CN107948004A - A kind of video CD N transfers optimization method and device - Google Patents
A kind of video CD N transfers optimization method and device Download PDFInfo
- Publication number
- CN107948004A CN107948004A CN201711486733.0A CN201711486733A CN107948004A CN 107948004 A CN107948004 A CN 107948004A CN 201711486733 A CN201711486733 A CN 201711486733A CN 107948004 A CN107948004 A CN 107948004A
- Authority
- CN
- China
- Prior art keywords
- target
- video server
- priority
- client
- dispatch
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 87
- 238000012546 transfer Methods 0.000 title claims abstract description 22
- 238000005457 optimization Methods 0.000 title claims abstract description 13
- 239000000126 substance Substances 0.000 claims description 12
- 239000000284 extract Substances 0.000 claims description 7
- 230000005540 biological transmission Effects 0.000 claims description 6
- 241000208340 Araliaceae Species 0.000 claims description 3
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims description 3
- 235000003140 Panax quinquefolius Nutrition 0.000 claims description 3
- 235000008434 ginseng Nutrition 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 17
- 230000001154 acute effect Effects 0.000 description 10
- 238000012545 processing Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 241001269238 Data Species 0.000 description 2
- 238000012512 characterization method Methods 0.000 description 2
- 230000007812 deficiency Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000011084 recovery Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- HUTDUHSNJYTCAR-UHFFFAOYSA-N ancymidol Chemical compound C1=CC(OC)=CC=C1C(O)(C=1C=NC=NC=1)C1CC1 HUTDUHSNJYTCAR-UHFFFAOYSA-N 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 238000005498 polishing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
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
- H04L67/01—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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/142—Network analysis or design using statistical or mathematical methods
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Algebra (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Mathematical Physics (AREA)
- Probability & Statistics with Applications (AREA)
- Pure & Applied Mathematics (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The present invention provides a kind of video CD N and transfers optimization method and device, this method is applied to dispatch server, utilize the target priority and destination probability of the priority identification model identification client pre-established, and then to determine the target video server cluster of client, and therefrom video server is distributed for client.Based on method disclosed by the invention, the rank of user can be identified, and provide targetedly dispatch service to the user according to recognition result.
Description
Technical field
The present invention relates to mobile communication technology field, and optimization method and dress are transferred more specifically to a kind of video CD N
Put.
Background technology
Video CD N (Content Delivery Network, content distributing network) system is one and is capable of providing video
Flow the system of service.In video CDN system, the effect of dispatch server is exactly to specify downloading video files burst for user
Video server.
At present, the common scheduling strategy of dispatch server has two kinds, and one kind is flow control policy, it is another then for based on
The scheduling strategy of user's speed of download.For second of scheduling strategy, dispatch server is by under the request scheduling of user to user
Carry the highest video server of speed.
And since the highest video server of speed of download is mainly voluntarily tested the speed by client definite, i.e., client is true
Fix and carry fastest video server information and feed back to dispatch server.Therefore, dispatch server also just can not be according to
User class provides specific aim dispatch service, for example, the member user for ordinary user and than ordinary user's priority higher,
The dispatch service that dispatch server is provided is consistent.
The content of the invention
In view of this, the present invention provides a kind of video CD N and transfers optimization method and device, to solve existing dispatch server
The problem of specific aim dispatch service can not be provided according to user class.Technical solution is as follows：
A kind of video CD N transfers optimization method, applied to dispatch server, including：
When receiving the dispatch request of client transmission, the target that the client is obtained according to the dispatch request is closed
Join daily record data；
According to the target association daily record data and the priority identification model that pre-establishes, the mesh of the client is determined
Destination probability corresponding to mark priority and the target priority；
According to the target priority and the destination probability, the target video server cluster of the client is determined；
It is that the client distributes video server, and the video that will be distributed from the target video server cluster
The label of server is sent to the client.
Preferably, priority identification model is pre-established, including：
Transfer the first daily record data of the dispatch server and the second daily record data of default video server；
The scheduling data of history access client are chosen from first daily record data, and from second daily record data
The middle downloading data for choosing the history access client；
The historical context daily record data of data and the downloading data as the client is dispatched using described, and according to pre-
Setting priority mark rule adds priority tag for the historical context daily record data, and the priority tag includes priority
And the probability corresponding to the priority；
The data characteristics of the historical context daily record data and the priority aspects of the priority tag are extracted respectively；
Based on the data characteristics, the priority aspects and default supervised learning algorithm, priority identification mould is established
Type.
Preferably, it is described according to the target priority and the destination probability, determine the target video of the client
Server cluster, including：
Judge whether the destination probability is less than probability threshold value；
It it is the target association day according to pre-set priority mark rule if the destination probability is less than the probability threshold value
Will data redefine the client fresh target priority and the fresh target priority corresponding to fresh target probability；
Judge whether the fresh target probability is less than the probability threshold value；
If the fresh target probability is not less than the probability threshold value, the client is determined according to the fresh target priority
Target video server cluster；
If the destination probability is not less than the probability threshold value, the mesh of the client is determined according to the target priority
Mark video server cluster.
Preferably, described is the client distribution video server from the target video server cluster, including：
Transferred described in the case that request carries target video server label and emergency parameter, determine the target
Target video server corresponding to video server label, the target video server label are used to identify the client
Predetermined video server, the emergency parameter are used for the urgency level for characterizing scheduling；
If the target video server is in the target video server cluster and the emergency parameter is used for table
Non-emergent scheduling is levied, extracts the dispatch request；
Generate the dispatch request queue being made of the dispatch request；
Regarded according to the preset flow upper limit value of video server and the target in the target video server cluster
Frequency server ticket, is that the dispatch request in the dispatch request queue distributes video server.
Preferably, it is described according to the preset flow upper limit value of video server in the target video server cluster and
The target video server label, is that the dispatch request in the dispatch request queue distributes video server, including：
According to the preset flow upper limit value of video server in the target video server cluster, calculate the target and regard
The flow control ratio of frequency server cluster；
It is empty dispatch request to choose target video server label substance in the dispatch request queue, and according to described
Flow control ratio distributes video server for the selected dispatch request；
It is described according to the dispatch request reallocation video server that the flow control ratio is unallocated video server
The dispatch request of unallocated video server include the dispatch request queue in do not choose transfer request and it is selected
The dispatch request for transferring unallocated video server in request.
Preferably, described is the client distribution video server from the target video server cluster, including：
Transferred described in the case that request carries target video server label and emergency parameter, determine the target
Target video server corresponding to video server label, the target video server label are used to identify the client
Predetermined video server, the emergency parameter are used for the urgency level for characterizing scheduling；
If the target video server in the target video server cluster, determines the mesh of the dispatch server
Mark operating status；
According to the preset flow upper limit of video server in the target operation state, the target video server cluster
Value, the target video server label and the emergency parameter, video server is distributed for the dispatch request.
Preferably, the target operation state for determining the dispatch server, including：
According to the preset flow upper limit value of video server in the target video server cluster, calculate the target and regard
The target flow control ratio of frequency server cluster；
Based on the present flow rate value of video server in the target video server cluster, the target video clothes are calculated
The present flow rate control ratio of business device cluster；
Judge whether the present flow rate control ratio is equal to target flow control ratio；
If the present flow rate control ratio is equal to target flow control ratio, by pre-set initial launch shape
State is determined as the target operation state of the dispatch server；
If the present flow rate control ratio is obtained at the dispatch server not equal to target flow control ratio
In the operation duration of current operating conditions；
The current scheduling request treating capacity of video server in the target video server cluster is obtained, and according to described
The current scheduling that current scheduling request treating capacity calculates the target video server cluster asks total treating capacity and current tune
Degree request distribution ratio；
If the operation duration is more than operation duration threshold value or the current scheduling asks total treating capacity please more than scheduling
Total treating capacity threshold value or the current scheduling is asked to ask distribution ratio to be not at preset schedule request distribution ratio range, according to
Preset schedule operation condition of server switching law determines the target operation state of the dispatch server；
If the operation duration asks total treating capacity please no more than scheduling no more than operation duration threshold value, the current scheduling
Seek total treating capacity threshold value and the current scheduling asks distribution ratio to be in preset schedule request distribution ratio range, by described in
The current operating conditions of dispatch server are determined as target operation state.
Preferably, it is described according to video server in the target operation state, the target video server cluster
Preset flow upper limit value, the target video server label and the emergency parameter, video is distributed for the dispatch request
Server, including：
Identify the Status Type of the target operation state, the Status Type includes full request responsive state or urgent
Ask responsive state or flow releveling state；
According to the preset flow upper limit value of video server in the target video server cluster, calculate the target and regard
The target flow control ratio of frequency server cluster；
In the case where the Status Type is the full request responsive state, the target video server label is judged
Label substance whether be empty；
If the label substance of the target video server label is sky, it is described to control ratio according to the target flow
Dispatch request distributes video server；
If the label substance of the target video server label is not sky, it is according to the target video server label
The dispatch request distributes video server；
In the case where the Status Type is the flash request responsive state, if the emergency parameter is for characterizing
First emergency parameter of non-emergent scheduling, controls ratio to distribute Video service for the dispatch request according to the target flow
Device；
If the emergency parameter is the second emergency parameter for characterizing intervention schedule, according to the target video server
Label distributes video server for the dispatch request；
In the case where the Status Type is the flow releveling state, the ratio is controlled to be according to the target flow
The dispatch request distributes video server.
A kind of video CD N transfers optimization device, including：Acquisition module, the first determining module, the second determining module and distribution
Module, first determining module include model foundation unit；
The acquisition module, for when receiving the dispatch request of client transmission, being obtained according to the dispatch request
The target association daily record data of the client；
The model foundation unit, for pre-establishing priority identification model；
First determining module, for identifying mould according to the target association daily record data and the priority pre-established
Type, determine the client target priority and the target priority corresponding to destination probability；
Second determining module, for according to the target priority and the destination probability, determining the client
Target video server cluster；
The distribution module, for being the client distribution Video service from the target video server cluster
Device, and the label of the video server distributed is sent to the client.
Preferably, the model foundation unit, is specifically used for：
Transfer the first daily record data of the dispatch server and the second daily record data of default video server；From institute
The scheduling data that history access client is chosen in the first daily record data are stated, and are gone through from second daily record data described in selection
The downloading data of history access client；The historical context day of data and the downloading data as the client is dispatched using described
Will data, and priority tag is added for the historical context daily record data according to pre-set priority mark rule, it is described preferential
Probability of the level label corresponding to including priority and the priority；The data of the historical context daily record data are extracted respectively
The priority aspects of feature and the priority tag；Based on the data characteristics, the priority aspects and default prison
Learning algorithm is superintended and directed, establishes priority identification model.
Compared to the prior art, what the present invention realized has the beneficial effect that：
The above present invention provides a kind of video CD N and transfers optimization method and device, and this method is applied to dispatch server, profit
The target priority and destination probability of client are identified with the priority identification model pre-established, and then to determine client
Target video server cluster, and therefrom for client distribute video server.Based on method disclosed by the invention, Ke Yishi
The rank of other user, and provide targetedly dispatch service to the user according to recognition result.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is attached drawing needed in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
The embodiment of invention, for those of ordinary skill in the art, without creative efforts, can also basis
The attached drawing of offer obtains other attached drawings.
Fig. 1 is the method flow diagram of video user method for optimizing scheduling provided in an embodiment of the present invention；
Fig. 2 is " to pre-establish priority knowledge in video user method for optimizing scheduling step S20 provided in an embodiment of the present invention
The method flow diagram of other model " process；
Fig. 3 is the method flow diagram of video user method for optimizing scheduling step S30 provided in an embodiment of the present invention；
Fig. 4 is " from target video server in video user method for optimizing scheduling step S40 provided in an embodiment of the present invention
In cluster for client distribute video server " method flow diagram；
Fig. 5 is the method flow diagram of video user method for optimizing scheduling step S404 provided in an embodiment of the present invention；
Fig. 6 is " from target video server in video user method for optimizing scheduling step S40 provided in an embodiment of the present invention
In cluster for client distribute video server " process method flow diagram；
Fig. 7 is " to determine dispatch server in video user method for optimizing scheduling step S406 provided in an embodiment of the present invention
Target operation state " process method flow diagram；
Fig. 8 is dispatch server cycle of states illustrated example provided in an embodiment of the present invention；
Fig. 9 is the method flow diagram of video user method for optimizing scheduling step S407 provided in an embodiment of the present invention；
Figure 10 is the structure diagram of video user optimizing scheduling device provided in an embodiment of the present invention.
Embodiment
Below in conjunction with the attached drawing in the embodiment of the present invention, the technical solution in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other without making creative work
Embodiment, belongs to the scope of protection of the invention.
The present invention implements a kind of video CD N of offer and transfers optimization method, and this method is applied to dispatch server, method flow
Figure is as shown in Figure 1, include the following steps：
S10, when receiving the dispatch request of client transmission, the target association day of client is obtained according to dispatch request
Will data；
In the present embodiment, record has the tune of all clients of access scheduling server in the daily record data of dispatch server
Degrees of data, record has the downloading data for all clients for accessing video server in the daily record data of video server.
Uniquely marked in addition, client includes client in the request sent to dispatch server and video server
Know, such as client ip address, also alternative includes request video content ID certainly and other are relevant with user behavior
Parameter, for example, preceding once browsed video ID, since being entered the page to request scheduling when untill in video page/broadcasting
Page number of clicks, institute are averaged duration/finish watching rate etc. using type, historical viewings.Therefore, using client unique mark
The scheduling data of client are obtained from the daily record data of dispatch server, client is obtained from the daily record data of video server
The downloading data at end, and then it is made of scheduling data and downloading data the target association daily record data of client.
S20, according to target association daily record data and the priority identification model pre-established, determines that the target of client is excellent
Destination probability corresponding to first level and target priority；
During specific implementation, the process that " pre-establishes priority identification model " in step S20 can be used specifically
Following steps, method flow diagram are as shown in Figure 2：
S201, transfers the first daily record data of dispatch server and the second daily record data of default video server；
Record has the scheduling data of all clients of access scheduling server in the first daily record data in the present embodiment, the
Record has the downloading data for all clients for accessing video server in two daily record datas.
S202, chooses the scheduling data of history access client from the first daily record data, and from the second daily record data
Choose the downloading data of history access client；
During step S202 is performed, using the known client unique mark of history access client, from tune
Spend in server log data and obtain the scheduling data of history access client, obtaining this from video server daily record data goes through
The downloading data of history access client.
S203, will dispatch the historical context daily record data of data and downloading data as history access client, and according to
Pre-set priority mark rule for historical context daily record data add priority tag, the priority tag include priority with
And the probability corresponding to priority；
During step S203 is performed, pre-set priority mark rule may be configured as " watching non-hot broadcast play and history
The user that duration is averagely browsed less than duration threshold value is normal priority user, and viewing hot broadcast is acute or history averagely browses duration
User more than duration threshold value is high-priority users ", then history access client is extracted from historical context daily record data
In the related data of " video content of viewing (hot broadcast is acute or non-hot broadcast is acute) " and " historical viewings be averaged duration " the two features,
And respective priority label is added according to above-mentioned rule, for example, the priority of high-priority users addition is " 1 ", normal precedence
The priority added of level user be " 0 ", in the process, also need mark history access client as high-priority users or
The probability of person normal priority user.
Certainly, if not including the related data of above-mentioned two feature in historical context daily record data, history is accessed into visitor
The priority tag at family end is determined as sky, i.e., uncertain priority, does not know the probability corresponding to priority yet.
In addition, the feature in pre-set priority mark rule can also select other features, the present embodiment according to being actually needed
It is not specifically limited, can be specifically chosen according to time needs, within protection domain of the embodiment of the present invention.
Further, since data processing amount is very big, can be to the historical context daily record data batch of whole history access clients
Processing：
First, clustering processing is carried out to the historical context daily record data of whole history access clients according to default feature,
Wherein, clustering algorithm can use simplest nearest neighbor algorithm (kNN, k-NearestNeighbor, k arest neighbors sorting algorithm)；Into
And the priority tag of history access client is determined according to cluster result.For example, using above-mentioned two preset feature " viewing
Video content (hot broadcast is acute or non-hot broadcast is acute) " and " historical viewings be averaged duration ", then detect more than quantity in cluster result
History access client of the duration less than duration threshold value that the non-hot broadcast of viewing of threshold value is acute and historical viewings be averaged, then by this part
History access client divides normal priority user into, accordingly, viewing hot broadcast play or historical viewings is averaged duration in duration
More than threshold value high-priority users are divided into history access client, and are preset according to some and be characterized as that history accesses client
End sets probability；And remaining history access client is then the user of uncertain priority, directly priority tag is set
For sky.
S204, extracts the data characteristics of historical context daily record data and the priority aspects of priority tag respectively；
S205, based on data characteristics, priority aspects and default supervised learning algorithm, establishes priority identification model.
It should be noted that default supervised learning algorithm includes but is not limited to decision tree, support vector machines or nerve
Network, user can be specifically chosen according to being actually needed, and the present embodiment does not limit.
It should also be noted that, since normal priority user is generally brush amount user, malicious attack user or machine
People etc., will not be complained, if some client is determined as normal priority user using the priority identification model, and then
The dispatch request of client is scheduled according to this classification results, can be by this after the calling information of this client is received
The classification results of client are adjusted to " high-priority users " by " normal priority user ", and then according to result after adjustment to excellent
First level identification model is modified.
S30, according to target priority and destination probability, determines the target video server cluster of client；
During step S30 is performed, such as, the priority recognition result of client is " this client of the probability exceeded
Hold as high-priority users ", determine the target priority that priority identification model is identified first using destination probability value " x% "
Whether " high priority " be accurate, and then utilizes the correspondence between the priority pre-established and video server cluster, than
Such as, high priority corresponds to the video server cluster in high-quality computer room, to determine the target video server cluster of client.
During specific implementation, step S30 " according to target priority and destination probability, determines that the target of client regards
The process of frequency server cluster ", can specifically use following steps, and method flow diagram is as shown in Figure 3：
S301, judges whether destination probability is less than probability threshold value；If destination probability is less than probability threshold value, step is performed
S302；If destination probability is not less than probability threshold value, step S305 is performed；
S302, the fresh target for redefining client for target association daily record data according to pre-set priority mark rule are excellent
Fresh target probability corresponding to first level and fresh target priority；
In the present embodiment, pre-set priority mark rule can be with consistent for training the rule of priority identification model, in advance
Setting priority mark rule may be configured as " watch that non-hot broadcast is acute and history averagely to browse duration less than the user of duration threshold value be general
Logical priority users, viewing hot broadcast is acute or history averagely browses user of the duration more than duration threshold value and used for high priority
Family ", then extract client at " video content of viewing (hot broadcast is acute or non-hot broadcast is acute) " from target association daily record data and " go through
History browses average duration " related datas of the two features, and according to the corresponding new priority tag of above-mentioned rule addition, for example,
The fresh target priority of high-priority users addition is " 1 ", and the fresh target priority that normal priority user is added is " 0 ",
In the process, mark client is also needed to become high-priority users or the fresh target probability of normal priority user.
Certainly, if not including the related data of above-mentioned two feature in association daily record data, by the new preferential of client
Level label is determined as sky, i.e., uncertain fresh target priority, does not know fresh target probability yet.
In addition, the feature in pre-set priority mark rule can also select other features, the present embodiment according to being actually needed
It is not specifically limited, can be specifically chosen according to time needs, within protection domain of the embodiment of the present invention.
S303, judges whether fresh target probability is less than probability threshold value；If fresh target probability is not less than probability threshold value, perform
Step S304；
S304, the target video server cluster of client is determined according to fresh target priority；
, can be according to the corresponding pass between the priority and video server cluster of foundation during step S304 is performed
System, such as, high priority corresponds to the video server cluster in high-quality computer room, to determine the target corresponding to fresh target priority
Video server cluster.
S305, the target video server cluster of client is determined according to target priority；
, can be according to pair between the priority and video server cluster pre-established during step S305 is performed
It should be related to, such as, high priority corresponds to the video server cluster in high-quality computer room, to determine the target video of target priority
Server cluster.
S40, video server, and the Video service that will be distributed are distributed from target video server cluster for client
The label of device is sent to client；
During specific implementation, " Video service is distributed for client in step S40 from target video server cluster
The process of device ", can specifically use following steps, and method flow diagram is as shown in Figure 4：
S401, in the case where transferring request and carrying target video server label and emergency parameter, determines that target regards
Target video server corresponding to frequency server ticket, the target video server label are true in advance for identifying client
Fixed video server, the emergency parameter are used for the urgency level for characterizing scheduling；
In the present embodiment, predetermined video server is the Video service of speed of download maximum under client angle
Device, can be determined by client using the method to test the speed.The target video server label substance can be sky, for representing not
Determine the video server of speed of download maximum.
And emergency parameter can detect video state to determine by client, such as, in the case of data cached abundance, this
Shi Wuxu obtains video data at once, and emergency parameter is provided for characterizing the first urgent ginseng of non-emergent scheduling by client
Number, for example, being arranged to " 0 ", for another example, in the case where video starts broadcasting, caches deficiency or interim card recovery occurs, needs at this time
Video data is obtained as early as possible, and emergency parameter is provided for the second emergency parameter of characterization intervention schedule by client, for example, setting
Be set to " 1 ", certainly, emergency parameter can also be sky, for represent it is uncertain whether emergent management.
S402, if target video server is in target video server cluster and emergency parameter is non-emergent for characterizing
Scheduling, extracts dispatch request；
S403, generates the dispatch request queue being made of dispatch request；
In the present embodiment, one or more dispatch requests are included in dispatch request queue；
S404, takes according to the preset flow upper limit value of video server in target video server cluster and target video
Business device label, is that the dispatch request in dispatch request queue distributes video server；
During specific implementation, step S404 is " according to the default stream of video server in target video server cluster
The mistake of amount upper limit value and target video server label, for the dispatch request distribution video server in dispatch request queue "
Journey, can specifically use following steps, and method flow diagram is as shown in Figure 5：
S1001, according to the preset flow upper limit value of video server in target video server cluster, calculates target video
The target flow control ratio of server cluster；
During step S1001 is performed, it is assumed that have two video servers in target video server cluster, be denoted as
The flow rate upper limit value of M and N, video server M are 2h, and the flow rate upper limit value of video server N is h, then the target video service
It is 2h that the target flow of device cluster, which controls ratio,:H=2:1.
S1002, it is empty dispatch request to choose target video server label substance in dispatch request queue, and according to mesh
Mark flow control ratio and distribute video server for selected dispatch request；
During step S1002 is performed, it is assumed that the non-emergent dispatch request in dispatch request queue has 90, its
In, the target video server label of 40 dispatch requests is M, and the target video server label of 40 dispatch requests is N, separately
There are 10 dispatch request target video server labels for sky, then it is not 80 empty tune first to choose target video server label
Degree request, the target flow being calculated according to step S1001 control ratio 2:1, it is M by 40 target video server labels
Dispatch request be fully allocated to video server M, extract 20 from the dispatch request that 40 target video server labels are N
Bar distributes to video server N, at this time, there remains dispatch request and 10 targets that 20 target video server labels are N
Video server label is empty dispatch request.
S1003, the dispatch request reallocation Video service that ratio is unallocated video server is controlled according to target flow
Device, request that what the dispatch request of the unallocated video server included not choosing in dispatch request queue transfer and selected
Transfer request in unallocated video server dispatch request；
During step S1003 is performed, the target flow being calculated according to step S1001 controls ratio 2:1, be
The dispatch request of unallocated video server in step S1002 --- the dispatch request that 20 target video server labels are N
It is empty dispatch request reallocation video server with 10 target video server labels, specifically, from 20 target videos
10 are chosen in the dispatch request that server ticket is N and distributes to video server N, and remaining unappropriated 10 targets are regarded
Frequency server ticket is N dispatch requests and 10 target video server labels are that empty dispatch request distributes to video server
M。
In addition, if target video server is not in target video server cluster or emergency parameter is used to characterize promptly
Scheduling, video server is distributed according to flow control policy for dispatch request, and the label of the video server distributed is sent out
Send client corresponding to dispatch request.
In the present embodiment, as existing flow control policy " is dispatch request according to the flow control ratio in dispatcher-controlled territory
The mode of distribution video server ".
During specific implementation, " Video service is distributed for client in step S40 from target video server cluster
The process of device ", can specifically use following steps, and method flow diagram is as shown in Figure 6：
S405, in the case where transferring request and carrying target video server label and emergency parameter, determines that target regards
Target video server corresponding to frequency server ticket, the target video server label are true in advance for identifying client
Fixed video server, the emergency parameter are used for the urgency level for characterizing scheduling；
In the present embodiment, predetermined video server is the Video service of speed of download maximum under client angle
Device, can be determined by client using the method to test the speed.The target video server label substance can be sky, for representing not
Determine the video server of speed of download maximum.
And emergency parameter can detect video state to determine by client, such as, in the case of data cached abundance, this
Shi Wuxu obtains video data at once, and emergency parameter is provided for characterizing the first urgent ginseng of non-emergent scheduling by client
Number, for example, being arranged to " 0 ", for another example, in the case where video starts broadcasting, caches deficiency or interim card recovery occurs, needs at this time
Video data is obtained as early as possible, and emergency parameter is provided for the second emergency parameter of characterization intervention schedule by client, for example, setting
Be set to " 1 ", certainly, emergency parameter can also be sky, for represent it is uncertain whether emergent management.
S406, if target video server in target video server cluster, determines the object run of dispatch server
State；
During specific implementation, the process of " target operation state for determining dispatch server " in step S406 can be with
Following steps specifically are used, method flow diagram is as shown in Figure 7：
S1004, according to the preset flow upper limit value of video server in target video server cluster, calculates target video
The target flow control ratio of server cluster；
During step S1004 is performed, it is assumed that have two video servers in target video server cluster, be denoted as
The flow rate upper limit value of M and N, video server M are 2h, and the flow rate upper limit value of video server N is h, then the target video service
It is 2h that the target flow of device cluster, which controls ratio,:H=2:1.
S1005, based on the present flow rate value of video server in target video server cluster, calculates target video service
The present flow rate control ratio of device cluster；
S1006, judges whether present flow rate control ratio is equal to target flow control ratio；If present flow rate controls ratio
Ratio is controlled equal to target flow, then performs step S1007；If present flow rate control ratio is not equal to target flow control ratio
Example, then perform step S1008；
Pre-set initial operating state, is determined as the target operation state of dispatch server by S1007；
During step S1007 is performed, if present flow rate control ratio is equal to target flow control ratio, then it is assumed that
Dispatch server reaches flow equalization point at the moment, then default initial schedule operation condition of server is determined as object run
State；Dispatch server state that may be present has three kinds " asking responsive state entirely ", " flash request responsive state " and " flow
Releveling state ", its cycle of states figure as shown in figure 8, the cycle of states of a dispatch server from " asking responsive state entirely "
Start, it is " asking responsive state entirely " to preset initial schedule operation condition of server, is determined so that " responsive state will be asked " entirely
For the target operation state of dispatch server.
When dispatch server is in full request responsive state, the dispatch request received can all be taken according to target video
Business device label distribution video server；When dispatch server is in flash request responsive state, only carry tight for characterizing
The dispatch request distribution video server for the first emergency parameter suddenly dispatched；When dispatch server is in flow releveling state,
Then only ratio is controlled to distribute video server for dispatch request according to target flow.
S1008, obtains the operation duration that dispatch server is in current operating conditions；
S1009, obtain target video server cluster in video server current scheduling request treating capacity, and according to work as
The current scheduling that preceding dispatch request treating capacity calculates target video server cluster asks total treating capacity and current scheduling request
Distribute ratio；
During step S1009 is performed, current request processing total amount is that video takes in target video server cluster
The sum of current scheduling request treating capacity of business device, current scheduling request distribution ratio is video clothes in target video server cluster
The ratio of the current scheduling request treating capacity of business device.
S1010, if operation duration is more than operation duration threshold value or current scheduling asks total treating capacity to be more than dispatch request
Total treating capacity threshold value or current scheduling request distribution ratio is not at preset schedule request distribution ratio range, according to default tune
Degree operation condition of server switching law determines the target operation state of dispatch server；
During step S1010 is performed, the cycle of states figure of dispatch server as shown in Figure 8, dispatch server
State that may be present has three kinds " asking responsive state entirely ", " flash request responsive state " and " flow releveling state ".
Assuming that there are two video servers in target video server cluster, M and N are denoted as, the target video server collection
It is 2h that the target flow of group, which controls ratio,:H=2:1, according to default flow excursion amount, it is assumed that be 2, then preset schedule request point
It is 1~4 with ratio range；Assuming that operation duration threshold value is 10S, the total treating capacity threshold value of dispatch request is 1000, then when running
It is long be more than 10S either current scheduling ask total treating capacity be more than 1000 or current scheduling request distribution ratio be not at 1~4 this
During one scope, if current operating conditions are " asking responsive state entirely ", target operation state is " flash request responsive state ",
Similarly, if current state is " flash request responsive state ", target operation state is " flow releveling state " ....
S1011, if operation duration asks total treating capacity to be not more than dispatch request no more than operation duration threshold value, current scheduling
Total treating capacity threshold value and current scheduling request distribute ratio and are in preset schedule request distribution ratio range, by dispatch server
Current operating conditions be determined as target operation state.
S407, according to the preset flow upper limit value of video server in target operation state, target video server cluster,
Target video server label and emergency parameter, video server is distributed for dispatch request；
During specific implementation, step S407 " takes according to video in target operation state, target video server cluster
Preset flow upper limit value, target video server label and the emergency parameter of business device, video server is distributed for dispatch request "
Process, can specifically use following steps, method flow diagram is as shown in Figure 9：
S1012, identifies the Status Type of target operation state, and the Status Type includes full request responsive state or tight
Anxious request responsive state or flow releveling state；
S1013, according to the preset flow upper limit value of video server in target video server cluster, calculates target video
The target flow control ratio of server cluster；
S1014, in the case where Status Type is full request responsive state, judges the label of target video server label
Whether content is empty；If the label substance of target video server label is sky, step S1015 is performed；If target video takes
The label substance of business device label is not sky, then performs step S1016；
S1015, controls ratio to distribute video server for dispatch request according to target flow；
S1016, video server is distributed according to target video server label for dispatch request；
S1017, in the case where Status Type is flash request responsive state, if emergency parameter is non-emergent for characterizing
First emergency parameter of scheduling, controls ratio to distribute video server for dispatch request according to target flow；If emergency parameter is
For characterizing the second emergency parameter of intervention schedule, Video service is distributed for dispatch request according to target video server label
Device；
S1018, in the case where Status Type is flow releveling state, controls ratio please for scheduling according to target flow
Seek distribution video server.
It should be noted that " ratio is controlled according to target flow to adjust in step S1015, step S1017, step S1018
The process of degree request distribution video server " is as follows：
Based on the present flow rate value of video server in target video server cluster, target video server cluster is calculated
Present flow rate control ratio；Compare present flow rate ratio and target flow control ratio, to the video server of underfed
Polishing is carried out, for example, there are two video servers in target video server cluster, is denoted as M and N, the target video server
The flow control ratio of cluster is 2h:H=2:1, video server M present flow rates value is 10, video server N present flow rate values
For 4, then dispatch request is distributed to video server N.Certainly, can be according to pre- if video server N present flow rates value is 5
If allocation rule is allocated dispatch request, such as according to " this default distribution of first video server M rear video servers N "
Rule, then should distribute dispatch request to video server M.
In addition, if target video server is scheduling according to flow control policy not in target video server cluster
Request distribution video server, and the label of the video server distributed is sent to client.
In the present embodiment, as existing flow control policy " is dispatch request according to the flow control ratio in dispatcher-controlled territory
The mode of distribution video server ".
Above step S201~step S205 is only " to be pre-established preferential in the step S20 that the embodiment of the present application discloses
A kind of preferable implementation of level identification model " process, the specific implementation in relation to this process can be according to the demand of oneself
It is any to set, do not limit herein.
Above step S301~step S305 is only step S30 that the embodiment of the present application discloses " according to target priority
And destination probability, determine the target video server cluster of client " a kind of preferable implementation of process, this related process
Specific implementation can arbitrarily be set according to the demand of oneself, do not limit herein.
Above step S401~step S404 is only " to be taken in the step S40 that the embodiment of the present application discloses from target video
Be engaged in device cluster in for client distribute video server " process a kind of preferable implementation, the specific reality in relation to this process
Existing mode can arbitrarily be set according to the demand of oneself, not limited herein.
Above step S1001~step S1003 is only step S404 that the embodiment of the present application discloses " according to target video
The preset flow upper limit value and target video server label of video server in server cluster, are in dispatch request queue
Dispatch request distribution video server " process a kind of preferable implementation, the specific implementation in relation to this process can
Arbitrarily set according to the demand of oneself, do not limited herein.
Above step S405~step S407 is only " to be taken in the step S40 that the embodiment of the present application discloses from target video
Be engaged in device cluster in for client distribute video server " process a kind of preferable implementation, the specific reality in relation to this process
Existing mode can arbitrarily be set according to the demand of oneself, not limited herein.
Above step S1004~step S1011 is only that the step S406 that the embodiment of the present application discloses " determines dispatch service
A kind of preferable implementation of the target operation state of device " process, the specific implementation in relation to this process can be according to oneself
Demand arbitrarily set, do not limit herein.
Above step S1012~step S1018 is only step S407 that the embodiment of the present application discloses " according to object run
The preset flow upper limit value of video server in state, target video server cluster, target video server label and tight
A kind of preferable implementation of anxious parameter, for dispatch request distribution video server " process, the specific implementation in relation to this process
Mode can arbitrarily be set according to the demand of oneself, not limited herein.
Video CD N provided in an embodiment of the present invention transfers optimization method, is known using the priority identification model pre-established
The target priority and destination probability of other client, and then to determine the target video server cluster of client, and therefrom
Video server is distributed for client.Based on method disclosed by the invention, the rank of user can be identified, and according to recognition result
Provide targetedly dispatch service to the user.
The video CD N method for optimizing scheduling provided based on above-described embodiment, the embodiment of the present invention then hold by the corresponding one kind that provides
The device of the above-mentioned video CD N method for optimizing scheduling of row, the structure diagram of video CD N optimizing scheduling devices is as shown in Figure 10,
Including：Acquisition module 10, the first determining module 20, the second determining module 30 and distribution module 40, the first determining module 20 include
Model foundation unit 201；
Acquisition module 10, for when receiving the dispatch request of client transmission, client to be obtained according to dispatch request
Target association daily record data；
Model foundation unit 201, for pre-establishing priority identification model；
First determining module 20, for according to target association daily record data and the priority identification model that pre-establishes, really
Determine the destination probability corresponding to the target priority and target priority of client；
Second determining module 30, for according to target priority and destination probability, determining the target video service of client
Device cluster；
Distribution module 40, for being that client distributes video server from target video server cluster, and will divide
The label for the video server matched somebody with somebody is sent to client.
In some other embodiment, model foundation unit 201, is specifically used for：
Transfer the first daily record data of dispatch server and the second daily record data of default video server；From first day
The scheduling data of history access client are chosen in will data, and are chosen from the second daily record data under history access client
Carry data；The historical context daily record data of data and downloading data as client will be dispatched, and is marked according to pre-set priority
Rule adds priority tag for historical context daily record data, and the priority tag is included corresponding to priority and priority
Probability；The data characteristics of historical context daily record data and the priority aspects of priority tag are extracted respectively；Based on data
Feature, priority aspects and default supervised learning algorithm, establish priority identification model.
In some other embodiment, the second determining module 30, is specifically used for：
Judge whether destination probability is less than probability threshold value；If destination probability is less than probability threshold value, according to pre-set priority mark
Note rule is redefined corresponding to the fresh target priority and fresh target priority of client for target association daily record data
Fresh target probability；Judge whether fresh target probability is less than probability threshold value；If fresh target probability is not less than probability threshold value, according to new mesh
Mark priority determines the target video server cluster of client；If destination probability is not less than probability threshold value, preferred according to target
The target video server cluster of the definite client of level.
In some other embodiment, distribution module 40, is specifically used for：
In the case where transferring request and carrying target video server label and emergency parameter, target video service is determined
Target video server corresponding to device label, for identifying, client is predetermined to be regarded the target video server label
Frequency server, the emergency parameter are used for the urgency level for characterizing scheduling；If target video server is in target video server
In cluster and emergency parameter is used to characterize non-emergent scheduling, extracts dispatch request；The scheduling that generation is made of dispatch request please
Ask queue；According to the preset flow upper limit value and target video server mark of video server in target video server cluster
Label, are that the dispatch request in dispatch request queue distributes video server.
In some other embodiment, distribution module 40, is specifically used for：
In the case where transferring request and carrying target video server label and emergency parameter, target video service is determined
Target video server corresponding to device label, for identifying, client is predetermined to be regarded the target video server label
Frequency server, the emergency parameter are used for the urgency level for characterizing scheduling；If target video server is in target video server
In cluster, the target operation state of dispatch server is determined；According to video in target operation state, target video server cluster
Preset flow upper limit value, target video server label and the emergency parameter of server, Video service is distributed for dispatch request
Device.
Video CD N provided in an embodiment of the present invention transfers optimization device, is known using the priority identification model pre-established
The target priority and destination probability of other client, and then to determine the target video server cluster of client, and therefrom
Video server is distributed for client.Based on device disclosed by the invention, the rank of user can be identified, and according to recognition result
Provide targetedly dispatch service to the user.
Detailed Jie has been carried out to video user method for optimizing scheduling, device and dispatch server provided by the present invention above
Continue, specific case used herein is set forth the principle of the present invention and embodiment, and the explanation of above example is only
It is the method and its core concept for being used to help understand the present invention；Meanwhile for those of ordinary skill in the art, according to this hair
Bright thought, there will be changes in specific embodiments and applications, in conclusion this specification content should not manage
Solve as limitation of the present invention.
It should be noted that each embodiment in this specification is described by the way of progressive, each embodiment weight
Point explanation is all difference with other embodiment, between each embodiment identical similar part mutually referring to.
For device disclosed in embodiment, since it is corresponded to the methods disclosed in the examples, so fairly simple, the phase of description
Part is closed referring to method part illustration.
It should also be noted that, herein, relational terms such as first and second and the like are used merely to one
Entity or operation are distinguished with another entity or operation, without necessarily requiring or implying between these entities or operation
There are any actual relationship or order.Moreover, term " comprising ", "comprising" or its any other variant are intended to contain
Lid non-exclusive inclusion, so that the key element that process, method, article or equipment including a series of elements are intrinsic,
Either further include as these processes, method, article or the intrinsic key element of equipment.In the absence of more restrictions,
The key element limited by sentence "including a ...", it is not excluded that in the process including the key element, method, article or equipment
In also there are other identical element.
The foregoing description of the disclosed embodiments, enables professional and technical personnel in the field to realize or use the present invention.
A variety of modifications to these embodiments will be apparent for those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, it is of the invention
The embodiments shown herein is not intended to be limited to, and is to fit to and the principles and novel features disclosed herein phase one
The most wide scope caused.
Claims (10)
1. a kind of video CD N transfers optimization method, it is characterised in that applied to dispatch server, including：
When receiving the dispatch request of client transmission, the target association day of the client is obtained according to the dispatch request
Will data；
According to the target association daily record data and the priority identification model pre-established, determine that the target of the client is excellent
Destination probability corresponding to first level and the target priority；
According to the target priority and the destination probability, the target video server cluster of the client is determined；
It is that the client distributes video server, and the Video service that will be distributed from the target video server cluster
The label of device is sent to the client.
2. according to the method described in claim 1, it is characterized in that, pre-establish priority identification model, including：
Transfer the first daily record data of the dispatch server and the second daily record data of default video server；
The scheduling data of history access client are chosen from first daily record data, and are selected from second daily record data
Take the downloading data of the history access client；
The historical context daily record data of data and the downloading data as the history access client is dispatched using described, and is pressed
Priority tag is added for the historical context daily record data according to pre-set priority mark rule, the priority tag includes excellent
Probability corresponding to first level and the priority；
The data characteristics of the historical context daily record data and the priority aspects of the priority tag are extracted respectively；
Based on the data characteristics, the priority aspects and default supervised learning algorithm, priority identification model is established.
It is 3. according to the method described in claim 1, it is characterized in that, described general according to the target priority and the target
Rate, determines the target video server cluster of the client, including：
Judge whether the destination probability is less than probability threshold value；
It is the target association daily record number according to pre-set priority mark rule if the destination probability is less than the probability threshold value
According to the fresh target probability corresponding to the fresh target priority and the fresh target priority for redefining the client；
Judge whether the fresh target probability is less than the probability threshold value；
If the fresh target probability is not less than the probability threshold value, the mesh of the client is determined according to the fresh target priority
Mark video server cluster；
If the destination probability is not less than the probability threshold value, determine that the target of the client regards according to the target priority
Frequency server cluster.
4. according to the method described in claim 1, it is characterized in that, described is described from the target video server cluster
Client distributes video server, including：
Transferred described in the case that request carries target video server label and emergency parameter, determine the target video
Target video server corresponding to server ticket, the target video server label are advance for identifying the client
Definite video server, the emergency parameter are used for the urgency level for characterizing scheduling；
If the target video server is in the target video server cluster and the emergency parameter is non-for characterizing
Intervention schedule, extracts the dispatch request；
Generate the dispatch request queue being made of the dispatch request；
Taken according to the preset flow upper limit value of video server in the target video server cluster and the target video
Business device label, is that the dispatch request in the dispatch request queue distributes video server.
It is 5. according to the method described in claim 4, it is characterized in that, described according to video in the target video server cluster
The preset flow upper limit value of server and the target video server label, are the tune in the dispatch request queue
Degree request distribution video server, including：
According to the preset flow upper limit value of video server in the target video server cluster, the target video clothes are calculated
The target flow control ratio of business device cluster；
It is empty dispatch request to choose target video server label substance in the dispatch request queue, and according to the target
Flow control ratio distributes video server for the selected dispatch request；
The dispatch request reallocation video server that ratio is unallocated video server is controlled according to the target flow, it is described
The dispatch request of unallocated video server include the dispatch request queue in do not choose transfer request and it is selected
The dispatch request for transferring unallocated video server in request.
6. according to the method described in claim 1, it is characterized in that, described is described from the target video server cluster
Client distributes video server, including：
Transferred described in the case that request carries target video server label and emergency parameter, determine the target video
Target video server corresponding to server ticket, the target video server label are advance for identifying the client
Definite video server, the emergency parameter are used for the urgency level for characterizing scheduling；
If the target video server in the target video server cluster, determines the target fortune of the dispatch server
Row state；
According to the preset flow upper limit value of video server in the target operation state, the target video server cluster,
The target video server label and the emergency parameter, video server is distributed for the dispatch request.
7. the according to the method described in claim 6, it is characterized in that, object run shape for determining the dispatch server
State, including：
According to the preset flow upper limit value of video server in the target video server cluster, the target video clothes are calculated
The target flow control ratio of business device cluster；
Based on the present flow rate value of video server in the target video server cluster, the target video server is calculated
The present flow rate control ratio of cluster；
Judge whether the present flow rate control ratio is equal to target flow control ratio；
If the present flow rate control ratio is equal to target flow control ratio, and pre-set initial operating state is true
It is set to the target operation state of the dispatch server；
If the present flow rate control ratio obtains the dispatch server and is in and work as not equal to target flow control ratio
The operation duration of preceding operating status；
The current scheduling request treating capacity of video server in the target video server cluster is obtained, and according to described current
The current scheduling that dispatch request treating capacity calculates the target video server cluster asks total treating capacity and current scheduling please
Seek distribution ratio；
If the operation duration is more than operation duration threshold value or the current scheduling asks total treating capacity total more than dispatch request
Treating capacity threshold value or current scheduling request distribution ratio are not at preset schedule request distribution ratio range, according to default
Dispatch server operating status switching law determines the target operation state of the dispatch server；
If the operation duration asks total treating capacity total no more than dispatch request no more than operation duration threshold value, the current scheduling
Treating capacity threshold value and current scheduling request distribute ratio and are in preset schedule request distribution ratio range, by the scheduling
The current operating conditions of server are determined as target operation state.
8. according to the method described in claim 6, it is characterized in that, described regard according to the target operation state, the target
The preset flow upper limit value of video server, the target video server label and the urgent ginseng in frequency server cluster
Number, video server is distributed for the dispatch request, including：
Identify the Status Type of the target operation state, the Status Type includes full request responsive state or flash request
Responsive state or flow releveling state；
According to the preset flow upper limit value of video server in the target video server cluster, the target video clothes are calculated
The target flow control ratio of business device cluster；
In the case where the Status Type is the full request responsive state, the mark of the target video server label is judged
Whether label content is empty；
If the label substance of the target video server label is sky, it is the scheduling to control ratio according to the target flow
Request distribution video server；
It is described according to the target video server label if the label substance of the target video server label is not sky
Dispatch request distributes video server；
In the case where the Status Type is the flash request responsive state, if the emergency parameter is non-tight for characterizing
The first emergency parameter suddenly dispatched, controls ratio to distribute video server for the dispatch request according to the target flow；
If the emergency parameter is the second emergency parameter for characterizing intervention schedule, according to the target video server label
Video server is distributed for the dispatch request；
In the case where the Status Type is the flow releveling state, it is described to control ratio according to the target flow
Dispatch request distributes video server.
9. a kind of video CD N transfers optimization device, it is characterised in that including：Acquisition module, the first determining module, second determine
Module and distribution module, first determining module include model foundation unit；
The acquisition module, for when receive client transmission dispatch request when, according to the dispatch request obtain described in
The target association daily record data of client；
The model foundation unit, for pre-establishing priority identification model；
First determining module, for according to the target association daily record data and the priority identification model that pre-establishes,
Determine the client target priority and the target priority corresponding to destination probability；
Second determining module, for according to the target priority and the destination probability, determining the mesh of the client
Mark video server cluster；
The distribution module, for being the client distribution video server from the target video server cluster, and
The label of the video server distributed is sent to the client.
10. device according to claim 9, it is characterised in that the model foundation unit, is specifically used for：
Transfer the first daily record data of the dispatch server and the second daily record data of default video server；From described
The scheduling data of history access client are chosen in one daily record data, and chooses the history from second daily record data and visits
Ask the downloading data of client；The historical context daily record number of data and the downloading data as the client is dispatched using described
According to, and according to pre-set priority mark rule priority tag, the priority mark are added for the historical context daily record data
Label include priority and the probability corresponding to the priority；The data characteristics of the historical context daily record data is extracted respectively
And the priority aspects of the priority tag；Learned based on the data characteristics, the priority aspects and default supervision
Algorithm is practised, establishes priority identification model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711486733.0A CN107948004B (en) | 2017-12-29 | 2017-12-29 | Video CDN (content delivery network) calling optimization method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711486733.0A CN107948004B (en) | 2017-12-29 | 2017-12-29 | Video CDN (content delivery network) calling optimization method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107948004A true CN107948004A (en) | 2018-04-20 |
| CN107948004B CN107948004B (en) | 2021-06-22 |
Family
ID=61937126
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711486733.0A Active CN107948004B (en) | 2017-12-29 | 2017-12-29 | Video CDN (content delivery network) calling optimization method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107948004B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109547546A (en) * | 2018-11-12 | 2019-03-29 | 网宿科技股份有限公司 | A kind of dispatching method and control centre's server of request task |
| CN109768879A (en) * | 2018-12-14 | 2019-05-17 | 北京爱奇艺科技有限公司 | The determination method, apparatus and server of target service server |
| CN110213623A (en) * | 2019-06-03 | 2019-09-06 | 上海哔哩哔哩科技有限公司 | Video method for uploading, video upload device and computer readable storage medium |
| CN112134811A (en) * | 2020-09-30 | 2020-12-25 | 安徽极玩云科技有限公司 | CDN cloud platform flow scheduling method |
| WO2021142971A1 (en) * | 2020-01-19 | 2021-07-22 | 深圳壹账通智能科技有限公司 | Transmission rate control method and apparatus, computer system, and readable storage medium |
| CN113556291A (en) * | 2021-07-08 | 2021-10-26 | 北京奇艺世纪科技有限公司 | Flow tracking method, device, equipment and computer readable medium |
| WO2022242498A1 (en) * | 2021-05-21 | 2022-11-24 | 北京字跳网络技术有限公司 | Cdn scheduling method and apparatus, device and storage medium |
| CN118921406A (en) * | 2024-07-09 | 2024-11-08 | 北京火山引擎科技有限公司 | Data access method, device, equipment and storage medium in content distribution network |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1801819A (en) * | 2005-01-04 | 2006-07-12 | 国际商业机器公司 | Confirming apparatus of priority, service treating and allocating device, and control method |
| CN101261592A (en) * | 2007-03-07 | 2008-09-10 | 国际商业机器公司 | Method and apparatus for scheduling task requests |
| US7680897B1 (en) * | 2003-04-08 | 2010-03-16 | Novell, Inc. | Methods and systems for managing network traffic |
| CN105898402A (en) * | 2015-11-13 | 2016-08-24 | 乐视云计算有限公司 | Network scheduling method and system |
| CN105898383A (en) * | 2015-11-26 | 2016-08-24 | 乐视云计算有限公司 | Bandwidth allocation method and system |
| CN105897837A (en) * | 2015-12-07 | 2016-08-24 | 乐视云计算有限公司 | Content distribution task submitting method and system |
| CN106652545A (en) * | 2015-10-28 | 2017-05-10 | 赵依军 | Method and system for optimizing processing of parking space reservation requests |
-
2017
- 2017-12-29 CN CN201711486733.0A patent/CN107948004B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7680897B1 (en) * | 2003-04-08 | 2010-03-16 | Novell, Inc. | Methods and systems for managing network traffic |
| CN1801819A (en) * | 2005-01-04 | 2006-07-12 | 国际商业机器公司 | Confirming apparatus of priority, service treating and allocating device, and control method |
| CN101261592A (en) * | 2007-03-07 | 2008-09-10 | 国际商业机器公司 | Method and apparatus for scheduling task requests |
| CN106652545A (en) * | 2015-10-28 | 2017-05-10 | 赵依军 | Method and system for optimizing processing of parking space reservation requests |
| CN105898402A (en) * | 2015-11-13 | 2016-08-24 | 乐视云计算有限公司 | Network scheduling method and system |
| CN105898383A (en) * | 2015-11-26 | 2016-08-24 | 乐视云计算有限公司 | Bandwidth allocation method and system |
| CN105897837A (en) * | 2015-12-07 | 2016-08-24 | 乐视云计算有限公司 | Content distribution task submitting method and system |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109547546A (en) * | 2018-11-12 | 2019-03-29 | 网宿科技股份有限公司 | A kind of dispatching method and control centre's server of request task |
| CN109547546B (en) * | 2018-11-12 | 2020-06-05 | 网宿科技股份有限公司 | Scheduling method of request task and scheduling center server |
| CN109768879A (en) * | 2018-12-14 | 2019-05-17 | 北京爱奇艺科技有限公司 | The determination method, apparatus and server of target service server |
| CN110213623A (en) * | 2019-06-03 | 2019-09-06 | 上海哔哩哔哩科技有限公司 | Video method for uploading, video upload device and computer readable storage medium |
| WO2021142971A1 (en) * | 2020-01-19 | 2021-07-22 | 深圳壹账通智能科技有限公司 | Transmission rate control method and apparatus, computer system, and readable storage medium |
| CN112134811A (en) * | 2020-09-30 | 2020-12-25 | 安徽极玩云科技有限公司 | CDN cloud platform flow scheduling method |
| CN112134811B (en) * | 2020-09-30 | 2022-08-09 | 安徽极玩云科技有限公司 | CDN cloud platform flow scheduling method |
| WO2022242498A1 (en) * | 2021-05-21 | 2022-11-24 | 北京字跳网络技术有限公司 | Cdn scheduling method and apparatus, device and storage medium |
| CN113556291A (en) * | 2021-07-08 | 2021-10-26 | 北京奇艺世纪科技有限公司 | Flow tracking method, device, equipment and computer readable medium |
| CN113556291B (en) * | 2021-07-08 | 2024-04-30 | 北京奇艺世纪科技有限公司 | Flow tracking method, device, equipment and computer readable medium |
| CN118921406A (en) * | 2024-07-09 | 2024-11-08 | 北京火山引擎科技有限公司 | Data access method, device, equipment and storage medium in content distribution network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107948004B (en) | 2021-06-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107948004A (en) | A kind of video CD N transfers optimization method and device | |
| CN112988390B (en) | Computing power resource allocation method and device | |
| CN105637925B (en) | Dynamic frequency and power resource allocation with refinement tactical management | |
| US8191070B2 (en) | Dynamic resource allocation | |
| WO2017080172A1 (en) | Network scheduling method and system | |
| CN109343801A (en) | Data storage method, device, and computer-readable storage medium | |
| CN105337901A (en) | Router intelligent bandwidth allocation method and device | |
| US20240028415A1 (en) | Instance deployment method and apparatus, cloud system, computing device, and storage medium | |
| CN104796422A (en) | Online customer service staff equilibrium assignment method and online customer service staff equilibrium assignment device | |
| CN109189578B (en) | Storage server allocation method, device, management server and storage system | |
| CN101080001B (en) | Device for realizing balance of media content in network TV system and its method | |
| CN106851343A (en) | For the method and apparatus of net cast | |
| CN106933868A (en) | A kind of method and data server for adjusting data fragmentation distribution | |
| CN113543160B (en) | 5G slice resource allocation method, device, computing equipment and computer storage medium | |
| CN106445690A (en) | Dynamic scheduling method and device for virtual machine resource | |
| CN111131841A (en) | Live indirect access method and device, electronic equipment and storage medium | |
| CN107733805A (en) | Business load dispatching method and device | |
| CN107104956B (en) | A remote sensing distribution method and system based on user priority | |
| CN107295358B (en) | A kind of 3D Streaming Media storage method under cloud environment | |
| CN107612847A (en) | A kind of bandwidth allocation methods, device and video monitoring system | |
| CN103713852B (en) | A kind of information processing method, service platform and electronic equipment | |
| CN108093272A (en) | A kind of video CD N method for optimizing scheduling and device | |
| CN119232741A (en) | Distributed network resource optimal scheduling method and system under load balancing strategy | |
| CN108184149A (en) | A kind of video CD N method for optimizing scheduling and device | |
| CN106161284A (en) | A kind of resource regulating method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |