CN106557611A - The Dynamic Load-balancing Algorithm research of distributed traffic network simulation platform and application - Google Patents
The Dynamic Load-balancing Algorithm research of distributed traffic network simulation platform and application Download PDFInfo
- Publication number
- CN106557611A CN106557611A CN201610890375.9A CN201610890375A CN106557611A CN 106557611 A CN106557611 A CN 106557611A CN 201610890375 A CN201610890375 A CN 201610890375A CN 106557611 A CN106557611 A CN 106557611A
- Authority
- CN
- China
- Prior art keywords
- road
- expense
- network
- migrated
- dynamic load
- 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
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/02—CAD in a network environment, e.g. collaborative CAD or distributed simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Traffic Control Systems (AREA)
Abstract
The Dynamic Load-balancing Algorithm research of distributed traffic network simulation platform and application.The present invention is proposed based on game theoretic Dynamic Load-balancing Algorithm, in a distributed manner based on traffic simulation platform, use for reference game theoretic thought, it is proposed that based on game theoretic Dynamic Load-balancing Algorithm, and by emulation experiment and theory analysis demonstrating the superiority of algorithm.Algorithm is defined to the cost function of road first, and proves that the function has Nash Equilibrium by theoretical derivation；Then, carrying out initial segmentation to map network by the way of iteration growth；Finally, road will calculate the expense of itself according to cost function, and according to the expense of itself come the purpose terminal for deciding whether to be migrated and migrated, the equilibrium until completing whole road network；And when there is unbalanced situation again in analogue system operation, equally road can also be migrated according to the expense of road, and reach equilibrium state again.
Description
Technical field
Based on the present invention is according to the distributed microcosmic traffic network simulation platform of independent development, devise on the platform
The dynamic load leveling model of mixed model, model are completed using the combination of CS and P2P, with CS between master control end and emulation terminal
Pattern exist, and exist with the pattern of P2P between emulation terminal and emulation terminal.Use for reference the think of of Nash Equilibrium in game theory
Think, regard single road as player, and provide the calculating function of road cost, allow which to be made whether what is migrated in alternating way
Select.Recursive bisection algorithm by experiment to Dynamic Load-balancing Algorithm proposed by the present invention with classics is compared.
Background technology
Traffic problems it is increasingly serious so that the experts and scholars inside and outside industry all traffic problems have been made deep analysis and
Research.But due to the factor that traffic problems are related in itself it is numerous, therefore, using real traffic environment come to traffic problems
Basic reason carries out studying has certain difficulty, it is necessary to realized observation by means of traffic simulating system to traffic behavior, divided
The work such as analysis and research, is emulated to traffic route network and operation vehicle by traffic simulating system, not only can be right
Various parameters and environmental effect etc. of transportation network operation are evaluated, and can for relevant departments assess existing road network or
Plan that new road provides effective foundation, be nowadays to analyze and assess urban road network planning and the maximally effective side of traffic administration
Formula, has the advantage that conventional model cannot be by comparison.
In traffic simulation field, with single unit vehicle as object of study, it is right more accurately can to realize for microscopic traffic simulation
The reproduction of behavior of transportation network condition of road surface, traffic participant etc. in reality, thus be considered as that current analysis solves many
One of most accurate and effective method of traffic problems.But, in the case of with single traffic entity as object of study, microcosmic is handed over
The amount of calculation of logical analogue system is very big.When road network scale is less, the computational efficiency of serial microscopic simulation system still receives scope
It is interior.When road network scale is larger, its simulation result is difficult to use in carries out traffic control, and emulation the spent time has exceeded is imitated
The physical time of true traffic system, although calculating speed can be improved in performance using high performance computer system, is reduced
Simulation time, but general high-performance computer system spends higher, it is relatively low for common traffic simulation cost performance, because
This, is not promoted and popularizes.Generally more often, mass transportation road can be realized using distributed traffic simulating system
The emulation of road network.How as far as possible between dynamic multiple equilibria emulation terminal artificial tasks simultaneously reduce the communication between emulation terminal
Expense is the key issue in distributed traffic simulating.
In distributed traffic network simulation platform, in order to reduce the simulation time of system, the simulation efficiency of system is improved,
It is requisite to introduce dynamic load leveling.
The content of the invention
The present invention proposes the road net model based on distributed traffic simulating and designs, according to the distributed microcosmic of independent development
Based on traffic network simulation platform, the dynamic load leveling model of mixed model is devised on the platform.The model is adopted
The combination of CS and P2P is present with the pattern of CS between master control end and emulation terminal completing, and between emulation terminal and emulation terminal
Exist with the pattern of P2P.Using such mixed model, that is, the emulation for enabling to terminal room reaches synchronization, it is also possible to reduce
Terminal and the communication overhead at master control end, reach the final equilibrium of system in the way of local equilibrium.
The present invention is proposed based on game theoretic Dynamic Load-balancing Algorithm, uses for reference the thought of Nash Equilibrium in game theory,
Regard single road as player, and provide the calculating function of road cost, allow which to be made whether the choosing for migrating in alternating way
Select.In many load balancing class algorithms before, road is typically passively selected.And set forth herein algorithm in, road
Road actively decides whether migration as player.
Recursive bisection algorithm by experiment to Dynamic Load-balancing Algorithm proposed by the present invention with classics is compared,
By the distributed microscopic traffic simulation platform of independent research, by two groups of experiments, change number of terminals from fixed vehicle number respectively
And fixed terminal number change vehicle number comes the simulation velocity to algorithm, simulation time variance, call duration time expense summation, communication
Four aspects of time variance are compared.Test result indicate that, set forth herein being based on game theoretic Dynamic Load-balancing Algorithm
Passing for classics is an advantage over all at simulation velocity, simulation time variance, call duration time expense summation, four aspects of call duration time variance
Return to point algorithm.
Description of the drawings
Fig. 1 is road net model composition figure；
Fig. 2 is load-balancing algorithm comprising modules figure；
Fig. 3 is traffic network simulation road network exploded view；
Fig. 4 is system CS framework and P2P Organization Charts；
Fig. 5 is the dynamic load leveling illustraton of model of mixed model.
Specific embodiment
Below in conjunction with the accompanying drawings technical scheme is elaborated.
In microcosmic traffic network simulation, generally transportation network flow is emulated in units of vehicle individual.It is simultaneously right
The motion of vehicle and their interaction are described.Wherein road network module is mainly responsible for the structure of road network, using road network
Topological structure showing the structure of road network in real world, the main positional information for including road of structure of road network, track
Situation, the distribution of signal lighties and surrounding other information etc., as shown in Figure 1.
The optimal situation of dynamic load leveling, is can to reach in system under the premise of terminal emulation time identical, eventually
The traffic between end is minimum.Common dynamic load balancing includes three modules：Information collection module, decision-making module and task are moved
Shifting formwork block, as shown in Figure 2.
In order to be able to reach bigger simulation scale, while simplified system is designed, the distributed traffic network for adopting herein is imitated
True emulation of the system to transportation network is decomposed on artificial tasks and road network so that the difference in traffic network simulation is appointed
Business is completed by different terminals.Such as Simulation Control is completed by Simulation Control end, simulation calculation is completed, emulated by simulation calculation end
As a result show and completed by display end, each terminal is separate, while interacting.Decomposition on road network is referred to entirely to be needed
Transportation network to be emulated resolves into several road subnets and is emulated respectively, as shown in Figure 3.Traffic in subnet point
Do not completed by different emulation terminals, wherein needing the traffic for being emulated mainly to include：The state of vehicle updates, traffic
Control setting, collection of traffic statistics information of facility (such as traffic lights) etc..Meanwhile, as sub-net boundary vehicle between subnet
Turnover and produce the traffic.
The dynamic load leveling model of the mixed model for adopting herein is as shown in Figure 4.After Task-decomposing, emulation is eventually
End only needs to the emulation for being responsible for vehicle, shows come the result to emulating without expending resource.Can effectively improve
The simulation efficiency of system.System comes the start-stop of control system load balancing, parameter setting etc. using CS frameworks, using P2P frameworks
The information transmission mutual to realize terminal room, the load balancing for after provide valid data.Master control end is according to from emulation data
Whether the data inquired about in storehouse are at equilibrium to differentiate emulation terminal, when master control end judges that emulation terminal is not at balanced shape
When state, dynamic equalization order is sent to emulation terminal.Now, the emulation terminal for receiving dynamic equalization order will be next
After step emulation is completed, stop emulation, and the data etc. shared according to neighbor node to complete dynamic load leveling.When in system
Emulation terminal complete again equilibrium after, system start continue emulation, until complete master control end requirement simulation time.Specifically
As shown in Figure 5.
Claims (3)
1. propose based on game theoretic Dynamic Load-balancing Algorithm, it is characterised in that：
The cost function of road is defined, and proves that the function has Nash Equilibrium by theoretical derivation；
Initial segmentation is carried out to map network by the way of iteration growth；
Road will calculate the expense of itself according to cost function, and according to the expense of itself deciding whether to be migrated and be moved
The purpose terminal of shifting, the equilibrium until completing whole road network；
And when there is unbalanced situation again in analogue system operation, equally can also come to road according to the expense of road
Migrated, and reached equilibrium state again.
2. as claimed in claim 1, algorithm is defined to the cost function of road first, and proves the letter by theoretical derivation
There is Nash Equilibrium in number；Then, carrying out initial segmentation to map network by the way of iteration growth；Finally, road is by root
The expense of itself is calculated according to cost function, and according to the expense of itself come the purpose end for deciding whether to be migrated and migrated
End, the equilibrium until completing whole road network；And when there is unbalanced situation again in analogue system operation, equally can also
Road is migrated according to the expense of road, and reach equilibrium state again.
3. as claimed in claim 1, on the basis of the distributed microcosmic traffic network simulation platform of independent research, devise
The dynamic load leveling model of mixed model.The model is completed using the combination of CS and P2P, between master control end and emulation terminal with
The pattern of CS is present, and exists with the pattern of P2P between emulation terminal and emulation terminal.While in order to improve the imitative of analogue system
Road in road network is reduced to the point of Weight, and crossing is reduced to the side of Weight by true velocity, reduction data acquisition cost.
Road network is emulated using the model, emulation road net model can be simplified while the verity of emulation is not lost, improve imitative
The simulation velocity and simulation efficiency of true system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610890375.9A CN106557611A (en) | 2016-10-12 | 2016-10-12 | The Dynamic Load-balancing Algorithm research of distributed traffic network simulation platform and application |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610890375.9A CN106557611A (en) | 2016-10-12 | 2016-10-12 | The Dynamic Load-balancing Algorithm research of distributed traffic network simulation platform and application |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106557611A true CN106557611A (en) | 2017-04-05 |
Family
ID=58418434
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610890375.9A Pending CN106557611A (en) | 2016-10-12 | 2016-10-12 | The Dynamic Load-balancing Algorithm research of distributed traffic network simulation platform and application |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106557611A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107103169A (en) * | 2017-06-26 | 2017-08-29 | 上海交通大学 | It is a kind of to be used to meet the transportation network equilibrium calculation method that trip continuation of the journey is required |
| CN111177892A (en) * | 2019-12-11 | 2020-05-19 | 中电普信(北京)科技发展有限公司 | Distributed simulation system |
| CN112134807A (en) * | 2020-08-26 | 2020-12-25 | 天津城建大学 | SDN load balancing method, system, equipment and storage medium based on Nash bargaining |
| CN112149287A (en) * | 2020-09-08 | 2020-12-29 | 安徽中科龙安科技股份有限公司 | Traffic simulation road network graphical segmentation method and system oriented to load balancing |
| CN117435351A (en) * | 2023-12-20 | 2024-01-23 | 深圳市城市交通规划设计研究中心股份有限公司 | Load balancing method for road simulation distributed computation |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140278331A1 (en) * | 2013-08-08 | 2014-09-18 | Iteris, Inc. | Pavement condition analysis from modeling impact of traffic characteristics, weather data and road conditions on segments of a transportation network infrastructure |
| CN104575021A (en) * | 2014-12-17 | 2015-04-29 | 浙江工业大学 | Distributed model predictive control method for urban road network system based on neighborhood optimization |
| CN105119837A (en) * | 2015-05-12 | 2015-12-02 | 电子科技大学 | Distributed traffic simulation load balancing algorithm based on simulation terminal configuration optimization |
-
2016
- 2016-10-12 CN CN201610890375.9A patent/CN106557611A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140278331A1 (en) * | 2013-08-08 | 2014-09-18 | Iteris, Inc. | Pavement condition analysis from modeling impact of traffic characteristics, weather data and road conditions on segments of a transportation network infrastructure |
| CN104575021A (en) * | 2014-12-17 | 2015-04-29 | 浙江工业大学 | Distributed model predictive control method for urban road network system based on neighborhood optimization |
| CN105119837A (en) * | 2015-05-12 | 2015-12-02 | 电子科技大学 | Distributed traffic simulation load balancing algorithm based on simulation terminal configuration optimization |
Non-Patent Citations (1)
| Title |
|---|
| 张萌洁: "分布式交通网络仿真平台的动态负载均衡算法研究及应用", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107103169A (en) * | 2017-06-26 | 2017-08-29 | 上海交通大学 | It is a kind of to be used to meet the transportation network equilibrium calculation method that trip continuation of the journey is required |
| CN107103169B (en) * | 2017-06-26 | 2020-03-24 | 上海交通大学 | Traffic network balance calculation method for meeting travel continuation requirements |
| CN111177892A (en) * | 2019-12-11 | 2020-05-19 | 中电普信(北京)科技发展有限公司 | Distributed simulation system |
| CN111177892B (en) * | 2019-12-11 | 2023-05-02 | 中电普信(北京)科技发展有限公司 | Distributed simulation system |
| CN112134807A (en) * | 2020-08-26 | 2020-12-25 | 天津城建大学 | SDN load balancing method, system, equipment and storage medium based on Nash bargaining |
| CN112149287A (en) * | 2020-09-08 | 2020-12-29 | 安徽中科龙安科技股份有限公司 | Traffic simulation road network graphical segmentation method and system oriented to load balancing |
| CN117435351A (en) * | 2023-12-20 | 2024-01-23 | 深圳市城市交通规划设计研究中心股份有限公司 | Load balancing method for road simulation distributed computation |
| CN117435351B (en) * | 2023-12-20 | 2024-04-30 | 深圳市城市交通规划设计研究中心股份有限公司 | Load balancing method for road simulation distributed computation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106557611A (en) | The Dynamic Load-balancing Algorithm research of distributed traffic network simulation platform and application | |
| Han et al. | A dynamic resource allocation framework for synchronizing metaverse with iot service and data | |
| CN103994768B (en) | Method and system for seeking for overall situation time optimal path under dynamic time varying environment | |
| CN109141446A (en) | For obtaining the method, apparatus, equipment and computer readable storage medium of map | |
| CN102426424B (en) | Visual emulation method of near-space flying vehicle based on distributed framework | |
| Turek | Erlang-based desynchronized urban traffic simulation for high-performance computing systems | |
| WO2013167085A2 (en) | Traffic network division method based on grid mapping generation and configuration server | |
| US8612197B1 (en) | Large scale simulation architecture for distributed networking waveforms | |
| CN104880961A (en) | Real-time simulation experiment system of multi-unmanned plane distributed synergetic hardware in loop | |
| CN110415521A (en) | Prediction technique, device and the computer readable storage medium of traffic data | |
| TWI512679B (en) | Method of computing a collision-free velocity for an agent in a crowd simulation environment | |
| CN106844980A (en) | A kind of scheduling system and its dispatching method for supporting systemic-function dynamic reorganization | |
| CN107172650A (en) | A kind of emulation mode and system of large-scale complex wireless communication system | |
| Simonov et al. | Multi-agent crowd simulation on large areas with utility-based behavior models: Sochi Olympic Park Station use case | |
| Fu et al. | Demonstrating geosparksim: A scalable microscopic road network traffic simulator based on apache spark | |
| CN111773717A (en) | Object control method and apparatus, storage medium, and electronic apparatus | |
| Ibn-Khedher et al. | Next-generation edge computing assisted autonomous driving based artificial intelligence algorithms | |
| CN110012021A (en) | A kind of adaptive polo placement moving method under mobile edge calculations | |
| CN103955801A (en) | Electric power system distributed parallel computing management method based on time-space dimension | |
| CN106209483A (en) | Satellite communication emulation mode and device | |
| Lozano et al. | A new system architecture for crowd simulation | |
| US10484249B1 (en) | Dynamic distribution of simulation load | |
| CN114676541B (en) | Traffic simulation method, related device, equipment and storage medium | |
| Wang et al. | Spatial conflict resolution in a multi-agent process by the use of a snake model | |
| US20220391558A1 (en) | Multi-agent simulation system and method |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170405 |