CN107295110A - Processing method, fringe node, service server and the system of calculating task - Google Patents
Processing method, fringe node, service server and the system of calculating task Download PDFInfo
- Publication number
- CN107295110A CN107295110A CN201710703806.0A CN201710703806A CN107295110A CN 107295110 A CN107295110 A CN 107295110A CN 201710703806 A CN201710703806 A CN 201710703806A CN 107295110 A CN107295110 A CN 107295110A
- Authority
- CN
- China
- Prior art keywords
- task
- fringe node
- service server
- goal
- calculating
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- 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
-
- 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)
- Information Transfer Between Computers (AREA)
- Computer And Data Communications (AREA)
Abstract
The invention discloses a kind of processing method of calculating task, fringe node, service server and system, wherein, methods described includes：The access request that subscription client is sent is received, the access request points to pending calculating task；Judge whether in the complete calculating task of processing locality；If it is not, the mission bit stream of the calculating task is sent to service server；At least include the task total amount of the calculating task in the mission bit stream；The task allocation information of the service server feedback is received, the task allocation information includes goal task amount, and the goal task amount is used to characterize the task amount for being responsible for processing in the calculating task by the service server of the task；Goal task equal with the goal task amount in the calculating task is uploaded to the service server, and the task in calculating task described in processing locality in addition to the goal task.The technical scheme that the application is provided, it is possible to increase the treatment effeciency of calculating task.
Description
Technical field
The present invention relates to Internet technical field, more particularly to a kind of processing method of calculating task, fringe node, business
Server and system.
Background technology
With continuing to develop for Internet technology, the information in network also increases in explosion type.In order that it is whole to obtain user
End can rapidly get required information from server, can pass through content distributing network (Content at present
Delivery Network, CDN) server and user terminal are attached.
Some fringe nodes can be included in CDN, physical computer can be disposed on each fringe node, can also be
Multiple virtual machines are divided in physical computer.Generally, the fringe node in CDN can be divided according to geographical position.
When user terminal initiates to access to server, corresponding side is accessed in geographical position that can be according to residing for user terminal nearby
Edge node.So, user terminal just can get the required network information in preferable network environment.
Currently, the request that user terminal is sent to fringe node, it is sometimes desirable to the corresponding calculating task of fringe node processing.
For example, user terminal to fringe node when asking video data, may be in the absence of regarding needed for user terminal in fringe node
Frequency form, therefore, fringe node need to enter local video data row format conversion.Now, form conversion just can be side
The calculating task of processing needed for edge node.However, because fringe node is generally connected with multiple user terminals, therefore fringe node
Need while providing network data, the calculating task of one side processing format conversion to other user terminals.So, fringe node
Computing resource may be consumed largely, is not only resulted in the other user terminals of no normal direction and is provided smooth network number
According to can also reduce the treatment effeciency of calculating task.
The content of the invention
The purpose of the application is to provide a kind of processing method of calculating task, fringe node, service server and system,
The treatment effeciency of calculating task can be improved.
To achieve the above object, on the one hand the application provides a kind of processing method of calculating task, and methods described includes：Connect
The access request that subscription client is sent is received, the access request points to pending calculating task；Judge whether originally be located in
The complete calculating task of reason；If it is not, the mission bit stream of the calculating task is sent to service server；The task letter
At least include the task total amount of the calculating task in breath；The task allocation information of the service server feedback is received, it is described
Task allocation information includes goal task amount, and the goal task amount is used to characterize in the calculating task to be taken by the business
Business device is responsible for the task amount of the task of processing；Goal task equal with the goal task amount in the calculating task is uploaded
The extremely service server, and the task in calculating task described in processing locality in addition to the goal task.
To achieve the above object, on the other hand the application also provides a kind of fringe node, and the fringe node includes：Access
Request reception unit, for receiving the access request that subscription client is sent, the access request is pointed to pending calculating and appointed
Business；Mission bit stream reporting unit, for judging whether in the complete calculating task of processing locality；If it is not, being calculated described
The mission bit stream of task is sent to service server；At least include the task total amount of the calculating task in the mission bit stream；
Task allocation information receiving unit, the task allocation information for receiving the service server feedback, the task distribution letter
Breath includes goal task amount, and the goal task amount is used to characterize in the calculating task to be responsible for by the service server
The task amount of the task of reason；Goal task uploading unit, for will be equal with the goal task amount in the calculating task
Goal task is uploaded to the service server, and the appointing in addition to the goal task in calculating task described in processing locality
Business.
To achieve the above object, on the other hand the application also provides a kind of processing method of calculating task, methods described bag
Include：The mission bit stream that at least one fringe node is sent is received, the mission bit stream includes pending in the fringe node
Calculating task task total amount；It is respectively that each described fringe node distributes corresponding target times based on the mission bit stream
Business amount；Receive the goal task that each described fringe node is uploaded；Wherein, the goal task that the fringe node is uploaded
Task amount goal task amount corresponding with the fringe node is equal；Handling the goal task that the fringe node is uploaded
Afterwards, corresponding result is fed back into the fringe node.
To achieve the above object, on the other hand the application also provides a kind of service server, and the service server includes：
Mission bit stream receiving unit, for receiving the mission bit stream that at least one fringe node is sent, the mission bit stream includes institute
State the task total amount of calculating task pending in fringe node；Goal task amount allocation unit, for based on task letter
Breath, is respectively that each described fringe node distributes corresponding goal task amount；Goal task receiving unit, for receiving each institute
State the goal task of fringe node upload；Wherein, the task amount for the goal task that the fringe node is uploaded and the side
The corresponding goal task amount of edge node is equal；Result feedback unit, for handling the mesh that the fringe node is uploaded
After mark task, corresponding result is fed back into the fringe node.
To achieve the above object, on the other hand the application also provides a kind of processing system of calculating task, the processing system
System includes subscription client, fringe node and service server, wherein：The subscription client, for the edge section
Point sends access request, and the access request points to pending calculating task；The fringe node, for judging whether at this
The complete calculating task of ground processing；If it is not, the mission bit stream of the calculating task is sent to the service server；Institute
Stating at least includes the task total amount of the calculating task in mission bit stream；Receive the task distribution letter of the service server feedback
Breath, the task allocation information includes goal task amount, and the goal task amount is used to characterize in the calculating task by institute
State the task amount that service server is responsible for the task of processing；By target equal with the goal task amount in the calculating task
Task is uploaded to the service server, and the task in calculating task described in processing locality in addition to the goal task；
The service server, is respectively each described fringe node for the mission bit stream sent based on fringe node each described
Distribute corresponding goal task amount；Receive the goal task that each described fringe node is uploaded；Handling the fringe node
After the goal task of upload, corresponding result is fed back into the fringe node.
Therefore, in the technical scheme that the application is provided, when fringe node receives the access that subscription client is sent
During request, the whole calculating tasks that can be pointed in the processing locality access request are can decide whether first.If local
Whole calculating tasks can not be handled, then the mission bit stream of the calculating task can be sent to service server, to cause industry
A part of task in server process of the being engaged in calculating task.So, fringe node with service server by cooperateing with processing should
Calculating task, can mitigate the local computing resource for needing to put into, so, on the one hand not interfere with and carried to other subscription clients
For network data, on the other hand, the calculating task is simultaneously and concurrently handled in local and service server, it is possible to increase calculate
The treatment effeciency of task.
Brief description of the drawings
Technical scheme in order to illustrate the embodiments of the present invention more clearly, makes required in being described below to embodiment
Accompanying drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the present invention, for
For those of ordinary skill in the art, on the premise of not paying creative work, other can also be obtained according to these accompanying drawings
Accompanying drawing.
Fig. 1 is the process flow figure of the calculating task in the embodiment of the present invention one；
Fig. 2 is the structural representation of system in the present invention；
Fig. 3 is the functional block diagram of fringe node in the embodiment of the present invention two；
Fig. 4 is the process flow figure of calculating task in the embodiment of the present invention three；
Fig. 5 is the functional block diagram of service server in the embodiment of the present invention four.
Embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with accompanying drawing to embodiment party of the present invention
Formula is described in further detail.
Embodiment one
One embodiment of the application provides a kind of processing method of calculating task, and this method can apply in CDN
On fringe node.Referring to Fig. 1, methods described may comprise steps of.
S11：The access request that subscription client is sent is received, the access request points to pending calculating task.
In the present embodiment, user can initiate the access request for a certain service by subscription client.The use
Family client can be connected with CDN, can be on CDN some sides in order to improve the speed that subscription client obtains network data
In edge node, the fringe node that same geographical position is in the subscription client is connected with the subscription client.This
Sample, when subscription client initiates the access request, the fringe node being connected with the subscription client just can receive this
Access request.
In the present embodiment, the access request can be initiated for Internet resources.The Internet resources are for example
Can be the data that video, voice, picture, word etc. need to obtain from CDN.Specifically, the access request can be pin
To the load request of HD video.In subscription client, the thumbnail of some video can be shown, when user clicks on the breviary
After figure, subscription client just can send the access request for loading HD video to fringe node.
In the present embodiment, fringe node is after the access request is received, and can whether there is in local search should
The HD video that access request is pointed to.If it does, the HD video just can be fed back to subscription client；If locally only
There is the video (such as SD video) of other forms, then fringe node needs that the video of other forms is carried out at transcoding
Reason, to obtain corresponding HD video.Now, fringe node just needs to handle the calculating task of transcoding.The calculating task just may be used
To be the pending calculating task described in step S11.
Certainly, it is necessary to which explanation, the embodiment of above-mentioned transcoding is only example illustrative embodiments in the application, in reality
In the application scenarios of border, the pending calculating task can also be image cropping, content identification, image recognition, data processing
Etc. a series of the needing to take computing resource of the tasks, the application is to this and is not construed as limiting.
S13：Judge whether in the complete calculating task of processing locality；If it is not, the task of the calculating task is believed
Breath is sent to service server；At least include the task total amount of the calculating task in the mission bit stream.
In the present embodiment, whether fringe node is after the access request is received, it can be determined that can be local
The complete calculating task of processing.If the current computing resource of the fringe node is abundant, can be complete in processing locality
The calculating task, and result is fed back into subscription client.If when the current computing resource of the fringe node not
Plentifully, if the computing resource of redundancy is all used to handle the calculating task, it can lead to not glibly to other user clients
End provides network data.In this case, fringe node can ask the service server in CDN so that service server
The processing calculating task is cooperateed with fringe node.The service server can be that multiple fringe nodes are connected and possess higher
The server of computing capability.The service server does not generally directly carry out data interaction, but and edge with subscription client
Node carries out data interaction.Specifically, the structural representation of service server, fringe node and subscription client refers to figure
2.As shown in Fig. 2 fringe node carries out data interaction with subscription client, and service server then carries out data with fringe node
Interaction.
In the present embodiment, can be in the complete calculating task of processing locality in judgement, fringe node can be obtained
Current time locally available computing resource.Wherein, the computing resource can include hard disk bus resource, memory source, in
At least one of central processor (CPU) resource.Then, the fringe node may determine that the current time is locally available
Can computing resource handle the task total amount of the calculating task.If can, then just judge can it is complete in processing locality described in
Calculating task；If can not, then just judging can not be in the complete calculating task of processing locality.Typically, fringe node,
Recorded according to the performance of fringe node and to the processing of historic task, remaining calculate of the fringe node can be summed up and provided
Mapping relations between source and the task amount that can be handled.It is local that the mapping relations can be stored in fringe node.So work as side
Edge node just can read this and reflect when judging that can current time available computing resource handle the task total amount of calculating task
Relation is penetrated, and determines the actual task amount of the calculating task that can be handled corresponding with current time available computing resource.Should
Actual task amount is if less than the task total amount of the calculating task, and just showing can not be in the whole calculating task of processing locality.
If the actual task amount is more than or equal to the task total amount of the calculating task, just showing can be in processing locality whole
Calculating task.
In addition, in the present embodiment, can be in the complete calculating task of processing locality in judgement, fringe node may be used also
So that current time locally available computing resource to be compared with specified resource threshold.The specified resource threshold can be pre-
The resource numerical value or resource ratio value first set, this specifies resource threshold to can be used for judging the remaining calculating money of fringe node
Whether source is sufficient.So, when the computing resource is more than the specified resource threshold, just show that fringe node is current vacant
Computing resource is more sufficient, then just judging can be in the complete calculating task of processing locality.When the computing resource is less than
Or during equal to the specified resource threshold, just show that the current vacant computing resource of fringe node is inadequate in dealing with whole meters
Calculation task, then just judging can not be in the complete calculating task of processing locality.
It should be noted that can generally possess header file field in access request, generally may be used in header file field
Type, access request of calculating task with the page address cited in the dated access request, pointed by the access request
The information such as the address of pointed Internet resources, the version of subscription client for sending the access request.So, fringe node exists
When parsing the header file field, just can know should carry out the processing procedure of which kind of calculating task for which Internet resources,
So that it is determined that the type of calculating task and corresponding task total amount.
In the present embodiment, just can be to when fringe node judges locally handle complete calculating task
Service server request is assisted.Now, the fringe node can send the mission bit stream of the calculating task to the industry
It is engaged at server.The task total amount of the calculating task can be included in the mission bit stream, so that service server is known
Dawn, current amount to needed task amount to be processed.Further, since service server is connected with multiple fringe nodes, in order to recognize
The fringe node of mission bit stream is sent, the mark of fringe node can also be included in the mission bit stream.The mark is for example
Can be preset numbers of the fringe node in CDN, or the corresponding IP address of fringe node etc. can uniquely characterize it is described
The character string of fringe node.Certainly, in actual application, more information, example can also be included in the mission bit stream
Such as, the type of calculating task can be included in the mission bit stream, the type is such as can include transcoding, cutting, identification.
S15：The task allocation information of the service server feedback is received, the task allocation information includes target and appointed
Business amount, the goal task amount is used to characterize the task for being responsible for processing in the calculating task by the service server of the task
Amount.
In the present embodiment, service server is after the mission bit stream that the fringe node is sent is received, Ke Yicong
The task total amount of the calculating task is extracted in the mission bit stream.Now, service server can be directed to the task total amount, and
With reference to the available computational resources of self residual, it is determined that the task amount of the task by itself being responsible for processing.The task amount can be
Specific data value or ratio.For example, service server can be responsible for handling in the task total amount of the calculating task
300MB task amount in 80% task amount, or the task total amount of the processing calculating task.Wherein, by the business
Server is responsible for the task amount of the task of processing, just can be the goal task amount described in step S15.
In the present embodiment, service server can determine the goal task using the algorithm of 0-1 knapsack models
Amount.Specifically, can receive simultaneously can in the mission bit stream that multiple fringe nodes are reported, each mission bit stream for service server
With the task total amount of the calculating task handled including corresponding edge node requesting cooperative.Now, in order to ensure in service server
The maximization of computing resource utilization rate, and the efficiency of processing calculating task are maximized, and service server can utilize 0-1 knapsacks
Model, iterates to calculate out the goal task amount for distributing to each fringe node and sets corresponding for each described fringe node
Target computational resource, wherein, the target computational resource is used to handle corresponding goal task.For example, the service server
Can be by local remaining 20% computing resource, the calculating task for handling fringe node 1 corresponding 40%；Can be by this
The computing resource on ground remaining 80%, the calculating task for handling fringe node 2 corresponding 60%.In the present embodiment,
The end condition of 0-1 knapsacks model iterative calculation can be that each described target computational resource sum is not more than specified money in guarantee
In the case of the threshold value of source so that the duration sum needed for handling the goal task that each described fringe node is uploaded is minimum.Wherein,
The specified resource threshold can be the preassigned resource threshold of service server.For example, the specified resource threshold can be with
It is 80% in the computing resource of service server current residual or the computing resource of service server current residual
Computing resource.When distributing target computational resource and goal task amount for each fringe node, service server can simulate profit
The duration needed for corresponding goal task amount is handled with target computational resource.So, duration sum is minimum, may indicate that business takes
The efficiency for device processing calculating task of being engaged in is higher.
In the present embodiment, service server will can distribute for the goal task amount that fringe node is distributed as task
Information, feeds back to the fringe node.So, fringe node just can be known in currently pending calculating task, and which is
It is responsible for the task of processing by the service server.
S17：Goal task equal with the goal task amount in the calculating task is uploaded to the business service
Device, and the task in calculating task described in processing locality in addition to the goal task.
In the present embodiment, after fringe node receives the task allocation information of service server feedback, Ke Yicong
It is middle to extract the goal task amount.So, fringe node just can will be equal with the goal task amount in the calculating task
Goal task be uploaded to the service server, to cause service server to handle the calculating task of this part.In addition, edge
Node can start the task in addition to the goal task in calculating task described in processing locality.
In the present embodiment, can be by when service server has handled the goal task that the fringe node is uploaded
The corresponding result of the goal task feeds back to fringe node.In addition, the result of goal task can also be by edge section
Point active request is obtained.Specifically, when can include in the task allocation information that service server feeds back to fringe node
Prolong information, the Delay is used for the duration for characterizing the required by task that the service server handles the goal task amount.
Certainly, in actual applications, the data interaction between service server and fringe node is also to need the time, therefore, described
Delay can interact the task that required duration handles the goal task amount with the service server with characterize data
Required duration sum.So, fringe node can count described after goal task to be uploaded to the service server
The duration that the service server processing goal task is consumed.Reach what the Delay was characterized in the duration of statistics
In the case of duration, fringe node can send task result to the service server and obtain request, to be taken from the business
The result of the goal task is obtained in business device.In addition, in the present embodiment, the fringe node can also be using wheel
The mode of inquiry, according to the specified time cycle, sends task result to the service server and obtains request.So, the business
Server is when receiving the task result acquisition request that fringe node is sent, it can be determined that whether the goal task is complete
Into processing.If the goal task does not complete processing, can ignore that the task result that the fringe node sends obtains please
Ask.And if the goal task completion is handled, the service server just can be by the processing knot of the goal task
Fruit feeds back to fringe node.So, fringe node just can obtain the processing of the goal task from the service server
As a result.
In the present embodiment, when fringe node gets result from service server, and locally also complete
Into after processing procedure, just it can appoint what is fed back in the result locally completed and the service server for the target
The result of business is integrated, and the result of integration is supplied into the subscription client.
Embodiment two
Referring to Fig. 3, the application also provides a kind of fringe node, the fringe node includes：
Access request receiving unit 100, for receiving the access request that subscription client is sent, the access request is pointed to
Pending calculating task；
Mission bit stream reporting unit 200, for judging whether in the complete calculating task of processing locality；If it is not, will
The mission bit stream of the calculating task is sent to service server；Appointing at least including the calculating task in the mission bit stream
Business total amount；
Task allocation information receiving unit 300, the task allocation information for receiving the service server feedback is described
Task allocation information includes goal task amount, and the goal task amount is used to characterize in the calculating task to be taken by the business
Business device is responsible for the task amount of the task of processing；
Goal task uploading unit 400, for target equal with the goal task amount in the calculating task to be appointed
Business is uploaded to the service server, and the task in calculating task described in processing locality in addition to the goal task.
In the present embodiment, Delay is also included in the task allocation information, the Delay is used to characterize
The service server handles the duration of the required by task of the goal task amount；Correspondingly, the fringe node also includes：
Duration statistic unit, the duration that the goal task is consumed is handled for counting the service server；
Result acquiring unit, the situation for reaching the duration that the Delay is characterized in the duration of statistics
Under, send task result to the service server and obtain request, to obtain the goal task from the service server
Result.Certainly, poll units can also be included in the fringe node, the poll units can be according to the specified time
In the cycle, send task result to the service server and obtain request, appointed with obtaining the target from the service server
The result of business.
In the present embodiment, the fringe node also includes：
As a result unit is provided, for that will be fed back in the result and the service server locally completed for described
The result of goal task is integrated, and the result of integration is supplied into the subscription client.
Fringe node described above is only schematical, wherein the unit illustrated as separating component can be with
It is or may not be physically separate, the part shown as unit can be or may not be physical location,
A place can be located at, or can also be distributed on multiple NEs.It can select according to the actual needs wherein
Some or all of module realize the purpose of this embodiment scheme.Those of ordinary skill in the art are not paying creativeness
In the case of work, you can to understand and implement.
It should be noted that the concrete function that fringe node disclosed in above-mentioned embodiment is realized, can be with the application
The embodiment of the processing method of calculating task contrasts explanation in embodiment one, just repeats no more here.
Embodiment three
The application also provides a kind of processing method of calculating task, and methods described can apply in service server, please
Refering to Fig. 4, methods described includes：
S21：The mission bit stream that at least one fringe node is sent is received, the mission bit stream includes the fringe node
In pending calculating task task total amount.
S23：It is respectively that each described fringe node distributes corresponding goal task amount based on the mission bit stream；
S25：Receive the goal task that each described fringe node is uploaded；Wherein, the mesh that the fringe node is uploaded
The task amount of mark task goal task amount corresponding with the fringe node is equal；
S27：After the goal task that the fringe node is uploaded has been handled, corresponding result is fed back into the side
Edge node.
In the present embodiment, service server can be adopted after the mission bit stream that each fringe node is sent is received
It is defined as the goal task amount of each fringe node distribution with the algorithm of 0-1 knapsack models.Specifically, service server can be with
Corresponding edge node request association can be included by receiving simultaneously in the mission bit stream that multiple fringe nodes are reported, each mission bit stream
With the task total amount of the calculating task of processing.Now, in order to ensure in service server calculate resource utilization maximization, with
And the efficiency of processing calculating task is maximized, service server can utilize 0-1 knapsack models, iterate to calculate out and distribute to each
The goal task amount of fringe node and corresponding target computational resource is set for each described fringe node, wherein, the mesh
Mark computing resource is used to handle corresponding goal task.For example, the service server can be by local remaining 20% meter
Calculate resource, the calculating task for handling fringe node 1 corresponding 40%；Can by local remaining 80% computing resource,
Calculating task for handling fringe node 2 corresponding 60%.
In the present embodiment, the end condition of 0-1 knapsacks model iterative calculation can ensure each described target
In the case that computing resource sum is not more than specified resource threshold so that handle the goal task that each described fringe node is uploaded
Required duration sum is minimum.Wherein, the specified resource threshold can be the preassigned resource threshold of service server.Example
Such as, the specified resource threshold can be service server current residual computing resource or service server it is current
80% computing resource in remaining computing resource.Target computational resource and goal task amount are being distributed for each fringe node
When, service server can simulate the duration handled using target computational resource needed for corresponding goal task amount.So, duration
Sum is minimum, may indicate that the efficiency of service server processing calculating task is higher.
In the present embodiment, service server will can distribute for the goal task amount that fringe node is distributed as task
Information, feeds back to the fringe node.So, fringe node just can be known in currently pending calculating task, and which is
It is responsible for the task of processing by the service server.
In the present embodiment, can be by when service server has handled the goal task that the fringe node is uploaded
The corresponding result of the goal task feeds back to fringe node.In addition, the result of goal task can also be by edge section
Point active request is obtained.Specifically, service server can also feed back corresponding time delay letter respectively to fringe node each described
Breath, the Delay is used for the duration for characterizing the required by task of the processing goal task amount.Certainly, in actual applications,
Data interaction between service server and fringe node is also to need the time, therefore, and the Delay can also be characterized
Duration needed for data interaction handles the duration sum of the required by task of the goal task amount with the service server.This
Sample, service server can reach what the Delay was characterized in the duration for handling the goal task that the fringe node is uploaded
In the case of duration, the task result acquisition request that the fringe node is sent is received, and obtain in response to the task result
Request, to fringe node feedback in the result locally completed.In addition, the fringe node can also be using poll
Mode, according to the specified time cycle, sends task result to the service server and obtains request.So, the business service
Device is when receiving the task result acquisition request that fringe node is sent, it can be determined that whether the goal task at completion
Reason.If the goal task does not complete processing, the task result acquisition request that the fringe node is sent can be ignored.And
If the goal task completion is handled, the service server just can feed back the result of the goal task
To fringe node.
Example IV
Referring to Fig. 5, the application also provides a kind of service server, the service server includes：
Mission bit stream receiving unit 110, for receiving the mission bit stream that at least one fringe node is sent, the task letter
Breath includes the task total amount of calculating task pending in the fringe node；
Goal task amount allocation unit 210, for based on the mission bit stream, being respectively each described fringe node distribution
Corresponding goal task amount；
Goal task receiving unit 310, for receiving the goal task that each described fringe node is uploaded；Wherein, it is described
The task amount goal task amount corresponding with the fringe node for the goal task that fringe node is uploaded is equal；
Result feedback unit 410, will be corresponding for after the goal task that the fringe node is uploaded has been handled
Result feeds back to the fringe node.
In the present embodiment, the service server also includes：
Target computational resource setup unit, for setting corresponding target computational resource for fringe node each described, its
In, the target computational resource is used to handle the goal task that corresponding fringe node is uploaded.
In the present embodiment, the target computational resource setup unit includes：
Duration minimum ensures module, is not more than specified resource threshold for each described target computational resource sum in guarantee
In the case of so that the duration sum needed for handling the goal task that each described fringe node is uploaded is minimum.
In the present embodiment, the service server also includes：
Delay feedback unit, for feeding back corresponding Delay respectively to fringe node each described, when described
Prolong information be used for characterize handle the goal task amount required by task duration；
Correspondingly, the result feedback unit includes：
Respond module, for reaching the Delay table in the duration for handling the goal task that the fringe node is uploaded
In the case of the duration levied, the task result acquisition request that the fringe node is sent is received, and in response to the task result
Request is obtained, to fringe node feedback in the result locally completed.
Service server described above is only schematical, wherein the unit illustrated as separating component can
To be or may not be physically separate, the part shown as unit can be or may not be physics list
Member, you can with positioned at a place, or can also be distributed on multiple NEs.It can be selected according to the actual needs
In some or all of module realize the purpose of this embodiment scheme.Those of ordinary skill in the art are not paying creativeness
Work in the case of, you can to understand and implement.
It should be noted that the concrete function that service server disclosed in above-mentioned embodiment is realized, can be with this Shen
The embodiment of the processing method of calculating task explanation please be contrasted in embodiment three, just repeat no more here.
Embodiment five
Referring to Fig. 2, the application also provides a kind of processing system of calculating task, the processing system includes user client
End, fringe node and service server, wherein：
The subscription client, for sending access request to the fringe node, the access request is pointed to pending
Calculating task；
The fringe node, for judging whether in the complete calculating task of processing locality；If it is not, being calculated described
The mission bit stream of task is sent to the service server；At least the task including the calculating task is total in the mission bit stream
Amount；The task allocation information of the service server feedback is received, the task allocation information includes goal task amount, described
Goal task amount is used to characterize the task amount for being responsible for processing in the calculating task by the service server of the task；Will be described
The goal task equal with the goal task amount is uploaded to the service server in calculating task, and described in processing locality
Task in calculating task in addition to the goal task；
The service server, is respectively described in each for the mission bit stream sent based on fringe node each described
Fringe node distributes corresponding goal task amount；Receive the goal task that each described fringe node is uploaded；It is described having handled
After the goal task that fringe node is uploaded, corresponding result is fed back into the fringe node.
Therefore, in the technical scheme that the application is provided, when fringe node receives the access that subscription client is sent
During request, the whole calculating tasks that can be pointed in the processing locality access request are can decide whether first.If local
Whole calculating tasks can not be handled, then the mission bit stream of the calculating task can be sent to service server, to cause industry
A part of task in server process of the being engaged in calculating task.So, fringe node with service server by cooperateing with processing should
Calculating task, can mitigate the local computing resource for needing to put into, so, on the one hand not interfere with and carried to other subscription clients
For network data, on the other hand, the calculating task is simultaneously and concurrently handled in local and service server, it is possible to increase calculate
The treatment effeciency of task.
The embodiments of the present invention are for illustration only, and the quality of embodiment is not represented.
Each embodiment in this specification is described by the way of progressive, identical similar between each embodiment
Part mutually referring to what each embodiment was stressed is the difference with other embodiment.Especially, pin
For the embodiment of fringe node, service server and system, Jie of the embodiment of preceding method is referred to
The control that continues is explained.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
Realized by the mode of software plus required general hardware platform, naturally it is also possible to pass through hardware.Understood based on such, on
The part that technical scheme substantially in other words contributes to prior art is stated to embody in the form of software product, should
Computer software product can be stored in a computer-readable storage medium, such as ROM/RAM, magnetic disc, CD, including some fingers
Order is to cause a computer equipment (can be personal computer, server, or network equipment etc.) to perform each implementation
Method described in some parts of example or embodiment.
The foregoing is only presently preferred embodiments of the present invention, be not intended to limit the invention, it is all the present invention spirit and
Within principle, any modification, equivalent substitution and improvements made etc. should be included in the scope of the protection.
Claims (19)
1. a kind of processing method of calculating task, it is characterised in that methods described includes：
The access request that subscription client is sent is received, the access request points to pending calculating task；
Judge whether in the complete calculating task of processing locality；If it is not, by the mission bit stream of the calculating task send to
Service server；At least include the task total amount of the calculating task in the mission bit stream；
The task allocation information of the service server feedback is received, the task allocation information includes goal task amount, institute
Stating goal task amount is used to characterize the task amount for being responsible for processing in the calculating task by the service server of the task；
Goal task equal with the goal task amount in the calculating task is uploaded to the service server, and at this
Task in the ground processing calculating task in addition to the goal task.
2. according to the method described in claim 1, it is characterised in that judge whether in the complete calculating task of processing locality
Including：
Current time locally available computing resource is obtained, the computing resource includes hard disk bus resource, memory source, center
At least one of processor resource；
Judge that can the current time locally available computing resource handle the task total amount of the calculating task；If can, sentence
Surely can be in the complete calculating task of processing locality；If it is not, judging the calculating task that can not be in processing locality completely.
3. according to the method described in claim 1, it is characterised in that judge whether in the complete calculating task of processing locality
Including：
Current time locally available computing resource is obtained, the computing resource includes hard disk bus resource, memory source, center
At least one of processor resource；
The computing resource is compared with specified resource threshold, when the computing resource is more than the specified resource threshold
When, judging can be in the complete calculating task of processing locality；When the computing resource is less than or equal to the specified resource
During threshold value, judging can not be in the complete calculating task of processing locality.
4. according to the method described in claim 1, it is characterised in that also include Delay, institute in the task allocation information
State the duration that Delay is used to characterize the required by task that the service server handles the goal task amount；Correspondingly, exist
Goal task equal with the goal task amount in the calculating task is uploaded to after the service server, the side
Method also includes：
Task result is sent to the service server and obtains request, to obtain the goal task from the service server
Result.
5. according to the method described in claim 1, it is characterised in that by the calculating task with the goal task amount phase
Deng goal task be uploaded to after the service server, methods described also includes：
According to the specified time cycle, send task result to the service server and obtain and ask, with from the service server
The middle result for obtaining the goal task.
6. according to the method described in claim 1, it is characterised in that methods described also includes：
The result for the goal task fed back in the result locally completed and the service server is entered
Row is integrated, and the result of integration is supplied into the subscription client.
7. a kind of fringe node, it is characterised in that the fringe node includes：
Access request receiving unit, for receiving the access request that subscription client is sent, the access request is pointed to pending
Calculating task；
Mission bit stream reporting unit, for judging whether in the complete calculating task of processing locality；If it is not, being calculated described
The mission bit stream of task is sent to service server；At least include the task total amount of the calculating task in the mission bit stream；
Task allocation information receiving unit, the task allocation information for receiving the service server feedback, the task point
Include goal task amount with information, the goal task amount is used to characterize in the calculating task to be born by the service server
Blame the task amount of the task of processing；
Goal task uploading unit, for goal task equal with the goal task amount in the calculating task to be uploaded to
The service server, and the task in calculating task described in processing locality in addition to the goal task.
8. fringe node according to claim 7, it is characterised in that also include time delay in the task allocation information and believe
Breath, the Delay is used for the duration for characterizing the required by task that the service server handles the goal task amount；Accordingly
Ground, the fringe node also includes：
Result acquiring unit, request is obtained for sending task result to the service server, to be taken from the business
The result of the goal task is obtained in business device.
9. fringe node according to claim 7, it is characterised in that the fringe node also includes：
Poll units, obtain for according to the specified time cycle, task result to be sent to the service server and ask, with from institute
State the result that the goal task is obtained in service server.
10. fringe node according to claim 7, it is characterised in that the fringe node also includes：
As a result unit is provided, for being directed to the target by what is fed back in the result locally completed and the service server
The result of task is integrated, and the result of integration is supplied into the subscription client.
11. a kind of processing method of calculating task, it is characterised in that methods described includes：
The mission bit stream that at least one fringe node is sent is received, the mission bit stream includes pending in the fringe node
Calculating task task total amount；
It is respectively that each described fringe node distributes corresponding goal task amount based on the mission bit stream；
Receive the goal task that each described fringe node is uploaded；Wherein, the goal task that the fringe node is uploaded
Task amount goal task amount corresponding with the fringe node is equal；
After the goal task that the fringe node is uploaded has been handled, corresponding result is fed back into the fringe node.
12. method according to claim 11, it is characterised in that be respectively that each described fringe node distribution is corresponding
After goal task amount, methods described also includes：
Corresponding target computational resource is set for fringe node each described, wherein, the target computational resource is used for processing pair
The goal task that the fringe node answered is uploaded.
13. method according to claim 12, it is characterised in that set corresponding target meter for each described fringe node
Calculating resource includes：
In the case where ensureing that each described target computational resource sum is not more than specified resource threshold so that handle described in each
Duration sum needed for the goal task that fringe node is uploaded is minimum.
14. method according to claim 11, it is characterised in that be respectively that each described fringe node distribution is corresponding
After goal task amount, methods described also includes：
Corresponding Delay is fed back respectively to fringe node each described, and the Delay is used to characterize the processing target
The duration of the required by task of task amount；
Correspondingly, corresponding result is fed back into the fringe node includes：
The task result acquisition request that the fringe node is sent is received, and request is obtained in response to the task result, to institute
Fringe node feedback is stated in the result locally completed.
15. a kind of service server, it is characterised in that the service server includes：
Mission bit stream receiving unit, is wrapped for receiving in the mission bit stream that at least one fringe node is sent, the mission bit stream
Include the task total amount of calculating task pending in the fringe node；
Goal task amount allocation unit, it is corresponding for based on the mission bit stream, being respectively each described fringe node distribution
Goal task amount；
Goal task receiving unit, for receiving the goal task that each described fringe node is uploaded；Wherein, the fringe node
The task amount goal task amount corresponding with the fringe node of the goal task uploaded is equal；
Result feedback unit, for after the goal task that the fringe node is uploaded has been handled, corresponding processing to be tied
Fruit feeds back to the fringe node.
16. service server according to claim 15, it is characterised in that the service server also includes：
Target computational resource setup unit, for setting corresponding target computational resource for fringe node each described, wherein, institute
Stating target computational resource is used to handle the goal task that corresponding fringe node is uploaded.
17. service server according to claim 16, it is characterised in that the target computational resource setup unit bag
Include：
Duration minimum ensures module, for ensureing that each described target computational resource sum is not more than the feelings of specified resource threshold
Under condition so that the duration sum needed for handling the goal task that each described fringe node is uploaded is minimum.
18. service server according to claim 15, it is characterised in that the service server also includes：
Delay feedback unit, for feeding back corresponding Delay, the time delay letter respectively to fringe node each described
Cease the duration of the required by task for characterizing the processing goal task amount；
Correspondingly, the result feedback unit includes：
Respond module, request is obtained for receiving the task result that the fringe node is sent, and in response to the task result
Request is obtained, to fringe node feedback in the result locally completed.
19. a kind of processing system of calculating task, it is characterised in that the processing system includes subscription client, fringe node
And service server, wherein：
The subscription client, for sending access request to the fringe node, the access request points to pending meter
Calculation task；
The fringe node, for judging whether in the complete calculating task of processing locality；If it is not, by the calculating task
Mission bit stream send to the service server；At least include the task total amount of the calculating task in the mission bit stream；
The task allocation information of the service server feedback is received, the task allocation information includes goal task amount, the mesh
Mark task amount is used to characterize the task amount for being responsible for processing in the calculating task by the service server of the task；By the meter
The goal task equal with the goal task amount is uploaded to the service server in calculation task, and is being counted described in processing locality
Task in calculation task in addition to the goal task；
The service server, is respectively each described edge for the mission bit stream sent based on fringe node each described
The corresponding goal task amount of node distribution；Receive the goal task that each described fringe node is uploaded；Handling the edge
After the goal task that node is uploaded, corresponding result is fed back into the fringe node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710703806.0A CN107295110B (en) | 2017-08-16 | 2017-08-16 | Processing method of computing task, edge node, service server and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710703806.0A CN107295110B (en) | 2017-08-16 | 2017-08-16 | Processing method of computing task, edge node, service server and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107295110A true CN107295110A (en) | 2017-10-24 |
| CN107295110B CN107295110B (en) | 2020-08-04 |
Family
ID=60106609
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710703806.0A Active CN107295110B (en) | 2017-08-16 | 2017-08-16 | Processing method of computing task, edge node, service server and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107295110B (en) |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108243246A (en) * | 2017-12-25 | 2018-07-03 | 北京市天元网络技术股份有限公司 | A kind of edge calculations resource regulating method, edge device and system |
| CN108616503A (en) * | 2018-03-12 | 2018-10-02 | 北京邮电大学 | Data transmission method based on UDP and system |
| CN108647085A (en) * | 2018-05-16 | 2018-10-12 | 中国联合网络通信集团有限公司 | Handle method, terminal and the system of data |
| CN108942346A (en) * | 2018-08-31 | 2018-12-07 | 湖北龙腾电子科技有限公司 | A kind of cutting machine with tool changing function |
| CN108965884A (en) * | 2018-07-13 | 2018-12-07 | 网宿科技股份有限公司 | A kind of distribution method and controlling equipment, transcoding device of transcoding task |
| CN109120679A (en) * | 2018-07-27 | 2019-01-01 | 平安科技（深圳）有限公司 | Method for allocating tasks and device |
| CN109347967A (en) * | 2018-11-02 | 2019-02-15 | 网宿科技股份有限公司 | A kind of method and device obtaining audio, video data |
| CN109598250A (en) * | 2018-12-10 | 2019-04-09 | 北京旷视科技有限公司 | Feature extracting method, device, electronic equipment and computer-readable medium |
| CN109639840A (en) * | 2019-02-25 | 2019-04-16 | 网宿科技股份有限公司 | A kind of data processing method and edge calculations system based on edge calculations |
| CN109889599A (en) * | 2019-03-07 | 2019-06-14 | 北京邮电大学 | Data processing method and system |
| CN110049350A (en) * | 2019-04-15 | 2019-07-23 | 深圳壹账通智能科技有限公司 | Video code conversion processing method, device, computer equipment and storage medium |
| CN110856045A (en) * | 2019-09-30 | 2020-02-28 | 咪咕视讯科技有限公司 | Video processing method, electronic device, and storage medium |
| CN110995827A (en) * | 2019-11-29 | 2020-04-10 | 腾讯科技（深圳）有限公司 | Communication processing method and device, computer readable medium and electronic equipment |
| CN111432159A (en) * | 2020-03-19 | 2020-07-17 | 深圳市鹏创软件有限公司 | Computing task processing method, device and system and computer readable storage medium |
| CN111565101A (en) * | 2020-04-28 | 2020-08-21 | 上海物融智能科技有限公司 | Processing method and device for computing task |
| CN112311820A (en) * | 2019-07-26 | 2021-02-02 | 腾讯科技（深圳）有限公司 | Edge device scheduling method, connection method, device and edge device |
| CN113660726A (en) * | 2020-05-12 | 2021-11-16 | 华为技术有限公司 | Resource allocation method and device |
| CN113891114A (en) * | 2021-11-18 | 2022-01-04 | 上海哔哩哔哩科技有限公司 | Transcoding task scheduling method and device |
| CN113992715A (en) * | 2021-10-26 | 2022-01-28 | 广东卓维网络有限公司 | Online state cloud monitoring and edge computing device and method |
| CN114679604A (en) * | 2022-04-11 | 2022-06-28 | 上海哔哩哔哩科技有限公司 | Resource processing method and device |
| CN115118766A (en) * | 2022-06-22 | 2022-09-27 | 中国银行股份有限公司 | Mobile office method based on edge calculation and related device |
| CN115348211A (en) * | 2022-07-04 | 2022-11-15 | 深圳市高德信通信股份有限公司 | Method for processing a computational task using network nodes available on a network process |
| CN116250219A (en) * | 2020-08-04 | 2023-06-09 | 北京大学深圳研究生院 | Communication method for network computing power or resource service |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101741912A (en) * | 2009-12-30 | 2010-06-16 | 中兴通讯股份有限公司 | Method, network apparatus and distributed network system for processing computation task |
| US20100153960A1 (en) * | 2008-12-15 | 2010-06-17 | Korea Advanced Institute Of Science And Technology | Method and apparatus for resource management in grid computing systems |
| CN102135949A (en) * | 2011-03-01 | 2011-07-27 | 浪潮(北京)电子信息产业有限公司 | Computing network system, method and device based on graphic processing unit |
| CN103955404A (en) * | 2014-03-28 | 2014-07-30 | 哈尔滨工业大学 | Load judgment method based on NoC multi-core homogeneous system and task immigration method based on method |
| CN105139130A (en) * | 2015-08-27 | 2015-12-09 | 国电南瑞科技股份有限公司 | Power system distributed task management method |
| CN105740085A (en) * | 2014-12-11 | 2016-07-06 | 华为技术有限公司 | Fault tolerance processing method and device |
-
2017
- 2017-08-16 CN CN201710703806.0A patent/CN107295110B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100153960A1 (en) * | 2008-12-15 | 2010-06-17 | Korea Advanced Institute Of Science And Technology | Method and apparatus for resource management in grid computing systems |
| CN101741912A (en) * | 2009-12-30 | 2010-06-16 | 中兴通讯股份有限公司 | Method, network apparatus and distributed network system for processing computation task |
| CN102135949A (en) * | 2011-03-01 | 2011-07-27 | 浪潮(北京)电子信息产业有限公司 | Computing network system, method and device based on graphic processing unit |
| CN103955404A (en) * | 2014-03-28 | 2014-07-30 | 哈尔滨工业大学 | Load judgment method based on NoC multi-core homogeneous system and task immigration method based on method |
| CN105740085A (en) * | 2014-12-11 | 2016-07-06 | 华为技术有限公司 | Fault tolerance processing method and device |
| CN105139130A (en) * | 2015-08-27 | 2015-12-09 | 国电南瑞科技股份有限公司 | Power system distributed task management method |
Cited By (34)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108243246A (en) * | 2017-12-25 | 2018-07-03 | 北京市天元网络技术股份有限公司 | A kind of edge calculations resource regulating method, edge device and system |
| CN108616503A (en) * | 2018-03-12 | 2018-10-02 | 北京邮电大学 | Data transmission method based on UDP and system |
| CN108647085A (en) * | 2018-05-16 | 2018-10-12 | 中国联合网络通信集团有限公司 | Handle method, terminal and the system of data |
| CN108965884A (en) * | 2018-07-13 | 2018-12-07 | 网宿科技股份有限公司 | A kind of distribution method and controlling equipment, transcoding device of transcoding task |
| CN108965884B (en) * | 2018-07-13 | 2020-12-25 | 网宿科技股份有限公司 | Distribution method of transcoding tasks, scheduling device and transcoding device |
| WO2020010709A1 (en) * | 2018-07-13 | 2020-01-16 | 网宿科技股份有限公司 | A transcoding task allocation method, scheduling device and transcoding device |
| CN109120679A (en) * | 2018-07-27 | 2019-01-01 | 平安科技（深圳）有限公司 | Method for allocating tasks and device |
| CN108942346A (en) * | 2018-08-31 | 2018-12-07 | 湖北龙腾电子科技有限公司 | A kind of cutting machine with tool changing function |
| CN109347967A (en) * | 2018-11-02 | 2019-02-15 | 网宿科技股份有限公司 | A kind of method and device obtaining audio, video data |
| CN109347967B (en) * | 2018-11-02 | 2021-08-03 | 网宿科技股份有限公司 | Method and device for acquiring audio and video data |
| CN109598250A (en) * | 2018-12-10 | 2019-04-09 | 北京旷视科技有限公司 | Feature extracting method, device, electronic equipment and computer-readable medium |
| CN109639840A (en) * | 2019-02-25 | 2019-04-16 | 网宿科技股份有限公司 | A kind of data processing method and edge calculations system based on edge calculations |
| CN109889599B (en) * | 2019-03-07 | 2020-08-28 | 北京邮电大学 | Data processing method and system |
| CN109889599A (en) * | 2019-03-07 | 2019-06-14 | 北京邮电大学 | Data processing method and system |
| CN110049350B (en) * | 2019-04-15 | 2022-10-11 | 深圳壹账通智能科技有限公司 | Video transcoding processing method and device, computer equipment and storage medium |
| CN110049350A (en) * | 2019-04-15 | 2019-07-23 | 深圳壹账通智能科技有限公司 | Video code conversion processing method, device, computer equipment and storage medium |
| CN112311820A (en) * | 2019-07-26 | 2021-02-02 | 腾讯科技（深圳）有限公司 | Edge device scheduling method, connection method, device and edge device |
| CN110856045A (en) * | 2019-09-30 | 2020-02-28 | 咪咕视讯科技有限公司 | Video processing method, electronic device, and storage medium |
| CN110856045B (en) * | 2019-09-30 | 2021-12-07 | 咪咕视讯科技有限公司 | Video processing method, electronic device, and storage medium |
| CN110995827A (en) * | 2019-11-29 | 2020-04-10 | 腾讯科技（深圳）有限公司 | Communication processing method and device, computer readable medium and electronic equipment |
| CN111432159A (en) * | 2020-03-19 | 2020-07-17 | 深圳市鹏创软件有限公司 | Computing task processing method, device and system and computer readable storage medium |
| CN111432159B (en) * | 2020-03-19 | 2022-05-17 | 深圳市鹏创软件有限公司 | Computing task processing method, device and system and computer readable storage medium |
| CN111565101A (en) * | 2020-04-28 | 2020-08-21 | 上海物融智能科技有限公司 | Processing method and device for computing task |
| CN113660726A (en) * | 2020-05-12 | 2021-11-16 | 华为技术有限公司 | Resource allocation method and device |
| CN116250219A (en) * | 2020-08-04 | 2023-06-09 | 北京大学深圳研究生院 | Communication method for network computing power or resource service |
| CN113992715A (en) * | 2021-10-26 | 2022-01-28 | 广东卓维网络有限公司 | Online state cloud monitoring and edge computing device and method |
| CN113992715B (en) * | 2021-10-26 | 2023-09-26 | 广东卓维网络有限公司 | Online state cloud monitoring and edge computing device and method |
| CN113891114A (en) * | 2021-11-18 | 2022-01-04 | 上海哔哩哔哩科技有限公司 | Transcoding task scheduling method and device |
| CN113891114B (en) * | 2021-11-18 | 2023-12-15 | 上海哔哩哔哩科技有限公司 | Transcoding task scheduling method and device |
| CN114679604A (en) * | 2022-04-11 | 2022-06-28 | 上海哔哩哔哩科技有限公司 | Resource processing method and device |
| CN114679604B (en) * | 2022-04-11 | 2023-12-19 | 上海哔哩哔哩科技有限公司 | Resource processing method and device |
| CN115118766A (en) * | 2022-06-22 | 2022-09-27 | 中国银行股份有限公司 | Mobile office method based on edge calculation and related device |
| CN115348211A (en) * | 2022-07-04 | 2022-11-15 | 深圳市高德信通信股份有限公司 | Method for processing a computational task using network nodes available on a network process |
| CN115348211B (en) * | 2022-07-04 | 2024-03-19 | 深圳市高德信通信股份有限公司 | Method for processing computing tasks using available network nodes |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107295110B (en) | 2020-08-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107295110A (en) | Processing method, fringe node, service server and the system of calculating task | |
| WO2020133967A1 (en) | Method for scheduling shared computing resources, shared computing system, server, and storage medium | |
| CN110557284B (en) | Data aggregation method and device based on client gateway | |
| CN103338252B (en) | Realizing method of distributed database concurrence storage virtual request mechanism | |
| US8938495B2 (en) | Remote management system with adaptive session management mechanism | |
| CN109729106A (en) | Handle the method, system and computer program product of calculating task | |
| IL240542B (en) | Distributed cloud services system and uses thereof | |
| US11741291B2 (en) | Systems and methods for providing error recovery in data transmissions | |
| CN109254854A (en) | Asynchronous invoking method, computer installation and storage medium | |
| CN110430274A (en) | A kind of document down loading method and system based on cloud storage | |
| CN109831524A (en) | A kind of load balance process method and device | |
| CN105872587A (en) | Video request processing method and device | |
| KR101033813B1 (en) | Cloud computing network system and file distrubuting method of the same | |
| CN113377507A (en) | Task processing method, device, equipment and computer readable storage medium | |
| CN104052679A (en) | Load balancing method and device for network flow | |
| CN105656794B (en) | Data distributing method, device and computer readable storage medium | |
| JP4789401B2 (en) | Content distribution system | |
| CN107343037B (en) | Data transmission method and device of distributed storage system | |
| CN116048825A (en) | Container cluster construction method and system | |
| WO2017185632A1 (en) | Data transmission method and electronic device | |
| WO2023213262A1 (en) | Mesh sidecar expansion method and device | |
| CN103502941B (en) | A kind of method for parallel processing and device | |
| KR20110109610A (en) | Cloud computing network system and file distribution method of the same | |
| CN109547291A (en) | A kind of method and device of quick positioning high frequency bandwidth consumption | |
| CN105578212B (en) | A kind of point-to-point Streaming Media method of real-time in big data under stream calculation platform |
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 |