMatrix T=(t are obtained by row normalizationij)n×k, wherein matrix element
Vector y is obtained by row to matrix T1,y2,…,yn, wherein vector yiSubscript i homographies T the i-th row, and i=1 ...,
n；
Step 5：To y1,...,ynK mean cluster processing is carried out, k cluster result is obtained for C1,C2,...,Ck, by each cluster
As a result the virtual node identifiers included obtain k node clustering set Am, m=1 ..., k；
Step 6：For each node clustering set Am, travel through AmIn each dummy node vr, calculate dummy node vrIn
The cost summation of heart nodeWherein cirRepresent from node viTo node vrLinkage length, that is, save
Point viTo node vrHop count, p be unit bandwidth cost；
Choose the minimum dummy nodes of cost summation cost and be used as each node clustering set AmCopy Centroid.
2. the method as described in claim 1, it is characterised in that in step 5, to y1,...,ynCarry out k mean clusters processing tool
Body is：
1. in n vector y1,...,ynIn, k vector is randomly selected as center vector；
2. in vectorial y1,...,ynIn, non-central vector is calculated to the distance of k center vector
3. to each non-central vector, divide its center vector minimum with distance into a class, obtain cluster result C1',...,
C'k, wherein the element in each cluster result represents yiSubscript value；
4. to each cluster result C'm, recalculate a center vector, the center vector to cluster result C'mIn it is non-in
Heart vector it is minimum apart from sum, obtain k new center vectors；
5. repeat step 2. -4., until k center vector newly updating with the range difference of center vector before updating is less than threshold epsilon.
3. method as claimed in claim 2, it is characterised in that the value of threshold epsilon is ε≤0.001.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710270020.4A CN107124306B (en) | 2017-04-24 | 2017-04-24 | Content delivery network server optimization dispositions method under network function virtualized environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710270020.4A CN107124306B (en) | 2017-04-24 | 2017-04-24 | Content delivery network server optimization dispositions method under network function virtualized environment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107124306A true CN107124306A (en) | 2017-09-01 |
| CN107124306B CN107124306B (en) | 2019-11-05 |
Family
ID=59725429
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710270020.4A Active CN107124306B (en) | 2017-04-24 | 2017-04-24 | Content delivery network server optimization dispositions method under network function virtualized environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107124306B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107749801A (en) * | 2017-09-28 | 2018-03-02 | 西南交通大学 | A kind of virtual network function laying method based on population Incremental Learning Algorithm |
| CN108156032A (en) * | 2017-12-22 | 2018-06-12 | 中国人民解放军战略支援部队信息工程大学 | The reference mode choosing method combined based on spectral clustering with random selection |
| CN110290165A (en) * | 2019-04-04 | 2019-09-27 | 平安科技（深圳）有限公司 | Traffic load regulation method, electronic device and readable storage medium storing program for executing between network host |
| CN110677306A (en) * | 2019-10-25 | 2020-01-10 | 上海交通大学 | Network topology replica server configuration method and device, storage medium and terminal |
| WO2020027743A1 (en) * | 2018-08-03 | 2020-02-06 | Medianova Internet Hizmetleri Ve Ticaret Anonim Sirketi | System used by cdn companies to improve the quality offered to the users and to optimize resource utilization |
| CN111181788A (en) * | 2019-12-31 | 2020-05-19 | 江苏省未来网络创新研究院 | SDN intelligent system, working method and remote server |
| CN111475250A (en) * | 2019-01-24 | 2020-07-31 | 阿里巴巴集团控股有限公司 | Network optimization method and device in cloud environment |
| CN115442229A (en) * | 2021-06-04 | 2022-12-06 | 中国移动通信集团浙江有限公司 | Method, device, storage medium and apparatus for networking communication core network |
| US20230221946A1 (en) * | 2020-09-22 | 2023-07-13 | Cisco Technology, Inc. | Identifying Execution Environments for Deploying Network Functions |
| CN115941992B (en) * | 2022-12-06 | 2023-11-03 | 南京奥看信息科技有限公司 | Channel condition-based cache-enabled multi-quality video distribution method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103179052A (en) * | 2011-12-20 | 2013-06-26 | 中国科学院声学研究所 | Virtual resource allocation method and system based on proximity centrality |
| CN103457852A (en) * | 2013-09-13 | 2013-12-18 | 电子科技大学 | Invulnerability mapping method of multicast virtual network |
| CN105262663A (en) * | 2015-08-19 | 2016-01-20 | 电子科技大学 | Cross-domain mapping method for hybrid virtual network (HVN) |
| CN105407080A (en) * | 2015-10-22 | 2016-03-16 | 华为技术有限公司 | Method and device used for making virtual machine disposition strategy |
-
2017
- 2017-04-24 CN CN201710270020.4A patent/CN107124306B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103179052A (en) * | 2011-12-20 | 2013-06-26 | 中国科学院声学研究所 | Virtual resource allocation method and system based on proximity centrality |
| CN103457852A (en) * | 2013-09-13 | 2013-12-18 | 电子科技大学 | Invulnerability mapping method of multicast virtual network |
| CN105262663A (en) * | 2015-08-19 | 2016-01-20 | 电子科技大学 | Cross-domain mapping method for hybrid virtual network (HVN) |
| CN105407080A (en) * | 2015-10-22 | 2016-03-16 | 华为技术有限公司 | Method and device used for making virtual machine disposition strategy |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107749801B (en) * | 2017-09-28 | 2019-09-06 | 西南交通大学 | A kind of virtual network function laying method based on population Incremental Learning Algorithm |
| CN107749801A (en) * | 2017-09-28 | 2018-03-02 | 西南交通大学 | A kind of virtual network function laying method based on population Incremental Learning Algorithm |
| CN108156032A (en) * | 2017-12-22 | 2018-06-12 | 中国人民解放军战略支援部队信息工程大学 | The reference mode choosing method combined based on spectral clustering with random selection |
| CN108156032B (en) * | 2017-12-22 | 2020-11-24 | 中国人民解放军战略支援部队信息工程大学 | Reference node selection method based on combination of spectral clustering and random selection |
| WO2020027743A1 (en) * | 2018-08-03 | 2020-02-06 | Medianova Internet Hizmetleri Ve Ticaret Anonim Sirketi | System used by cdn companies to improve the quality offered to the users and to optimize resource utilization |
| CN111475250B (en) * | 2019-01-24 | 2023-05-26 | 阿里巴巴集团控股有限公司 | Network optimization method and device in cloud environment |
| CN111475250A (en) * | 2019-01-24 | 2020-07-31 | 阿里巴巴集团控股有限公司 | Network optimization method and device in cloud environment |
| CN110290165B (en) * | 2019-04-04 | 2022-01-28 | 平安科技（深圳）有限公司 | Method for regulating and controlling communication load between network hosts, electronic device and readable storage medium |
| CN110290165A (en) * | 2019-04-04 | 2019-09-27 | 平安科技（深圳）有限公司 | Traffic load regulation method, electronic device and readable storage medium storing program for executing between network host |
| CN110677306A (en) * | 2019-10-25 | 2020-01-10 | 上海交通大学 | Network topology replica server configuration method and device, storage medium and terminal |
| CN110677306B (en) * | 2019-10-25 | 2021-09-03 | 上海交通大学 | Network topology replica server configuration method and device, storage medium and terminal |
| CN111181788A (en) * | 2019-12-31 | 2020-05-19 | 江苏省未来网络创新研究院 | SDN intelligent system, working method and remote server |
| US20230221946A1 (en) * | 2020-09-22 | 2023-07-13 | Cisco Technology, Inc. | Identifying Execution Environments for Deploying Network Functions |
| CN115442229A (en) * | 2021-06-04 | 2022-12-06 | 中国移动通信集团浙江有限公司 | Method, device, storage medium and apparatus for networking communication core network |
| CN115442229B (en) * | 2021-06-04 | 2023-09-22 | 中国移动通信集团浙江有限公司 | Communication core network networking method, equipment, storage medium and device |
| CN115941992B (en) * | 2022-12-06 | 2023-11-03 | 南京奥看信息科技有限公司 | Channel condition-based cache-enabled multi-quality video distribution method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107124306B (en) | 2019-11-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107124306A (en) | Content delivery network server optimization dispositions method under network function virtualized environment | |
| CN104468688B (en) | Method and apparatus for network virtualization | |
| CN102724063B (en) | Log acquisition server and packet delivery, Log Clustering method and network | |
| Dong et al. | Energy-performance tradeoffs in IaaS cloud with virtual machine scheduling | |
| Raghavendra et al. | Dynamic graph query primitives for sdn-based cloudnetwork management | |
| CN107111513A (en) | The InfiniBand system and method for SR IOV vSWITCH architectural frameworks are provided for high-performance cloud computing environment | |
| CN103514245B (en) | Creation user it is visible processing tracking can search for and global data base | |
| CN107710159A (en) | System and method for the efficient virtualization in lossless network | |
| Hu et al. | An Immune Cooperative Particle Swarm Optimization Algorithm for Fault‐Tolerant Routing Optimization in Heterogeneous Wireless Sensor Networks | |
| Pandurangan et al. | DEX: self-healing expanders | |
| Pu et al. | Information transport in multiplex networks | |
| CN108600101A (en) | A kind of network for the optimization of end-to-end time delay performance services cross-domain method of combination | |
| CN106953848A (en) | A kind of software defined network implementation method based on ForCES | |
| CN102006234A (en) | Method and device for optimizing P2P (Peer-to-Peer) network flow | |
| CN109995580A (en) | VN mapping method based on GA_PSO hybrid algorithm in 5G network slice | |
| Singh et al. | Efficient content retrieval in fog zone using Nano‐Caches | |
| Zhang et al. | Method to enhance traffic capacity for two-layer complex networks | |
| CN103281211A (en) | Large-scale network node grouping management system and management method | |
| Bouzidi et al. | Dynamic clustering of software defined network switches and controller placement using deep reinforcement learning | |
| US20200099735A1 (en) | System and method for mesh network streaming | |
| Liu et al. | Optimizing multicast routing tree on application layer via an encoding-free non-dominated sorting genetic algorithm | |
| Huang et al. | A topology-cognitive algorithm framework for virtual network embedding problem | |
| Han et al. | Identifying top-k influential nodes based on discrete particle swarm optimization with local neighborhood degree centrality | |
| Wang et al. | Cost-driven data caching in edge-based content delivery networks | |
| Mokhtarian et al. | Minimum-delay multicast algorithms for mesh overlays |
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 |