CN105187322B - A kind of virtual net mapping method based on load balancing - Google Patents
A kind of virtual net mapping method based on load balancing Download PDFInfo
- Publication number
- CN105187322B CN105187322B CN201510604413.5A CN201510604413A CN105187322B CN 105187322 B CN105187322 B CN 105187322B CN 201510604413 A CN201510604413 A CN 201510604413A CN 105187322 B CN105187322 B CN 105187322B
- Authority
- CN
- China
- Prior art keywords
- node
- load
- dimension
- load balancing
- link
- 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
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
- H04L47/10—Flow control; Congestion control
- H04L47/12—Avoiding congestion; Recovering from congestion
- H04L47/125—Avoiding congestion; Recovering from congestion by balancing the load, e.g. traffic engineering
-
- 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
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
Abstract
The invention discloses a kind of virtual net mapping method based on load balancing.The present invention includes the following steps: the two-dimentional Load Balancing Model of step 1, settling time and resource；Step 2 establishes the mathematical programming model that two-dimentional load balancing degrees minimize；Step 3 is based on two-dimentional Load Balancing Model and mathematical programming model, maps virtual net.Two-dimentional Load Balancing Model includes node two dimension intensity of load wn, link two dimension intensity of load we, node two dimension load equilibrium and link two dimension load equilibrium；The mathematical programming model that two-dimentional load balancing degrees minimize includes minimizing objective function u.The present invention can in real time map virtual net, can be improved virtual net mapping success rate, and bottom physical network is made to be chronically at the state of load balancing.
Description
Technical field
The invention belongs to technical field of network virtualization, and in particular to a kind of virtual net mapping side based on load balancing
Method.The present invention relates to the buildings of the virtual net of the two dimensional attributes of time response and spatial character, the angle benefit based on load balancing
The method for carrying out virtual net mapping with the method for discrete weightings.
Background technique
Come in the past few decades, the development of internet be fastly it is unprecedented, due to supporting Distributed Application and heterogeneous network
Network technology, internet achieve huge success, become people's quick obtaining information and transmit the main channel of information.It is greatly
The fast development and extensive use that have stimulated network technology.However, since there are multiple suppliers in internet, using new network
Architecture is not only the change of individual host and routing, and needs the joint agreement between Internet Service Provider.But
It is that Internet provider considers that number one is difficult to reach an agreement on to new technique large scale deployment, therefore also just leads
Network has been caused to ossify problem.Network virtualization technology provides up-and-coming mode and solves the problems, such as that network ossifys.In network void
In quasi-ization, multiple service providers construct isomery virtual net in the bottom-layer network that one or more infrastructure suppliers lease
Network, and provide network service end to end, so as to reach can simultaneously individually carry out network technology innovation and service promotion.
With the rise of network virtualization, researcher proposes a series of virtual net developing algorithm, these algorithms are big
It is all based in the one-dimensional space in the same bottom-layer network while running multiple virtual net applications.There is no consider the time
The influence that factor maps virtual net, because virtual net request has runing time and duration, so as to cause bottom-layer network
Resource deteriorates with the progress of time.
The present invention proposes the side of the two-dimensional discrete weighting of time and resource load from the angle of Network Load Balance
Method.This method can be very good metric physical network resource status, and physical network resource is located always on the whole
In the state of load balancing, to improve network resource utilization.
Summary of the invention
The purpose of the present invention is in view of the deficiencies of the prior art, propose a kind of virtual net mapping side based on load balancing
Method.
It is as follows that this hair solves technical solution used by its technical problem:
Step 1, the two-dimentional Load Balancing Model of settling time and resource
Step 2 establishes the mathematical programming model that two-dimentional load balancing degrees minimize
Step 3 is based on two-dimentional Load Balancing Model and mathematical programming model, maps virtual net
The foundation of the two-dimentional Load Balancing Model of time described in step 1 and resource, specific as follows:
Two-dimentional Load Balancing Model includes node two dimension intensity of load wn, link two dimension intensity of load we, node two dimension load
Harmonious and link two dimension load equilibrium；
Node two dimension intensity of load wnCalculating it is as follows:
Formula (1)
Wherein, SnIt (t) is node load intensity,W (t) is the weight function of a monotone decreasing
And meetI is positive integer.
Link two dimension intensity of load weCalculating it is as follows:
Formula (2)
Wherein Se(t) t moment chain road intensity of load is indicated,W (t) is a monotone decreasing
Weight function and satisfactionI is positive integer.
Node two dimension load equilibrium NσCalculating it is as follows:
Formula (3)
Wherein,NσIndicate node two dimension load equilibrium on physical network, particularly respectively
The standard deviation of node two dimension intensity of load, NsIndicate physical node.
Link two dimension load equilibrium LσCalculating it is as follows:
Formula (4)
Wherein,LσIndicate physical network uplink two dimension load equilibrium, particularly respectively
The standard deviation of link two dimension intensity of load, EsIndicate physical link.
The mathematical programming model that two-dimentional load balancing degrees minimize described in step 2 includes minimizing objective function u:
Minu=α Nσ+βLσFormula (5)
Wherein, α, β are the parameter factors of 0-1；
Based on two-dimentional Load Balancing Model and mathematical programming model described in step 3, virtual net is mapped, specific mistake
Journey is as follows:
3-1. is arranged the dummy node of virtual request by the non-increasing of cpu resource, is successively mapped all dummy nodes,
Dummy node is mapped to the node two dimension intensity of load w for meeting resource constraintnOn the smallest physical node；
The resource constraint isThat is the cpu resource of dummy node request is surplus less than physical node
Remaining cpu resource.
3-2. is arranged the virtual link of virtual request by the non-increasing of bandwidth constraint, is successively reflected to all virtual links
It penetrates, the K shortest path that current virtual link meets bandwidth constraint and can map is found out from bottom-layer network, current virtual link
It is mapped to link two dimension intensity of load w in path seteOn the smallest path.
The K is positive integer, value range 3-7；
The bandwidth constraint isThat is it is remaining that the bandwidth resources of virtual link request are less than physical link
Bandwidth resources.
Mapping of the 3-3. based on step 3-1 and 3-2, the Mathematical Planning mould minimized according to the two-dimentional load balancing degrees of proposition
Type computational minimization objective function u.
Mapping of the 3-4. based on step 3-1 and 3-2, each dummy node are sequentially completed following process:
The unassigned physical node set M for meeting the constraint of current virtual node resource is found out, if one in set M
After a physical node replacement current virtual node mapped physical node, the value that can minimize objective function u is reduced, then is used
The physical node replaces current virtual node mapped physical node.
The present invention has the beneficial effect that:
The present invention can in real time map virtual net, can be improved virtual net mapping success rate, and make bottom object
Reason network is chronically at the state of load balancing.
Detailed description of the invention
Fig. 1 is virtual net map example schematic diagram of the present invention
Fig. 2 is discrete weightings right value function schematic diagram
Specific embodiment
The present invention will be further described below with reference to the drawings.
A kind of virtual net mapping method based on load balancing, includes the following steps:
Step 1, the two-dimentional Load Balancing Model of settling time and resource
Step 2 establishes the mathematical programming model that two-dimentional load balancing degrees minimize
Step 3 is based on two-dimentional Load Balancing Model and mathematical programming model, maps virtual net
The foundation of the two-dimentional Load Balancing Model of time described in step 1 and resource, specific as follows:
Two-dimentional Load Balancing Model includes node two dimension intensity of load wn, link two dimension intensity of load we, node two dimension load
Harmonious and link two dimension load equilibrium；
Node two dimension intensity of load wnCalculating it is as follows:
Formula (1)
Wherein, SnIt (t) is node load intensity,W (t) is the weight function of a monotone decreasing
And meetI is positive integer.
Link two dimension intensity of load weCalculating it is as follows:
Formula (2)
Wherein Se(t) t moment chain road intensity of load is indicated,W (t) is a monotone decreasing
Weight function and satisfactionI is positive integer.
Node two dimension load equilibrium NσCalculating it is as follows:
Formula (3)
Wherein,NσIndicate node two dimension load equilibrium on physical network, particularly
The standard deviation of each node two dimension intensity of load, NsIndicate physical node.
Link two dimension load equilibrium LσCalculating it is as follows:
Formula (4)
Wherein,LσIndicate physical network uplink two dimension load equilibrium, particularly respectively
The standard deviation of link two dimension intensity of load, EsIndicate physical link.
The mathematical programming model that two-dimentional load balancing degrees minimize described in step 2 includes minimizing objective function u:
Minu=α Nσ+βLσFormula (5)
Wherein, α, β are the parameter factors of 0-1；
Based on two-dimentional Load Balancing Model and mathematical programming model described in step 3, virtual net is mapped, specific mistake
Journey is as follows:
3-1. is arranged the dummy node of virtual request by the non-increasing of cpu resource, is successively mapped all dummy nodes,
Dummy node is mapped to the node two dimension intensity of load w for meeting resource constraintnOn the smallest physical node；
The resource constraint isThat is it is remaining that the cpu resource of dummy node request is less than physical node
Cpu resource.
3-2. is arranged the virtual link of virtual request by the non-increasing of bandwidth constraint, is successively reflected to all virtual links
It penetrates, the K shortest path that current virtual link meets bandwidth constraint and can map is found out from bottom-layer network, current virtual link
It is mapped to link two dimension intensity of load w in path seteOn the smallest path.
The K is positive integer, value range 3-7；
The bandwidth constraint isThat is it is remaining that the bandwidth resources of virtual link request are less than physical link
Bandwidth resources.
Mapping of the 3-3. based on step 3-1 and 3-2, the Mathematical Planning mould minimized according to the two-dimentional load balancing degrees of proposition
Type computational minimization objective function u.
Mapping of the 3-4. based on step 3-1 and 3-2, each dummy node are sequentially completed following process:
The unassigned physical node set M for meeting the constraint of current virtual node resource is found out, if one in set M
After a physical node replacement current virtual node mapped physical node, the value that can minimize objective function u is reduced, then is used
The physical node replaces current virtual node mapped physical node.
Embodiment:
As shown in Figure 1,1 interior joint mapping scheme of virtual network requests be { a → A, b → B }, link maps scheme be (a,
B) → (A, B) }, virtual net request 2 in mapping scheme be { c → D, d → C, f → F }, link maps scheme be (c, d) →
(D, B, C), (c, f) → (D, E, F) } by modification iteration map after mapping scheme can be changed to { c → D, d → A, f → C }, link
Mapping scheme is { (c, d) → (D, B, A), (c, f) → (D, E, C) }.
We cannot simply acquire the intensity of load phase Calais of timeslice each in a period of time the degree of load balancing degrees
Magnitude.Just start to be predistribution to virtual net resource due to us, if we need the resource pre-allocated to turn
When shifting, the cost of transfer closer from current time will be higher.As shown in Fig. 2, the method that can use discrete weightings, introduces one
The function w (t) of a monotone decreasing, and the function meetsAnd up and down using each timeslice beginning and ending time as integral
Limit will carry out the obtained value of resource load degree product of integral and the timeslice to weight function as timeslice intensity of load
Weight, and be based on
This is weighted summation, to obtain the two-dimentional intensity of load within the virtual net operation beginning and ending time.
Claims (2)
1. a kind of virtual net mapping method based on load balancing, it is characterised in that include the following steps:
Step 1, the two-dimentional Load Balancing Model of settling time and resource；
Step 2 establishes the mathematical programming model that two-dimentional load balancing degrees minimize；
Step 3 is based on two-dimentional Load Balancing Model and mathematical programming model, maps virtual net；
The foundation of the two-dimentional Load Balancing Model of time described in step 1 and resource, specific as follows:
Two-dimentional Load Balancing Model includes node two dimension intensity of load wn, link two dimension intensity of load we, node two dimension load balancing
Property and link two dimension load equilibrium；
Node two dimension intensity of load wnCalculating it is as follows:
Wherein, SnIt (t) is node load intensity,W (t) is the weight function and satisfaction of a monotone decreasingI is positive integer；
Link two dimension intensity of load weCalculating it is as follows:
Wherein Se(t) t moment chain road intensity of load is indicated,W (t) is the power of a monotone decreasing
Value function and satisfactionI is positive integer；
Node two dimension load equilibrium NσCalculating it is as follows:
Wherein,NσIndicate node two dimension load equilibrium on physical network, particularly each node
The standard deviation of two-dimentional intensity of load, NsIndicate the set of physical node；
Link two dimension load equilibrium LσCalculating it is as follows:
Wherein,LσIndicate physical network uplink two dimension load equilibrium, particularly each link
The standard deviation of two-dimentional intensity of load, EsIndicate the set of physical link；
The mathematical programming model that two-dimentional load balancing degrees minimize described in step 2 includes minimizing objective function u:
Min (u=α Nσ+βLσ) formula (5)
Wherein, α, β are the parameter factors of 0-1.
2. a kind of virtual net mapping method based on load balancing as described in claim 1, it is characterised in that described in step 3
Based on two-dimentional Load Balancing Model and mathematical programming model, virtual net is mapped, detailed process is as follows:
3-1. is arranged the dummy node of virtual request by the non-increasing of cpu resource, is successively mapped all dummy nodes, void
Quasi- node is mapped to the node two dimension intensity of load w for meeting resource constraintnOn the smallest physical node；
The resource constraint isThat is it is remaining that the cpu resource of dummy node request is less than physical node
Cpu resource；
3-2. is arranged the virtual link of virtual request by the non-increasing of bandwidth constraint, is successively mapped all virtual links,
The K shortest path that current virtual link meets bandwidth constraint and can map is found out from bottom-layer network, and current virtual link is reflected
It is mapped to link two dimension intensity of load w in path seteOn the smallest path；
The K is positive integer, value range 3-7；
The bandwidth constraint isThat is the bandwidth resources of virtual link request are less than the remaining band of physical link
Wide resource；
Mapping of the 3-3. based on step 3-1 and 3-2, the mathematical programming model meter minimized according to the two-dimentional load balancing degrees of proposition
It calculates and minimizes objective function u；
Mapping of the 3-4. based on step 3-1 and 3-2, each dummy node are sequentially completed following process:
The unassigned physical node set M for meeting the constraint of current virtual node resource is found out, if an object in set M
After managing node replacement current virtual node mapped physical node, the value that can minimize objective function u is reduced, then with the object
It manages node and replaces current virtual node mapped physical node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510604413.5A CN105187322B (en) | 2015-09-22 | 2015-09-22 | A kind of virtual net mapping method based on load balancing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510604413.5A CN105187322B (en) | 2015-09-22 | 2015-09-22 | A kind of virtual net mapping method based on load balancing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105187322A CN105187322A (en) | 2015-12-23 |
| CN105187322B true CN105187322B (en) | 2019-06-14 |
Family
ID=54909173
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510604413.5A Active CN105187322B (en) | 2015-09-22 | 2015-09-22 | A kind of virtual net mapping method based on load balancing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105187322B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106330557A (en) * | 2016-08-30 | 2017-01-11 | 杭州电子科技大学 | Multipath virtual network mapping method based on space-time association |
| CN106487707B (en) * | 2016-09-29 | 2019-06-18 | 北京邮电大学 | A kind of virtual fault tolerant network mapping method of power communication and device |
| CN108173777B (en) * | 2017-12-29 | 2020-08-04 | 杭州电子科技大学 | SDN-based virtual network mapping two-dimensional resource fragment measurement method |
| CN111824216B (en) * | 2020-06-19 | 2021-11-05 | 北京交通大学 | Train running scheme evaluation method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102223281A (en) * | 2011-06-24 | 2011-10-19 | 清华大学 | Method for remapping resource demand dynamic change based on virtual network |
| CN103457752A (en) * | 2012-05-30 | 2013-12-18 | 中国科学院声学研究所 | Virtual network mapping method |
| CN104104621A (en) * | 2013-04-07 | 2014-10-15 | 中国科学院声学研究所 | Dynamic adaptive adjustment method of virtual network resources based on nonlinear dimensionality reduction |
| EP2804343A1 (en) * | 2013-05-16 | 2014-11-19 | NTT DoCoMo, Inc. | Method for mapping a network topology request to a physical network, computer program product, mobile communication system, and network configuration platform |
-
2015
- 2015-09-22 CN CN201510604413.5A patent/CN105187322B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102223281A (en) * | 2011-06-24 | 2011-10-19 | 清华大学 | Method for remapping resource demand dynamic change based on virtual network |
| CN103457752A (en) * | 2012-05-30 | 2013-12-18 | 中国科学院声学研究所 | Virtual network mapping method |
| CN104104621A (en) * | 2013-04-07 | 2014-10-15 | 中国科学院声学研究所 | Dynamic adaptive adjustment method of virtual network resources based on nonlinear dimensionality reduction |
| EP2804343A1 (en) * | 2013-05-16 | 2014-11-19 | NTT DoCoMo, Inc. | Method for mapping a network topology request to a physical network, computer program product, mobile communication system, and network configuration platform |
Non-Patent Citations (3)
| Title |
|---|
| Virtual Network Mapping Algorithm for Large-scale Network Environment;ZHANG Shun-li.etc;《IEEE》;20120227;全文 |
| 基于时间因素的虚拟网映射策略研究;赵志洋;《中国优秀硕士论文电子期刊网》;20131215;全文 |
| 虚拟网络的映射技术研究;孙罡;《中国博士学位论文电子期刊网》;20130515;全文 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105187322A (en) | 2015-12-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112367109B (en) | Incentive method for digital twin-driven federal learning in air-ground network | |
| Yao et al. | A novel reinforcement learning algorithm for virtual network embedding | |
| US11018979B2 (en) | System and method for network slicing for service-oriented networks | |
| CN105187322B (en) | A kind of virtual net mapping method based on load balancing | |
| Zhang et al. | A novel edge computing architecture based on adaptive stratified sampling | |
| Zhang et al. | Model and algorithms for the planning of fog computing networks | |
| CN107566194A (en) | A kind of method for realizing the mapping of cross-domain virtual network network | |
| Lim et al. | UAV-assisted communication efficient federated learning in the era of the artificial intelligence of things | |
| CN103684850A (en) | Service neighborhood based Web Service quality prediction method | |
| Gong et al. | Deep learning-based computation offloading with energy and performance optimization | |
| Hosseini et al. | Resource allocation optimization in cloud computing using the whale optimization algorithm | |
| Ghafouri et al. | Mobile-kube: Mobility-aware and energy-efficient service orchestration on kubernetes edge servers | |
| Huang et al. | Distance-aware hierarchical federated learning in blockchain-enabled edge computing network | |
| Nandhakumar et al. | Edgeaisim: A toolkit for simulation and modelling of ai models in edge computing environments | |
| Wang et al. | Game-theory-based task offloading and resource scheduling in cloud-edge collaborative systems | |
| Yang et al. | Towards 6g joint haps-mec-cloud 3c resource allocation for delay-aware computation offloading | |
| CN109640261B (en) | Positioning algorithm of heterogeneous wireless sensor network based on support vector regression machine | |
| Fan et al. | DRL-D: revenue-aware online service function chain deployment via deep reinforcement learning | |
| CN108833162A (en) | A kind of network node sort method and mapping method of virtual network | |
| Ge et al. | Dynamic hierarchical caching resource allocation for 5G-ICN slice | |
| Arul et al. | Integration of IoT and edge cloud computing for smart microgrid energy management in VANET using machine learning | |
| Zhang et al. | Offloading demand prediction-driven latency-aware resource reservation in edge networks | |
| Wu et al. | PECCO: A profit and cost‐oriented computation offloading scheme in edge‐cloud environment with improved Moth‐flame optimization | |
| Pavan Kumar et al. | Qos aware service selection in iot using anp | |
| KR20220150126A (en) | Coded and Incentive-based Mechanism for Distributed Training of Machine Learning in IoT |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 6, 310027 building, horizon building, 181 Tianmu Road, Hangzhou, Zhejiang, Xihu District Applicant after: Handing you Yu Internet Limited by Share Ltd Address before: 6, 310027 building, horizon building, 181 Tianmu Road, Hangzhou, Zhejiang, Xihu District Applicant before: Hakim Information Technology Co., Ltd. |
|
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: Room 1101, South Building, Handing International Building, No. 5 Yongfuqiao Road, Xiacheng District, Hangzhou City, Zhejiang Province Applicant after: Handing you Yu Internet Limited by Share Ltd Address before: 310027 Tianji Building, 181 Tianmushan Road, Xihu District, Hangzhou City, Zhejiang Province, 6th Floor Applicant before: Handing you Yu Internet Limited by Share Ltd |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: Floor 17, building 6, jinjingwan business operation center, Pingtan Comprehensive Experimental Zone, Fuzhou City, Fujian Province 350400 Patentee after: Straits innovation Internet Co.,Ltd. Address before: Room 1101, South Building, Handing International Building, No. 5 Yongfuqiao Road, Xiacheng District, Hangzhou City, Zhejiang Province Patentee before: HAKIM UNIQUE INTERNET Co.,Ltd. |