CN104539744B - A kind of the media edge cloud dispatching method and device of two benches cooperation - Google Patents
A kind of the media edge cloud dispatching method and device of two benches cooperation Download PDFInfo
- Publication number
- CN104539744B CN104539744B CN201510039231.8A CN201510039231A CN104539744B CN 104539744 B CN104539744 B CN 104539744B CN 201510039231 A CN201510039231 A CN 201510039231A CN 104539744 B CN104539744 B CN 104539744B
- Authority
- CN
- China
- Prior art keywords
- service
- migration
- deployment
- film
- 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.)
- Active
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
- H04L67/14—Session management
- H04L67/148—Migration or transfer of sessions
-
- 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
- H04L67/04—Protocols specially adapted for terminals or networks with limited capabilities; specially adapted for terminal portability
-
- 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
- H04L67/1004—Server selection for load balancing
- H04L67/101—Server selection for load balancing based on network conditions
Abstract
The invention discloses the media edge cloud dispatching methods and device of a kind of cooperation of two benches, belong to multimedia network communication field.This method and device neatly carry out scheduling of resource according to the different amplitude of variation of user access pattern using online service moving method or dynamic content dispositions method；Meanwhile during dynamic content is disposed, also the service flow impacted because of film deployment adjustment is handled using online service moving method.Therefore, this method and device are cooperated by online service moving method and dynamic content dispositions method, first, can be during user access pattern changes over time, realize scheduling process seamlessly transitting from migration phase to deployment phase, and ensure the high request receptance of media edge cloud with rational cost, second is that fully transparent property of the scheduling process to user service can be realized, third, can be effectively prevented from user access pattern occurs influence of the shake to request receptance.
Description
Technical field
The present invention relates to the media edge cloud dispatching parties that multimedia network communication field more particularly to a kind of two benches cooperate
Method and device.
Background technology
In recent years, along with the universal of the increase of number of network users and mobile terminal, large-scale concurrent user asks
The access way of access and multiple terminals is sought, the new challenge that stream media service system faces just is being become.However, traditional based on CDN
The stream media service system of (content distributing network) technology, P2P (peer-to-peer network) technologies or Clustering, autgmentability, can
By property, service encapsulation ability etc., there is certain limitations.
The rise and development of cloud computing promote streaming media service to cloud Morphological Transitions.Streaming Media cloud is Streaming Media and cloud meter
The product that calculation technology is combined can well solve traditional stream media service system problem encountered.Based on Streaming Media cloud
Service system (Streaming Media cloud service system) be a wide area system, by placing media edge respectively in different zones
Cloud shifts the requested streaming medium content of user onto network edge, to reduce user response delay and mitigate backbone network flow
Load.As the Streaming Media cloud of wide area, media edge cloud be also the resources such as huge storage and calculating are virtualized and
It is managed collectively, therefore the node for being responsible for providing a user streaming media service is Virtual Service node, and in traditional Streaming Media
In service system, the node for being responsible for providing streaming media service is then physical server node.On the other hand, what is be managed collectively is each
The strength of reliability of class physical resource is not of uniform size, and longtime running will necessarily increase media edge cloud and the general of node failure occurs
Rate；In order to improve service reliability, each Virtual Service node in media edge cloud is organized according to distributed mode, with
Avoid influence of the node failure problem to user service.
Although being responsible for providing a user streaming media service by Virtual Service node, with traditional stream media service system one
The problem of sample, media edge cloud is dispatched there is also system resource, the i.e. resources such as Optimum utilization storage, bandwidth, to remain
Height request receptance.In Streaming Media cloud service system, media edge cloud storage space, the limitation of service capacity and user visit
Ask that the inclination of pattern frequently can lead to following problem.1) film disposes unreasonable problem.It is visited in dynamic streaming media service
It asks in environment, since the popularity of film is continually changing, thus when user access pattern has greatly changed, originally
The media edge cloud that service performance is stablized, can be because current film deployment be not suitable for active user's access module, and generates and ask
The phenomenon that seeking receptance degradation.2) service flow irrational distribution problem.Although user access pattern does not generate larger
Variation, but due to the not perfect or user of request scheduling technology interrupts service flow etc., the matchmaker of script load distribution equilibrium
Body edge cloud, the phenomenon that just will produce service flow irrational distribution, and then influence the request receptance of whole system.
In order to solve the problems, such as because film dispose or service flow distribution it is unreasonable due to caused by request receptance decline, have
Many professionals are directed to traditional stream media service system, it has been proposed that corresponding dispatching method, including online services migrating
Method (being directed to service flow irrational distribution problem) and dynamic content dispositions method (disposing unreasonable problem for film).
Online service moving method is migrated service flow to low negative from high load node using online service migrating technology
Node is carried, to readjust distribution of the service flow on each Virtual Service node；Although online service moving method has adjustment
The small advantage of cost, but it does not tackle the problem at its root, i.e., the deployment of current film may be not suitable for having sent out
The user access pattern for changing；Therefore, online service moving method is only applicable to the smaller situation of user access pattern variation.
In addition, traditional online service migrating technology is application layer migrating technology, need by RTSP agent nodes as medium；In stream matchmaker
The quantity of body service access it is larger or migration service flow it is more when, agent node often become stream media service system property
It can bottleneck；And application layer migrating technology does not account for the problem of network layer state synchronized, thus it is only applicable to the cluster of local
Formula stream media service system.
Dynamic content dispositions method, quantity according to the variation tendency of user access pattern to all kinds of film copies and puts
It sets into Mobile state and adjusts, can dramatically improve the request receptance of whole system；But it does not account for service flow
The problem of migration, frequently can lead to some service flows using dynamic content dispositions method merely and be produced because of the deletion of film copy
It is raw to interrupt, and then influence QoS of customer.
Invention content
The object of the present invention is to provide the media edge cloud dispatching methods and device of a kind of cooperation of two benches, can be with smaller
Cost avoids the request receptance of media edge cloud from declining, and it is smooth from migration phase to deployment phase both to have realized scheduling process
Transition avoids excessively high scheduling cost further through the flexible selection of scheduling strategy；Meanwhile disposing process executing dynamic content
In, by the cooperation of online service moving method, realize fully transparent property of the film deployment adjustment to user service.
The purpose of the present invention is what is be achieved through the following technical solutions：
A kind of media edge cloud dispatching method of two benches cooperation, this method include：
It asks to access come real-time judge according to whether media edge cloud can access the current all users concurrently reached simultaneously
Whether state is normal；If the current all users request concurrently reached can not be accessed simultaneously, judges access state exception, execute
First online service moving method of migration phase realizes service stream migration and service flow distribution adjustment, so that media edge cloud
The current all users request concurrently reached can be accessed simultaneously；
Again periodically statistically in a period all kinds of films user's request data, using the method for prediction, under acquisition
Prediction user request in one period, and as the user access pattern predicted, and then judge the variation of user access pattern
Amplitude；If amplitude of variation range more than expected, by the dynamic content dispositions method of deployment phase, exist to adjust all kinds of films
Deployment on each Virtual Service node, and using the second online service moving method of migration phase, to handle because film is disposed
It adjusts and needs the service flow migrated.
Further, the first online service moving method of the migration phase is that the online service based on OpenFlow is moved
Shifting method comprising：Forecasting type moving method and non-prediction type moving method；
Wherein, forecasting type moving method will be about cyclic forecast information the opening as service stream migration for servicing flow distribution
Hairdo condition；Specifically：Current film deployment, current service flow distribution and maximum service capacity on each Virtual Service node
Under constraints, to minimize the migration cost as optimization aim, and according to about forecasting type optimal service flow distribution
Migration principle, carry out the calculating of migration strategy；
Non-prediction type moving method is a kind of migration optimization method based on linear programming, as an optimization by service flow distribution
Object；Specifically：Current film deployment, current service flow distribution and maximum service capacity-constrained on each Virtual Service node
Under the conditions of, to receive all users currently concurrently reached to ask and minimize the migration cost as optimization aim, use
The method of linear programming establishes Optimized model, and carries out the calculating of migration strategy.
Further, described to judge that the amplitude of variation of user access pattern includes：
By the prediction user request and current service stream merger of all kinds of films, and the synthetic user for seeing all kinds of films as is asked
It asks；
Current film deployment, maximum memory space and maximum service capacity constraints on each Virtual Service node
Under, it is established using the method for linear programming using distribution of the synthetic user request on each Virtual Service node as decision variable
The mathematical model of one media edge cloud resource scheduling problem；
According to the mathematical model established, the synthetic user request distribution that searching can meet all constraint equations is feasible
Solution；
If any one feasible solution can not be found, the amplitude of variation of user access pattern range more than expected is judged；
Otherwise, then judge the amplitude of variation of user access pattern without departing from expected range.
Further, the dynamic content dispositions method by deployment phase, to adjust all kinds of films in each virtual clothes
The deployment being engaged on node, and using the second online service moving method of migration phase, needed due to film is disposed and adjusted to handle
The service flow to be migrated includes：
Current film deployment, current service flow distribution, maximum memory space on each Virtual Service node and maximum clothes
It is engaged under capacity constraints, to receive all prediction users to ask and minimize deployment total cost as optimization aim, into Mobile state
The calculating of content disposition strategy, for determining the new deployment of the one kind of all kinds of films on each Virtual Service node, in order to complete shadow
Piece deployment adjustment and need the Virtual Service node of the film copy and place deleted, need increased film copy and right
The destination virtual service node answered；
Wherein, the deployment total cost includes：Cost, deletion and the increase film for copying film and bandwidth resources being occupied
Copy and to the cost of I/O resource consumptions, service stream migration caused by Virtual Service node between completion status synchronizing information
Cost and OpenFlow switch processes migration service streams cost；
The dynamic content deployment strategy includes online services migrating strategy, by the second online service migration side
Method is cooperated and is generated, and needs the service flow migrated and the target of migration service stream because film copy is deleted for determining
Virtual Service node.
Further, the second online service moving method of the migration phase is that the online service based on OpenFlow is moved
Shifting method comprising：
Execute dynamic content deployment during, when because film copy is deleted cause corresponding to service flow need to migrate
When, the current film on each Virtual Service node is disposed, under current service flow distribution and maximum service capacity constraints, with
It is optimization aim to minimize migration cost, is each migration from remaining Virtual Service node containing deleted film copy
Service flow selection target Virtual Service node.
A kind of media edge cloud dispatching device of two benches cooperation, the device include：
Real-time scheduler module, for the current all users concurrently reached whether can be accessed simultaneously according to media edge cloud
Whether request carrys out real-time judge access state normal；If the current all users request concurrently reached can not be accessed simultaneously, sentence
Determine access state exception, execute the first online service moving method of migration phase, realizes service stream migration and service flow distribution
Adjustment, so that media edge cloud can access the current all users request concurrently reached simultaneously；
Periodic scheduling module is adopted for user's request data of all kinds of films in a periodic statistically period again
With the method for prediction, the prediction user request in next period is obtained, and as the user access pattern predicted, and then judge
The amplitude of variation of user access pattern；If amplitude of variation range more than expected passes through the dynamic content deployment side of deployment phase
Method to adjust deployment of all kinds of films on each Virtual Service node, and utilizes the second online service migration side of migration phase
Method needs the service flow migrated to handle due to film is disposed and adjusted.
Further, the real-time scheduler module includes：First online service transferring module, for executing migration phase
The first online service moving method；This method is the online service moving method based on OpenFlow comprising：Forecasting type moves
Shifting method and non-prediction type moving method；
Wherein, forecasting type moving method will be about cyclic forecast information the opening as service stream migration for servicing flow distribution
Hairdo condition；Specifically：Current film deployment, current service flow distribution and maximum service capacity on each Virtual Service node
Under constraints, to minimize the migration cost as optimization aim, and according to about forecasting type optimal service flow distribution
Migration principle, carry out the calculating of migration strategy；
Non-prediction type moving method is a kind of migration optimization method based on linear programming, as an optimization by service flow distribution
Object；Specifically：Current film deployment, current service flow distribution and maximum service capacity-constrained on each Virtual Service node
Under the conditions of, to receive all users currently concurrently reached to ask and minimize the migration cost as optimization aim, use
The method of linear programming establishes Optimized model, and carries out the calculating of migration strategy.
Further, the periodic scheduling module includes：User access pattern amplitude of variation judgment module, for judging
The amplitude of variation of user access pattern comprising：
By the prediction user request and current service stream merger of all kinds of films, and the synthetic user for seeing all kinds of films as is asked
It asks；
Current film deployment, maximum memory space and maximum service capacity constraints on each Virtual Service node
Under, it is established using the method for linear programming using distribution of the synthetic user request on each Virtual Service node as decision variable
The mathematical model of one media edge cloud resource scheduling problem；
According to the mathematical model established, the synthetic user request distribution that searching can meet all constraint equations is feasible
Solution；
If any one feasible solution can not be found, the amplitude of variation of user access pattern range more than expected is judged；
Otherwise, then judge the amplitude of variation of user access pattern without departing from expected range.
Further, the periodic scheduling module includes：Dynamic content is disposed and services migrating module, for passing through portion
The dynamic content dispositions method in administration's stage to adjust deployment of all kinds of films on each Virtual Service node, and executes migration rank
Second online service moving method of section needs the service flow migrated to handle due to film is disposed and adjusted comprising：
Current film deployment, current service flow distribution, maximum memory space on each Virtual Service node and maximum clothes
It is engaged under capacity constraints, to receive all prediction users to ask and minimize deployment total cost as optimization aim, into Mobile state
The calculating of content disposition strategy, for determining the new deployment of the one kind of all kinds of films on each Virtual Service node, in order to complete shadow
Piece deployment adjustment and need the Virtual Service node of the film copy and place deleted, need increased film copy and right
The destination virtual service node answered；
Wherein, the deployment total cost includes：Cost, deletion and the increase film for copying film and bandwidth resources being occupied
Copy and to the cost of I/O resource consumptions, service stream migration caused by Virtual Service node between completion status synchronizing information
Cost and OpenFlow switch processes migration service streams cost；
The dynamic content deployment strategy includes online services migrating strategy, by the second online service migration side
Method is cooperated and is generated, and needs the service flow migrated and the target of migration service stream because film copy is deleted for determining
Virtual Service node.
Further, the dynamic content, which is disposed with services migrating module, includes：Second online service transferring module, is used for
The second online service moving method of migration phase is executed, this method is the online service moving method based on OpenFlow,
Including：
Execute dynamic content deployment during, when because film copy is deleted cause corresponding to service flow need to migrate
When, the current film on each Virtual Service node is disposed, under current service flow distribution and maximum service capacity constraints, with
It is optimization aim to minimize migration cost, is each migration from remaining Virtual Service node containing deleted film copy
Service flow selection target Virtual Service node.
As seen from the above technical solution provided by the invention, it is disposed by online service moving method and dynamic content
Method cooperates, first, scheduling process can be realized from migration rank during user access pattern changes over time
Section seamlessly transitting to deployment phase, and ensure the high request receptance of media edge cloud with rational cost, second is that can be real
Existing scheduling process occurs to shake to request to the fully transparent property of user service third, user access pattern can be effectively prevented from
The influence of receptance.
Description of the drawings
In order to illustrate the technical solution of the embodiments of the present invention more clearly, required use in being described below to embodiment
Attached drawing be briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for this
For the those of ordinary skill in field, without creative efforts, other are can also be obtained according to these attached drawings
Attached drawing.
Fig. 1 is a kind of flow chart of the media edge cloud dispatching method for two benches cooperation that the embodiment of the present invention one provides；
Fig. 2 is the flow chart for the execution real-time scheduling that the embodiment of the present invention one provides；
Fig. 3 is the flow chart for the execution periodic scheduling that the embodiment of the present invention one provides；
Fig. 4 is the flow chart for the amplitude of variation for judging user access pattern that the embodiment of the present invention one provides；
Fig. 5 is the Streaming Media cloud service system structural schematic diagram of a media edge cloud provided by Embodiment 2 of the present invention；
Fig. 6 is a kind of schematic diagram of the media edge cloud dispatching device of two benches cooperation provided by Embodiment 2 of the present invention.
Specific implementation mode
With reference to the attached drawing in the embodiment of the present invention, technical solution in the embodiment of the present invention carries out clear, complete
Ground describes, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Based on this
The embodiment of invention, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, belongs to protection scope of the present invention.
Access, the new challenges such as access way of multiple terminals, streaming media service are asked in order to cope with large-scale concurrent user
To cloud Morphological Transitions.Streaming Media cloud service system is asked user by being distributed in the media edge clouds of different zones
Content shift network edge onto；In media edge cloud, it is responsible for providing a user streaming media service by Virtual Service node.Dynamic
In the streaming media service access environment of state, when user access pattern changes, in order to avoid film deployment or service flow
The unreasonable influence to asking receptance of distribution, media edge cloud needs to utilize suitable scheduling strategy, according to active user
Access module dynamically adjusts the film deployment on each Virtual Service node or service flow distribution.
Existing dispatching method includes the online service moving method (being directed to service flow irrational distribution problem) of migration phase
With the dynamic content dispositions method of deployment phase (disposing unreasonable problem for film), but these methods are all difficult in scheduling generation
Accomplish reasonably to weigh between valence and dispatching effect.
The present invention considers the advantages of online service moving method and dynamic content dispositions method and deficiency, devises one kind
Migration phase and the co-operating media edge cloud scheduling scheme of deployment phase；The program can be according to user access pattern difference
Amplitude of variation, smoothly the smaller preferable dynamic content dispositions method of online service moving method and effect of cost is selected
Carry out scheduling of resource；It, can also be by online service moving method, to handle meanwhile during executing dynamic content deployment
The service flow migrated is needed because film copy is deleted.Therefore, the present invention passes through online service moving method and dynamic content portion
Arranging method cooperates, first, scheduling process can be realized from migration during user access pattern changes over time
Stage seamlessly transits to deployment phase, and ensures the high request receptance of media edge cloud with rational cost, second is that can
Fully transparent property of the scheduling process to user service is realized, third, can be effectively prevented from user access pattern occurs shake to asking
Ask the influence of receptance；It is described in detail below by embodiment.
Embodiment one
Fig. 1 is a kind of flow chart of the media edge cloud dispatching method for two benches cooperation that the embodiment of the present invention one provides.
As shown in Figure 1, it mainly includes the following steps：
Step 11 executes real-time scheduling.
The step specifically includes：The current all users request concurrently reached whether can be accessed simultaneously according to media edge cloud
Whether normal carry out real-time judge access state；If the current all users request concurrently reached can not be accessed simultaneously, judgement connects
Enter abnormal state, execute the first online service moving method of migration phase, realizes service stream migration and service flow distribution adjustment,
So that media edge cloud can access the current all users request concurrently reached simultaneously.
Step 12 executes periodic scheduling.
The step specifically includes：User's request data of all kinds of films in a periodic statistically period, using prediction
Method, obtain the prediction user request in next period, and as the user access pattern predicted, then judge user's visit
Ask the amplitude of variation of pattern；If amplitude of variation range more than expected, by the dynamic content dispositions method of deployment phase, to adjust
Deployment of the whole all kinds of films on each Virtual Service node, and using the second online service moving method of migration phase, to locate
Reason needs the service flow migrated due to film is disposed and adjusted.
In order to make it easy to understand, 2-3 does detailed introduction to above-mentioned two step below in conjunction with the accompanying drawings.
As shown in Fig. 2, to execute the flow chart of real-time scheduling, mainly include the following steps：
Step 21 judges whether the access state of user's request is normal in real time；Specifically, media edge is monitored in real time
The access situation that cloud asks concurrent user, and it is useful whether to access the institute currently concurrently reached simultaneously according to media edge cloud
Whether family request carrys out real-time judge access state normal；If normal, it is transferred to step 23；Otherwise, it is transferred to step 22.
If step 22 can not access the current all users request concurrently reached simultaneously, access state exception is judged,
Execute the first online service moving method of migration phase.
In the embodiment of the present invention, the first online service moving method of the migration phase is for realizing service stream migration
It is adjusted with service flow distribution, so that media edge cloud can access the current all users request concurrently reached simultaneously.
Illustratively, can be a kind of lazy formula trigger mechanism in terms of trigger mechanism；Specifically：Lazy formula triggering
Mechanism asks the state of access media edge cloud to be perceived in real time user, only when the user concurrently reached asks nothing
Method simultaneously by media edge cloud access when, then the first online service moving method can just be triggered.In technology and strategy
Aspect can be a kind of online service moving method based on OpenFlow, i.e., by OpenFlow technologies, realize that service flow is moved
It moves and service flow distribution adjusts；Specifically：Migration strategy based on OpenFlow, the migration cost of variable as an optimization,
The cost being different from traditional online service moving method includes not only completion status synchronizing information between Virtual Service node
Cost further includes the cost of OpenFlow switch processes migration service streams.
In the embodiment of the present invention, the first online service moving method may include：Forecasting type moving method and non-prediction type
One kind in moving method.Specifically：
1) forecasting type moving method using about service flow distribution cyclic forecast information as service stream migration inspiration
Formula condition；Specifically：Current film deployment, current service flow distribution and maximum service capacity on each Virtual Service node are about
Under the conditions of beam, to minimize the migration cost as optimization aim, and according to about forecasting type optimal service flow distribution
Principle is migrated, the calculating of migration strategy is carried out；On the one hand for user's request selecting to media edge cloud can not be linked into
On the other hand suitable access node is used to determine that user to ask access and needs the service flow migrated, migration clothes in order to complete
The destination virtual service node of business stream.
Wherein, the forecasting type optimal service flow distribution has recorded the optimal of all kinds of films on each Virtual Service node
Fluxion is serviced, is after periodically predicting user access pattern, in current service flow distribution, current film deployment or warp
It crosses under the new film deployment conditions of adjustment, to receive all prediction user requests, target calculates as an optimization, is used for conduct
The heuristic evidence of online services migrating in next period.The migration principle about forecasting type optimal service flow distribution,
It specifically refers to, if in the access node being fully loaded with, the service fluxion of the corresponding film of user's request is less than optimal service stream
Number, then illustrate that remaining film occupies its bandwidth resources, needs a service stream migration of remaining film in access node extremely
Other Virtual Service node, so that enough bandwidth resources are vacateed in the access asked for user.
2) non-prediction type moving method is a kind of migration optimization method based on linear programming, using service flow distribution as excellent
Change object；Specifically：Current film deployment, current service flow distribution and maximum service capacity on each Virtual Service node are about
Under the conditions of beam, to receive all users currently concurrently reached to ask and minimize the migration cost as optimization aim, adopt
Optimized model is established with the method for linear programming, and carries out the calculating of migration strategy.It is used to determine current service stream in each void
A kind of new distribution on quasi- service node needs the service flow migrated and migration clothes to complete the adjustment of service flow distribution
The destination virtual service node of business stream.
Step 23, judgement access state are normal.
At this point it is possible to judge whether timing cycle arrives；If timing cycle arrives, periodic scheduling step is executed；
Otherwise, return to step 21 continue to execute real-time scheduling.
As shown in figure 3, to execute the flow chart of periodic scheduling, mainly include the following steps：
Step 31, prediction user access pattern.
In the embodiment of the present invention, historical requests statistical number was recorded in user's request data of all kinds of films in a upper period
According to inner, then according to the historical requests statistical data recorded, using the method for prediction, it is calculated in time in next period
It predicts user's request, that is, predicts new user access pattern；In order to ensure that user asks the accuracy and promptness of prediction, Ke Yixuan
A smaller time window is selected as statistics and predetermined period；In view of statistics and the smaller characteristic of predetermined period, twice phase
User access pattern between the adjacent period is there are a degree of continuity and proximity, therefore a kind of simple but very practical meter
Calculation method is directly to be asked using the user that is actually reached in a upper period, as the prediction user request in next period.
Step 32, the amplitude of variation for judging user access pattern；If amplitude of variation is more than expected range, it is transferred to step
33。
Judgment step is as shown in figure 4, include mainly：
Obtain the current film deployment in media edge cloud on each Virtual Service node, current service flow distribution, maximum are deposited
Store up space and maximum service capacity information；
By the prediction user request and current service stream merger of all kinds of films, and the synthetic user for seeing all kinds of films as is asked
It asks；
Current film deployment, maximum memory space and maximum service capacity constraints on each Virtual Service node
Under, it is established using the method for linear programming using distribution of the synthetic user request on each Virtual Service node as decision variable
The mathematical model of one media edge cloud resource scheduling problem；
According to the mathematical model established, the synthetic user request distribution that searching can meet all constraint equations is feasible
Solution；
If any one feasible solution can not be found, the amplitude of variation of user access pattern range more than expected is judged；
Otherwise, then judge the amplitude of variation of user access pattern without departing from expected range.
Illustratively, with reference to an example, to judging that the amplitude of variation method of user access pattern carries out specifically
It is bright.Assuming that each physical quantity of media edge cloud includes, and films types set V, Virtual Service node set N, a specific film
Type i ∈ V, a specific Virtual Service node j ∈ N, current film deployment matrix D (element dij=1/0 indicates Virtual Service section
In the presence/absence of the copy of film i on point j), current service flow distribution matrix S (element sijIndicate shadow on Virtual Service node j
The service fluxion of piece i), the prediction user of all kinds of films asks matrix P (element piIndicate the prediction user number of request of film i),
The synthetic user of all kinds of films asks matrix PS (elementsIndicate the synthetic user number of request of film i), respectively
Memory capacity Matrix C (the element c of Virtual Service nodejIndicate that Virtual Service node j can most open ended film copies
Number), service capacity matrix L (the element l of each Virtual Service nodejIndicate the at most open ended service flows of Virtual Service node j
Number), decision variable is that synthetic user asks distribution matrix X (element xijIndicate the synthetic user of film i on Virtual Service node j
Number of request).So, the mathematical model of the media edge cloud resource scheduling problem based on linear programming is as follows：
If there is the feasible solution of a synthetic user request matrix X, it disclosure satisfy that above-mentioned all constraint equations, then sentence
The amplitude of variation for determining user access pattern is smaller (without departing from expected range)；Conversely, then judging the variation width of user access pattern
It spends larger (range more than expected).
Step 33, immediately trigger deployment phase dynamic content dispositions method, by with the second online service moving method
Cooperation, generate and execute dynamic content deployment strategy, pellucidly complete on each Virtual Service node film deployment adjustment, and
The request receptance of media edge cloud is fundamentally avoided to decline.
Specifically：Current film deployment on each Virtual Service node, current service flow distribution, maximum memory space
Under maximum service capacity constraints, to receive all prediction users requests and minimum deployment total cost as optimization aim,
The calculating for carrying out dynamic content deployment strategy, for determining that the one kind of all kinds of films on each Virtual Service node newly disposes, is
It completes film deployment adjustment and needs the Virtual Service node of the film copy and place deleted, needs increased film pair
Sheet and corresponding destination virtual service node；
Wherein, the deployment total cost includes：Cost, deletion and the increase film for copying film and bandwidth resources being occupied
Copy and to the cost of I/O resource consumptions, service stream migration caused by Virtual Service node between completion status synchronizing information
Cost and OpenFlow switch processes migration service streams cost；
The dynamic content deployment strategy includes online services migrating strategy, by the second online service migration side
Method is cooperated and is generated, and needs the service flow migrated and the target of migration service stream because film copy is deleted for determining
Virtual Service node.Second online service moving method of above-mentioned migration phase is also based on the online service migration of OpenFlow
Method, but it is different from the first online service moving method in terms of algorithm policy；Its detailed process is as follows：Executing dynamic content
During deployment, when because film copy is deleted cause corresponding to service flow need migration when, on each Virtual Service node
Current film deployment, be optimization to minimize migration cost under current service flow distribution and maximum service capacity constraints
Target virtually takes from remaining Virtual Service node containing deleted film copy for each migration service stream selection target
Business node.
When amplitude of variation is less than expected range, then real-time scheduling shown in Fig. 2 is directly returned.That is, according to user
The access state of request, triggers the first online service moving method of migration phase in real time, and avoids matchmaker with smaller cost
The request receptance of body edge cloud declines.
In the embodiment of the present invention, by cooperating for online service moving method and dynamic content dispositions method, when
It can realize that scheduling process is smoothed from migration phase to deployment phase during user access pattern changes over time
It crosses, and media edge cloud is made to remain higher request receptance；It is imitated with scheduling second is that both efficiently solving scheduling cost
Contradiction between fruit, and realize fully transparent property of the scheduling process to user service；Third, user can also be effectively prevented from
Influence of the shake to request receptance occurs for access module.
Through the above description of the embodiments, those skilled in the art can be understood that above-described embodiment can
By software realization, the mode of necessary general hardware platform can also be added to realize by software.Based on this understanding,
The technical solution of above-described embodiment can be expressed in the form of software products, the software product can be stored in one it is non-easily
In the property lost storage medium (can be CD-ROM, USB flash disk, mobile hard disk etc.), including some instructions are with so that a computer is set
Standby (can be personal computer, server or the network equipment etc.) executes the method described in each embodiment of the present invention.
Embodiment two
In order to solve the new challenge of streaming media service appearance and the existing issue of traditional stream media service system, Streaming Media
Service is to cloud Morphological Transitions.Streaming Media cloud service system by placing media edge cloud respectively in different zones, by user
Requested content shifts network edge onto, i.e. user only needs the media edge cloud by accessing this area, it will be able to obtain institute
The streaming media service needed.Media edge cloud towards be the user that various terminals access is used in this area, be responsible for straight
It connects and provides a user reliable streaming media service.Therefore, good scheduling of resource will effectively improve the service of media edge cloud
Performance, the request receptance for increasing media edge cloud and the streaming media service quality for improving this area user.
Fig. 5 is containing there are one the Streaming Media cloud service system structural schematic diagrams of media edge cloud.Whole system is mainly by matchmaker
Body edge cloud, user terminal, OpenFlow networks (including service managing server and OpenFlow interchangers) and media edge cloud
Dispatching device forms.Specifically, media edge cloud is organized in the way of cloud computing by a large amount of Virtual Service nodes,
It is responsible for directly providing streaming media service to this area user；User in user terminal can access media edge by various terminals
Cloud；By the auxiliary of OpenFlow networks, media edge cloud can provide transparent service for the user in user terminal；Wherein,
On the one hand service managing server has the function of accessing user's request, using the access way tried one's best, be responsible for use
The request at family end is assigned to suitable Virtual Service node, on the other hand, as OpenFlow controllers, is responsible for OpenFlow
Interchanger carries out flow table configuration.OpenFlow interchangers correctly forward matching service flow according to flow table rule, and
Complete the network layer state synchronized of migration service stream；Media edge cloud dispatching device, storage, band for optimizing media edge cloud
The utilization of the resources such as width generates and executes the strategy of film deployment adjustment or service flow distribution adjustment, to avoid asking in time
Ask the decline of receptance.
For the storage resource and bandwidth resources in preferably Optimum utilization media edge cloud, an embodiment of the present invention provides
A kind of media edge cloud dispatching device of two benches cooperation, as shown in fig. 6, the device mainly includes：
Real-time scheduler module 61, it is useful for the current institute concurrently reached whether can be accessed simultaneously according to media edge cloud
Whether family request carrys out real-time judge access state normal；If the current all users request concurrently reached can not be accessed simultaneously,
Judge that access state is abnormal, execute the first online service moving method of migration phase, realizes service stream migration and service flow point
Cloth adjusts, so that media edge cloud can access the current all users request concurrently reached simultaneously；
Periodic scheduling module 62 is used for again user's request data of all kinds of films in a periodic statistically period,
Using the method for prediction, the prediction user request in next period is obtained, and as the user access pattern predicted, and then sentence
The amplitude of variation of disconnected user access pattern；If amplitude of variation range more than expected, disposed by the dynamic content of deployment phase
Method to adjust deployment of all kinds of films on each Virtual Service node, and is migrated using the second online service of migration phase
Method needs the service flow migrated to handle due to film is disposed and adjusted.
Further, the real-time scheduler module 61 includes：First online service transferring module 611 is moved for executing
The first online service moving method in shifting stage；This method is the online service moving method based on OpenFlow comprising：In advance
Survey type moving method and non-prediction type moving method；
Wherein, forecasting type moving method will be about cyclic forecast information the opening as service stream migration for servicing flow distribution
Hairdo condition；Specifically：Current film deployment, current service flow distribution and maximum service capacity on each Virtual Service node
Under constraints, to minimize the migration cost as optimization aim, and according to about forecasting type optimal service flow distribution
Migration principle, carry out the calculating of migration strategy；
Non-prediction type moving method is a kind of migration optimization method based on linear programming, as an optimization by service flow distribution
Object；Specifically：Current film deployment, current service flow distribution and maximum service capacity-constrained on each Virtual Service node
Under the conditions of, to receive all users currently concurrently reached to ask and minimize the migration cost as optimization aim, use
The method of linear programming establishes Optimized model, and carries out the calculating of migration strategy.
Further, the periodic scheduling module 62 includes：User access pattern amplitude of variation judgment module 621 is used
In the amplitude of variation for judging user access pattern comprising：
By the prediction user request and current service stream merger of all kinds of films, and the synthetic user for seeing all kinds of films as is asked
It asks；
Current film deployment, maximum memory space and maximum service capacity constraints on each Virtual Service node
Under, it is established using the method for linear programming using distribution of the synthetic user request on each Virtual Service node as decision variable
The mathematical model of one media edge cloud resource scheduling problem；
According to the mathematical model established, the synthetic user request distribution that searching can meet all constraint equations is feasible
Solution；
If any one feasible solution can not be found, the amplitude of variation of user access pattern range more than expected is judged；
Otherwise, then judge the amplitude of variation of user access pattern without departing from expected range.
Further, the periodic scheduling module 62 includes：Dynamic content is disposed and services migrating module 622, is used for
By the dynamic content dispositions method of deployment phase, to adjust deployment of all kinds of films on each Virtual Service node, and execute
Second online service moving method of migration phase needs the service flow migrated to handle due to film is disposed and adjusted comprising：
Current film deployment, current service flow distribution, maximum memory space on each Virtual Service node and maximum clothes
It is engaged under capacity constraints, to receive all prediction users to ask and minimize deployment total cost as optimization aim, into Mobile state
The calculating of content disposition strategy, for determining the new deployment of the one kind of all kinds of films on each Virtual Service node, in order to complete shadow
Piece deployment adjustment and need the Virtual Service node of the film copy and place deleted, need increased film copy and right
The destination virtual service node answered；
Wherein, the deployment total cost includes：Cost, deletion and the increase film for copying film and bandwidth resources being occupied
Copy and to the cost of I/O resource consumptions, service stream migration caused by Virtual Service node between completion status synchronizing information
Cost and OpenFlow switch processes migration service streams cost；
The dynamic content deployment strategy includes online services migrating strategy, by the second online service migration side
Method is cooperated and is generated, and needs the service flow migrated and the target of migration service stream because film copy is deleted for determining
Virtual Service node.
Further, the dynamic content, which is disposed with services migrating module 622, includes：Second online service transferring module
6221, the second online service moving method for executing migration phase, this method is that the online service based on OpenFlow is moved
Shifting method comprising：
During executing dynamic content deployment, corresponding service flow is caused to need to migrate when film copy is deleted
When, the current film on each Virtual Service node is disposed, under current service flow distribution and maximum service capacity constraints, with
It is optimization aim to minimize migration cost, is each migration from remaining Virtual Service node containing deleted film copy
Service flow selection target Virtual Service node.
It should be noted that the specific implementation of the functions implemented by the various functional modules for including in above-mentioned apparatus exists
It has had a detailed description in each embodiment of front, therefore has repeated no more herein.
It is apparent to those skilled in the art that for convenience and simplicity of description, only with above-mentioned each function
The division progress of module, can be as needed and by above-mentioned function distribution by different function moulds for example, in practical application
Block is completed, i.e., the internal structure of device is divided into different function modules, to complete all or part of work(described above
Energy.
The foregoing is only a preferred embodiment of the present invention, but scope of protection of the present invention is not limited thereto,
Any one skilled in the art is in the technical scope of present disclosure, the change or replacement that can be readily occurred in,
It should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with the protection model of claims
Subject to enclosing.
Claims (10)
1. a kind of media edge cloud dispatching method of two benches cooperation, which is characterized in that this method includes：
It is asked come real-time judge access state according to whether media edge cloud can access the current all users concurrently reached simultaneously
It is whether normal；If the current all users request concurrently reached can not be accessed simultaneously, judges access state exception, execute migration
The first online service moving method in stage realizes service stream migration and service flow distribution adjustment, so that media edge cloud can be same
When access currently concurrently reach all users request；
User's request data of all kinds of films obtains next week using the method for prediction in a periodic statistically period again
Prediction user request in phase, and as the user access pattern predicted, and then judge the amplitude of variation of user access pattern；
If amplitude of variation range more than expected, by the dynamic content dispositions method of deployment phase, to adjust all kinds of films in each void
Deployment on quasi- service node, and using the second online service moving method of migration phase, adjusted to handle to dispose because of film
And need the service flow migrated.
2. according to the method described in claim 1, it is characterized in that, the first online service moving method of the migration phase is
Online service moving method based on OpenFlow comprising：Forecasting type moving method and non-prediction type moving method；
Wherein, forecasting type moving method using about service flow distribution cyclic forecast information as service stream migration it is heuristic
Condition；Specifically：Current film deployment, current service flow distribution and maximum service capacity-constrained on each Virtual Service node
Under the conditions of, to minimize migration cost as optimization aim, and according to the migration principle about forecasting type optimal service flow distribution,
Carry out the calculating of migration strategy；Wherein, migration cost includes：The cost of completion status synchronizing information between Virtual Service node,
And the cost of OpenFlow switch processes migration service streams；
Non-prediction type moving method is a kind of migration optimization method based on linear programming, and service flow distribution is right as an optimization
As；Specifically：Current film deployment, current service flow distribution and maximum service capacity-constrained item on each Virtual Service node
Under part, to receive all users currently concurrently reached to ask and minimize the migration cost as optimization aim, using line
Property planning method establish Optimized model, and carry out the calculating of migration strategy.
3. according to the method described in claim 1, it is characterized in that, described judge that the amplitude of variation of user access pattern includes：
By the prediction user request and current service stream merger of all kinds of films, and see the synthetic user request of all kinds of films as；
Under current film deployment, maximum memory space and maximum service capacity constraints on each Virtual Service node, with
Distribution of the synthetic user request on each Virtual Service node establishes one as decision variable using the method for linear programming
The mathematical model of media edge cloud resource scheduling problem；
According to the mathematical model established, the synthetic user request distribution feasible solution that can meet all constraint equations is found；
If any one feasible solution can not be found, the amplitude of variation of user access pattern range more than expected is judged；Otherwise,
Then judge the amplitude of variation of user access pattern without departing from expected range.
4. according to the method described in claim 1, it is characterized in that, the dynamic content dispositions method by deployment phase,
Adjust deployment of all kinds of films on each Virtual Service node, and using the second online service moving method of migration phase,
The service flow migrated is needed to include due to film is disposed and adjusted to handle：
Current film deployment, current service flow distribution, maximum memory space and maximum service on each Virtual Service node are held
It measures under constraints, to receive all prediction users to ask and minimize deployment total cost as optimization aim, carries out dynamic content
The calculating of deployment strategy, for determining the new deployment of the one kind of all kinds of films on each Virtual Service node, in order to complete film portion
Administration's adjustment and need the Virtual Service node of the film copy and place deleted, need increased film copy and corresponding
Destination virtual service node；
Wherein, the deployment total cost includes：Cost, deletion and the increase film copy for copying film and bandwidth resources being occupied
And to the cost of I/O resource consumptions, service stream migration caused by Virtual Service node between completion status synchronizing information generation
The cost of valence and OpenFlow switch processes migration service streams；
The dynamic content deployment strategy includes online services migrating strategy, by the second online service moving method into
Row cooperation and generates, need the destination virtual of the service flow migrated and migration service stream because film copy is deleted for determination
Service node.
5. according to the method described in claim 4, it is characterized in that, the second online service moving method of the migration phase is
Online service moving method based on OpenFlow comprising：
Execute dynamic content deployment during, when because film copy is deleted cause corresponding to service flow need migrate when,
Under current film deployment, current service flow distribution and maximum service capacity constraints on each Virtual Service node, with most
It is optimization aim that smallization, which migrates cost, from remaining Virtual Service node containing deleted film copy, is taken for each migration
Business stream selection target Virtual Service node；Wherein, migration cost includes：Completion status synchronizing information between Virtual Service node
The cost of cost and OpenFlow switch processes migration service streams.
6. a kind of media edge cloud dispatching device of two benches cooperation, which is characterized in that the device includes：
Real-time scheduler module, for being asked according to whether media edge cloud can access the current all users concurrently reached simultaneously
Whether normal carry out real-time judge access state；If the current all users request concurrently reached can not be accessed simultaneously, judgement connects
Enter abnormal state, execute the first online service moving method of migration phase, realizes service stream migration and service flow distribution adjustment,
So that media edge cloud can access the current all users request concurrently reached simultaneously；
Periodic scheduling module, for user's request data of all kinds of films in a periodic statistically period again, using pre-
The method of survey obtains the prediction user request in next period, and as the user access pattern predicted, and then judges user
The amplitude of variation of access module；If amplitude of variation range more than expected, by the dynamic content dispositions method of deployment phase, come
Deployment of all kinds of films on each Virtual Service node is adjusted, and using the second online service moving method of migration phase, is come
Processing needs the service flow migrated due to film is disposed and adjusted.
7. device according to claim 6, which is characterized in that the real-time scheduler module includes：First online service
Transferring module, the first online service moving method for executing migration phase；This method is the online clothes based on OpenFlow
Business moving method comprising：Forecasting type moving method and non-prediction type moving method；
Wherein, forecasting type moving method using about service flow distribution cyclic forecast information as service stream migration it is heuristic
Condition；Specifically：Current film deployment, current service flow distribution and maximum service capacity-constrained on each Virtual Service node
Under the conditions of, to minimize migration cost as optimization aim, and according to the migration principle about forecasting type optimal service flow distribution,
Carry out the calculating of migration strategy；Wherein, migration cost includes：The cost of completion status synchronizing information between Virtual Service node,
And the cost of OpenFlow switch processes migration service streams；
Non-prediction type moving method is a kind of migration optimization method based on linear programming, and service flow distribution is right as an optimization
As；Specifically：Current film deployment, current service flow distribution and maximum service capacity-constrained item on each Virtual Service node
Under part, to receive all users currently concurrently reached to ask and minimize the migration cost as optimization aim, using line
Property planning method establish Optimized model, and carry out the calculating of migration strategy.
8. device according to claim 6, which is characterized in that the periodic scheduling module includes：User access pattern
Amplitude of variation judgment module, the amplitude of variation for judging user access pattern comprising：
By the prediction user request and current service stream merger of all kinds of films, and see the synthetic user request of all kinds of films as；
Under current film deployment, maximum memory space and maximum service capacity constraints on each Virtual Service node, with
Distribution of the synthetic user request on each Virtual Service node establishes one as decision variable using the method for linear programming
The mathematical model of media edge cloud resource scheduling problem；
According to the mathematical model established, the synthetic user request distribution feasible solution that can meet all constraint equations is found；
If any one feasible solution can not be found, the amplitude of variation of user access pattern range more than expected is judged；Otherwise,
Then judge the amplitude of variation of user access pattern without departing from expected range.
9. device according to claim 6, which is characterized in that the periodic scheduling module includes：Dynamic content is disposed
With services migrating module, the dynamic content dispositions method for passing through deployment phase, to adjust all kinds of films in each Virtual Service
Deployment on node, and the second online service moving method of migration phase is executed, it is needed due to film is disposed and adjusted to handle
The service flow of migration comprising：
Current film deployment, current service flow distribution, maximum memory space and maximum service on each Virtual Service node are held
It measures under constraints, to receive all prediction users to ask and minimize deployment total cost as optimization aim, carries out dynamic content
The calculating of deployment strategy, for determining the new deployment of the one kind of all kinds of films on each Virtual Service node, in order to complete film portion
Administration's adjustment and need the Virtual Service node of the film copy and place deleted, need increased film copy and corresponding
Destination virtual service node；
Wherein, the deployment total cost includes：Cost, deletion and the increase film copy for copying film and bandwidth resources being occupied
And to the cost of I/O resource consumptions, service stream migration caused by Virtual Service node between completion status synchronizing information generation
The cost of valence and OpenFlow switch processes migration service streams；
The dynamic content deployment strategy includes online services migrating strategy, by the second online service moving method into
Row cooperation and generates, need the destination virtual of the service flow migrated and migration service stream because film copy is deleted for determination
Service node.
10. device according to claim 9, which is characterized in that the dynamic content, which is disposed with services migrating module, includes：
Second online service transferring module, the second online service moving method for executing migration phase, this method be based on
The online service moving method of OpenFlow comprising：
Execute dynamic content deployment during, when because film copy is deleted cause corresponding to service flow need migrate when,
Under current film deployment, current service flow distribution and maximum service capacity constraints on each Virtual Service node, with most
It is optimization aim that smallization, which migrates cost, from remaining Virtual Service node containing deleted film copy, is taken for each migration
Business stream selection target Virtual Service node；Wherein, migration cost includes：Completion status synchronizing information between Virtual Service node
The cost of cost and OpenFlow switch processes migration service streams.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510039231.8A CN104539744B (en) | 2015-01-26 | 2015-01-26 | A kind of the media edge cloud dispatching method and device of two benches cooperation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510039231.8A CN104539744B (en) | 2015-01-26 | 2015-01-26 | A kind of the media edge cloud dispatching method and device of two benches cooperation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104539744A CN104539744A (en) | 2015-04-22 |
| CN104539744B true CN104539744B (en) | 2018-08-24 |
Family
ID=52855211
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510039231.8A Active CN104539744B (en) | 2015-01-26 | 2015-01-26 | A kind of the media edge cloud dispatching method and device of two benches cooperation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104539744B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106911511B (en) * | 2017-03-10 | 2019-09-13 | 网宿科技股份有限公司 | A kind of means of defence and system of CDN client source station |
| CN107396204B (en) * | 2017-06-12 | 2019-08-27 | 江苏大学 | A kind of P2P video on demand node selecting method based on linear programming and intensified learning |
| CN108509276B (en) * | 2018-03-30 | 2021-11-30 | 南京工业大学 | Video task dynamic migration method in edge computing environment |
| CN111447266B (en) * | 2020-03-24 | 2022-11-11 | 中国人民解放军国防科技大学 | Mobile edge calculation model based on chain and service request and scheduling method thereof |
| CN111464611B (en) * | 2020-03-30 | 2022-07-12 | 中科边缘智慧信息科技（苏州）有限公司 | Method for efficiently accessing service between fixed cloud and edge node in dynamic complex scene |
| CN112506644B (en) * | 2020-11-02 | 2022-09-30 | 合肥工业大学 | Task scheduling method and system based on cloud edge-side hybrid computing mode system |
| CN117014313B (en) * | 2023-09-26 | 2023-12-19 | 工业云制造(四川)创新中心有限公司 | Method and system for analyzing equipment data of edge cloud platform in real time |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101252546A (en) * | 2008-04-15 | 2008-08-27 | 中国科学技术大学 | Method and apparatus for migrating medium stream online service |
| EP2211508A1 (en) * | 2009-01-22 | 2010-07-28 | IBBT vzw | Method and device for characterising a data flow in a network for transferring media data |
| CN102739798A (en) * | 2012-07-05 | 2012-10-17 | 成都国腾实业集团有限公司 | Cloud platform resource scheduling method with network sensing function |
| CN103188813A (en) * | 2013-02-25 | 2013-07-03 | 南京邮电大学 | Distributed resource scheduling method combined with dynamic management of control information |
| CN104065663A (en) * | 2014-07-01 | 2014-09-24 | 复旦大学 | Auto-expanding/shrinking cost-optimized content distribution service method based on hybrid cloud scheduling model |
-
2015
- 2015-01-26 CN CN201510039231.8A patent/CN104539744B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101252546A (en) * | 2008-04-15 | 2008-08-27 | 中国科学技术大学 | Method and apparatus for migrating medium stream online service |
| EP2211508A1 (en) * | 2009-01-22 | 2010-07-28 | IBBT vzw | Method and device for characterising a data flow in a network for transferring media data |
| CN102739798A (en) * | 2012-07-05 | 2012-10-17 | 成都国腾实业集团有限公司 | Cloud platform resource scheduling method with network sensing function |
| CN103188813A (en) * | 2013-02-25 | 2013-07-03 | 南京邮电大学 | Distributed resource scheduling method combined with dynamic management of control information |
| CN104065663A (en) * | 2014-07-01 | 2014-09-24 | 复旦大学 | Auto-expanding/shrinking cost-optimized content distribution service method based on hybrid cloud scheduling model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104539744A (en) | 2015-04-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104539744B (en) | A kind of the media edge cloud dispatching method and device of two benches cooperation | |
| Silva Filho et al. | Approaches for optimizing virtual machine placement and migration in cloud environments: A survey | |
| Aujla et al. | MEnSuS: An efficient scheme for energy management with sustainability of cloud data centers in edge–cloud environment | |
| WO2022021176A1 (en) | Cloud-edge collaborative network resource smooth migration and restructuring method and system | |
| Boru et al. | Energy-efficient data replication in cloud computing datacenters | |
| Dong et al. | Energy-performance tradeoffs in IaaS cloud with virtual machine scheduling | |
| Li et al. | CaaS: Caching as a service for 5G networks | |
| Talebian et al. | Optimizing virtual machine placement in IaaS data centers: taxonomy, review and open issues | |
| CN103747274B (en) | A kind of video data center setting up cache cluster and cache resources dispatching method thereof | |
| CN105610715B (en) | A kind of cloud data center multi-dummy machine migration scheduling method of planning based on SDN | |
| Dong et al. | Virtual machine placement for improving energy efficiency and network performance in iaas cloud | |
| CN104092756A (en) | Cloud storage system resource dynamic allocation method based on DHT mechanism | |
| Wang et al. | Bandwidth guaranteed virtual network function placement and scaling in datacenter networks | |
| CN108196935A (en) | A kind of energy saving moving method of virtual machine towards cloud computing | |
| Silvestre et al. | Aren: a popularity aware replication scheme for cloud storage | |
| CN106681815A (en) | Concurrent migration method of virtual machines | |
| Ma et al. | Dynamic task scheduling in cloud-assisted mobile edge computing | |
| Yapicioglu et al. | A traffic-aware virtual machine placement method for cloud data centers | |
| CN107197039B (en) | A kind of PAAS platform service packet distribution method and system based on CDN | |
| CN109039933A (en) | A kind of cluster network optimization method, device, equipment and medium | |
| Liu et al. | Deadline guaranteed service for multi-tenant cloud storage | |
| Mahmoudi et al. | SDN-DVFS: an enhanced QoS-aware load-balancing method in software defined networks | |
| Wang et al. | Proactive caching in mobile networks with delay guarantees | |
| Mashaly et al. | Load balancing in cloud-based content delivery networks using adaptive server activation/deactivation | |
| Rawas et al. | LECC: Location, energy, carbon and cost-aware VM placement model in geo-distributed DCs |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |