CN110377352A - Task processing method and device based on mobile device cloud system - Google Patents
Task processing method and device based on mobile device cloud system Download PDFInfo
- Publication number
- CN110377352A CN110377352A CN201910417909.XA CN201910417909A CN110377352A CN 110377352 A CN110377352 A CN 110377352A CN 201910417909 A CN201910417909 A CN 201910417909A CN 110377352 A CN110377352 A CN 110377352A
- Authority
- CN
- China
- Prior art keywords
- task
- node
- probability
- path
- mobile device
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/445—Program loading or initiating
- G06F9/44594—Unloading
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
- G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
- G06F9/5038—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
A kind of task processing method and device based on mobile device cloud system, method include: the initiation node for obtaining task and the task to be unloaded；Carry out executing the original allocation of node in the mobile device cloud system；Unassigned task during original allocation is redistributed；It obtains and executes the smallest path of Probability Of Mission Success, and the task on the path is adjusted, realize that the Mission Success for carrying out maximum probability in the mobile device cloud system that opportunistic connects executes, ensure that the reliability of task execution.
Description
Technical field
The application design objective processing technology field, in particular to a kind of task processing based on mobile device cloud system
Method and apparatus.
Background technique
Although the performance with mobile device is continuously improved, for computation-intensive task, limited meter
Calculation ability, memory space and battery life are still difficult to meet the process demand of task.Moreover, in practical applications, needing
The complicated task of the calculating that mobile device is completed has been usually more than the processing capacity of individual equipment.Therefore, in mobile device capability
Limited and communications status shows in the case that chance is connection, how to handle this generic task as current critical issue.
Summary of the invention
The object of the present invention is to provide a kind of task processing methods based on mobile device cloud system, to realize in chance
The Mission Success that maximum probability is carried out in the mobile device cloud system of formula connection executes, and ensure that the reliability of task execution.
To solve the above problems, the first aspect of the present invention provides at a kind of task based on mobile device cloud system
Reason method obtains the initiation node of task and the task to be unloaded；Execution section is carried out in the mobile device cloud system
The original allocation of point；Unassigned task during original allocation is redistributed；Obtain execute the task at
The smallest path of function probability, and the task on the path is adjusted.
Further, the original allocation for carrying out executing node in the mobile device cloud system, comprising: obtain
Execute the shortest path of the task, wherein the shortest path does not include directapath；It is selected in the shortest path
For executing the execution node of the task；It, will be described when identifying that the energy for executing node meets the execution task
Task is distributed to the execution node.
Further, described that unassigned task during original allocation is redistributed, comprising: by institute
It states and is assigned to the path of task during original allocation and is ranked up, wherein sort by is usable probability；Described in selection
The maximum path of usable probability, and selection executes node in the maximum path of the usable probability.
Further, the acquisition executes the smallest path of Probability Of Mission Success, and to the task on the path
It is adjusted, further includes: obtain the probability of success of each path in mobile device cloud；Identify the maximum road of the probability of success
Diameter and the smallest path of the probability of success；Any task in the smallest path of the probability of success is adjusted to institute by identification
State the situation of change of probability of success when in the maximum path of the probability of success；If described increase at merits and demerits probability, this is controlled
Business is adjusted to the maximum path of the probability of success.
Further, when being allocated to the task, in system described in real-time update the energy of each node with can
By property.
Further, after the execution node completes the task, implementing result is fed back into the initiation node.
According to another aspect of the present invention, a kind of Task Processing Unit based on mobile device cloud system, comprising: obtain
Modulus block, for obtaining the initiation node of task and the task to be unloaded；Original allocation module, in the movement
Carry out executing the original allocation of node in equipment cloud system；Module is redistributed, for not divided during original allocation
Matching for task is redistributed；Module is adjusted, executes the smallest path of Probability Of Mission Success for obtaining, and to institute
Stating on path for task is adjusted.
According to another aspect of the invention, a kind of computer readable storage medium is stored thereon with program, described program
It is performed and realizes the task processing method based on mobile device cloud system.
Above-mentioned technical proposal of the invention has following beneficial technical effect: allowing arbitrary size data between node
It is transmitted several times, and introduces the model of dynamic evaluation mission reliability.Based on the mentioned frame of the application, the task of realizing is unloaded
Decision is carried, can guarantee the reliability of task execution while maximizing the probability that Mission Success executes.
Detailed description of the invention
Fig. 1 is the structural schematic diagram of the mobile device cloud system of one specific embodiment of this application；
Fig. 2 is the structural schematic diagram of the mobile device cloud system of another specific embodiment of this application；
Fig. 3 is the process schematic of task adjustment in one specific embodiment of this application；
Fig. 4 is the flow chart of the task processing method of the mobile device cloud system of one embodiment of the invention；
Fig. 5 is the flow chart of the task processing method of the mobile device cloud system of another embodiment of the present invention；
Fig. 6 is the flow chart of the task processing method of the mobile device cloud system of another embodiment of the invention；
Fig. 7 is the flow chart of the task processing method of the mobile device cloud system of further embodiment of the present invention；
Fig. 8 (a) is the contrast verification knot that the successful execution probability of first specific embodiment of the invention changes with the off period
Fruit figure；
Fig. 8 (b) is the contrast verification that the successful execution probability of second specific embodiment of the invention changes with size of data
Result figure；
Fig. 8 (c) is the contrast verification knot that the task copy of third specific embodiment of the present invention changes with reliability requirement
Fruit figure；
Fig. 8 (d) is that the remaining capacity of the 4th specific embodiment of the invention changes contrast verification result with reliability requirement
Figure；
Fig. 9 is the block diagram of the Task Processing Unit of the mobile device cloud system of one embodiment of the invention.
Specific embodiment
In order to make the objectives, technical solutions and advantages of the present invention clearer, With reference to embodiment and join
According to attached drawing, the present invention is described in more detail.It should be understood that these descriptions are merely illustrative, and it is not intended to limit this hair
Bright range.In addition, in the following description, descriptions of well-known structures and technologies are omitted, to avoid this is unnecessarily obscured
The concept of invention.
Schematic diagram of a layer structure according to an embodiment of the present invention is shown in the attached drawings.These figures are not drawn to scale
, wherein for purposes of clarity, some details are magnified, and some details may be omitted.It is shown in the drawings various
Region, the shape of layer and relative size, positional relationship between them are merely exemplary, in practice may be due to manufacture
Tolerance or technical restriction and be deviated, and those skilled in the art may be additionally designed as required have difference
Shape, size, the regions/layers of relative position.
Obviously, described embodiments are some of the embodiments of the present invention, instead of all the embodiments.Based on this hair
Embodiment in bright, every other reality obtained by those of ordinary skill in the art without making creative efforts
Example is applied, shall fall within the protection scope of the present invention.
In the description of the present invention, it should be noted that term " first ", " second ", " third " are only used for description mesh
, it is not understood to indicate or imply relative importance.
In addition, as long as technical characteristic involved in invention described below different embodiments is each other not
Constituting conflict can be combined with each other.
Hereinafter reference will be made to the drawings, and the present invention will be described in more detail.In various figures, identical element is using similar attached
Icon is remembered to indicate.For the sake of clarity, the various pieces in attached drawing are not necessarily to scale.
Many specific details of the invention, such as structure, material, size, the processing work of device is described hereinafter
Skill and technology, to be more clearly understood that the present invention.But it just as the skilled person will understand, can not
The present invention is realized according to these specific details.Unless hereinafter particularly pointing out, the various pieces in semiconductor devices can
To be made of material well known to those skilled in the art.
It should be noted that mobile device carries more important work in specific area.For example, in battlefield
On, the mobile device for being equipped with sound transducer may need to identify from background shot to position sniper or to gun type
Classify, to provide Informational support for subsequent take action；The mobile device for being preinstalled with Target Identification Software may need
It is identified after taking pictures to the target in battlefield, to exclude potential danger as early as possible；In view of the requirement of battlefield particular surroundings
(artillery fire sound, which shrouds lower voice messaging, possibly can not obtain, and may need to keep silent when standing facing each other with place), soldier can pass through
Voice command from commanding is converted to text by the speech recognition software of mobile device assembly, to realize acquisition of information.
But these calculate the processing capacity that complicated task has been usually more than single mobile device, therefore, the application
Propose a kind of task processing method based on mobile device cloud.
The definition designed in the application is introduced below by table 1.
Firstly, mobile device cloud is different from assuming in the related technology between mobile device and mobile device is counted with long-range
According to, there are stabilizing network connection, mobile device cloud thinks that the network connection between equipment is opportunistic, intermittence between center
's.
Therefore, there are a central node (such as satellite or base station) and at least one shiftings in the application building system
Dynamic node (mobile device), central node are connected at least one mobile node, and the state letter of each node of periodic harvest
Breath, so that the migration for task provides decision-making foundation.Fig. 1 is the mobile device cloud system of one specific embodiment of this application
Structural schematic diagram.As shown in Figure 1, A, B, C and D are mobile node, in opportunistic connection and central node point between them
It is not connected with them, central node constantly collects its contact information and real-time status with other nodes from each node
Information (such as task dispatching of electricity, distribution).For the task method of salary distribution shown in Fig. 1, A node initiate task 1 with
Task 2, and task 1 is sent to node C by node B, so that node C passes through node B for task after having executed task 1 again
As a result node A is fed back to.In such a mode, the mode that the node in mobile device cloud shows cooperation carries out task computation,
This mode is known as mobile device cloud system cooperated computing by the application.
Optimal mobile node is selected to execute task and the most suitable path of selection in mobile cloud system in order to realize
Data are transmitted, then face following problem.Firstly, connecting between mobile node in opportunistic, the decision at current time is to be based on
Current connection state, thus current optimal decision may become failure decision subsequent；Simultaneously as when being contacted between node
Between transience and dynamic, data relevant with task cannot be guaranteed once contact in transmit completely, then need to consider more
Secondary contact transmission.Secondly as usually there is each task off period time of completion and implementing result to require return task
Node is initiated, which in turns increases the complex natures of the problem.Furthermore, it is contemplated that the mobility and typical scene of node (such as
Disaster, battlefield) environment abominable, the case where unavoidable node fails.For example, mobile node in battlefield because by
It attacks and damages to enemy.Therefore, it also needs to consider fault tolerant mechanism when being designed mobile device cloud system.Finally, due to
The limitation of battery capacity in the finiteness and mobile device of calculated performance, distribute task when also need to consider power saving, with
Mobile device is set to be able to maintain long-play.
In order to realize that the purpose of the application, the application have carried out following analysis first.
Fig. 2 is the application scenario diagram of the application one embodiment.As shown in Fig. 2, in certain circumstances, some movable joints
Point random movement in respective region, can be by the point-to-point communication interface (WiFi or bluetooth etc.) that is configured between them
It is communicated.It is in opportunistic communication state between node due to the limitation of the mobility and range for wireless communication of mobile node.It examines
Consider between node can mutual shared resource, thus can be considered an opportunistic mobile device cloud system.When appointing for some node
When calculating demand of being engaged in is more than the ability of this node, which can be offloaded to other nodes for local calculating task.When two
In a node motion to communication range, it can be considered primary contact, can carry out data transmission between node.It should be understood that machine
Meeting formula (opportunistic) communication is only just to be considered as primary contact in two node motions to communication range.Therefore, the application
By the opportunistic Contact modeling between mobile device at a contact netWherein,For node set, ε is node
Between contact line set.It is set as eij∈ ε node i is to the side of node j, mainly by contact interval duration (inter-
Contact duration distribution) distribution p conijIt determines (for example, the node contact probability of node a and node b
For 0.1).Wherein, the contact between node is mutual, therefore scheming G is non-directed graph.Moreover, pconij=0 mean node i with
The possibility directly contacted is not present in node j, thus works as pconij> 0 is that there are side e between node i and node jij.However, this
The node being not directly contacted with a bit can be considered as the relaying of task execution.In addition, identical two nodes continue in different contacts
Time is also different.The application sets node to<i, j>between once the minimum data volume that can be transmitted of contact be δij.Such as Fig. 2 institute
Show, task, which initiates node a, directly to be contacted with node b, d, e, and node c is not contacted directly with node a, but can play the part of number
According to biography loser and potential performer.Due to task execution it is complete after data must also be back to initiation node,
All execution routes are all such as path<a, b from node is initiated to initiating node, a>.In view of the cooperation between node needs
Want the contact network figure etc. between the global information in mobile device cloud system, such as node state, allocated task, node.
The application assumes that there are a central nodes (satellite or base station in mobile device cloud system etc.) and carries out the period to these data
Property mobile phone.Since the usual rate of the link of bandwidth between mobile node and central node is lower, usually satellite link therefore will
Being transmitted to purpose mobile node after multiplexed transport to central node again will cause intolerable transmission delay, thus must adopt
With point-to-point transmission.Further, since the application is by taking specific application scene such as war and disaster as an example, then there is movable joint
Point can be moved according to specific rule and plan, and therefore, the touching act in the application between mobile node can be pre-
It surveys.
Such analysis it is found that task initiate node task may due to or off period larger content limitation etc. without
Method is executed by a mobile node, and therefore, the task that task initiates the initiation of node needs to be broken down into multiple subtasks, example
Such as, voice is turned in the application of text, a big voice document can be divided into multiple small documents, then by multiple movements
Node simultaneously converts multiple small documents, to shorten voice conversion time.The application sets subtask setEach
SubtaskIt is described as such as executing load W comprising following informationn(CPU number of processes), deadline Tn, input
Data volume DinnWith output data quantity DoutnDeng.In view of data volume size is typically different before data processing and after processing, for
For specific application, it is assumed that DinnWith DoutnBetween there are a parameter factors δ, that is, Dinn=δ Doutn.Be placed in how
These details of acquisition task are the prior art, are repeated no more in this application.How the application mainly solves by task
It initiates multiple subtasks that node is initiated to distribute in each node into mobile device cloud, to accelerate the completion speed of task.
Generally, it is only necessary to the result of task execution is back to task and initiates node, and in task execution
Between result need not return.However, due to being in that opportunistic connects, and to consider task completion time most between mobile node
Smallization, the correlative factors such as fault-tolerant and efficiency optimally distribute these tasks non-into the mobile node of each dynamic change
Often with there is challenge.
Further, each subtask is set as incoherent in the application, thus in the implementation procedure of subtask, it is in
Reveal starlike communication network architecture.For subtaskWe are scheduled according to the priority orders of task.It is right
For single subtask, offloading network model of the subtask between node is appointed as shown in Fig. 2, when meeting with other nodes
Business initiation node can be loaded to be offloaded to the node to meet and result is back to task after the completion of task execution and be initiated
Node.
Due to the mobility of node and the abominable of environment, especially under disaster or field environment, node can not be kept away
Exempt from the case where will appear failure, this will will lead to task execution actual effect.Therefore, consider during to sub- task immigration fault-tolerant
Mechanism is particularly important to the successful execution for guaranteeing task.In this regard, failure and mission reliability two side of the application from mobile node
It is analyzed in face.
For task failure analysis.The failure of usual node is attacked from hardware error or by external environment.It is known
The incidence of node mistake is obeyed Poisson distribution and is widely used in system Reliability Research.Therefore, for individual node
For i, the distribution of T moment interior nodes error probability can be formalized are as follows:
Wherein, uiIt is the parameter of Poisson function, s indicates the errors number occurred simultaneously in the T moment.
Therefore, as s=0, the probability that node i does not malfunction within the T moment is then indicated are as follows:
For mission reliability model.The application sets the reliability of incoming task as R, it is contemplated that includes N number of son in task
Task, and the importance of each subtask is impartial, thus can consider the reliability of each subtask also just as can calculate
ForFor the reliability for improving task execution, each task needs, which are assigned to the execution of multiple nodes, (to be resolved into more
Multiple subtasks are distributed to multiple nodes execute again by a subtask).Equally, it is contemplated that the finiteness of node resource, a section
Point may also can undertake multiple subtasks to guarantee the backup number of task.If on (i) is subtask the allocated in node i
Set, at current time, if distributing subtask n to node i, the execution time (TT) of total score be may be expressed as:
Wherein, ET (n, i) is execution time of the task n in node i, nkFor k-th of element on (i) set.It is known
The load W of task nnWith the processing speed c of node ii, then ET (n, i)=Wn/ci.Reference formula (2), can derive egress
Current reliability are as follows:
Wherein, R (n, i) is the probability that task n does not malfunction during node i executes.Therefore, it is wanted when R (n, i) is greater than
When the reliability r asked, then stops at and task n is allocated.
Further, after task initiates node initiation task, multiple sons can be divided by certain task description tool
Task is denoted asThe problem of being addressed below is how to gather subtaskIn each subtask be offloaded to movement set
Accelerate the execution of task for other nodes in cloud system.It should be noted that in the uninstall process of subtask, it is also necessary to full
Foot is claimed below: 1) off period for meeting each subtask requires；2) guarantee the reliability requirement of task execution；3) make as far as possible
The energy consumption of task execution is minimum.Finally, cooperate with the task unloading problem in mobile device cloud system that can be defined as maximizing institute
There is the probability of subtask successful execution and meets institute's Prescribed Properties.In turn, which can turn in digital form:
Wherein,To initiate node to the set of paths result receiving node from task.PS (n, p, i) is logical for task n
It crosses path p and carries out the probability for transmitting and being run succeeded within the off period by node i.xn,p,iFor a variable, whether subtask is indicated
The node i being assigned into path p.R (n) indicates the achieved reliability of subtasking n, and formula (6) ensure that subtask
The practical execution reliability of n is greater than required reliability.Formula (7) ensure that the charge level of node i not less than set
Threshold value.Formula (8) indicates xn,p,iIt is a two-valued variable.Formula (9) indicates that any task is at most distributed to same node point one
Secondary and each task is at least assigned primary.As it can be seen that the problem optimizes simultaneously under conditions of reliability and energy constraint
The distribution path of task, execute node and task by score, be a combined optimization problem.Due to execution route set with
Number of nodes be exponentially increased, therefore, ask the time complexity of Optimum Solution very high, related optimization algorithm
Be not suitable for running in mobile device cloud system.
Further, since node is usually carried by people or controlled, the Move Mode of mobile node can show certain
Rule.For example, unmanned plane is usually cruised when obtaining situation of battlefield with the route that commanding sets.Thus, it is false
If the touching act between node be it is predictable, since there are different data transmission and Mission Capability in different paths simultaneously
And mainly determined by the probability that its opportunistic connects, the application is by successful execution event of the task on particular path with one kind
Probabilistic manner analysis.Most current usually mainly considers two node lists when carrying out path probability analysis in the related technology
Secondary contact situation, and assume no matter time of contact how long data can transmission success.But single connects between the application assumes node
The transmittability of touching is limited, for biggish data, it may be necessary to which multiple-contact could end of transmission.
Wherein, chance route availability probability is the measurement that connection status is in path in specific time, it is task
The key factor of successful execution on path.The application measures the path using the contact interval duration of node each on path
Availability.Contact interval duration is primary before referring to contact the time terminated between contact start next time.Intuitively, right
For a node, contact interval is smaller, and contact duration is bigger, thus the usable probability in path is bigger.Based on this,
The probability of availability in chance path can be determined by the contact interval duration factor.The availability in path is analyzed, just be needed
Further investigate the contact process between node.
Specifically, since the contact process between node obeys Poisson distribution, then the contact interval duration obedience between node refers to
Number distribution.Consider double bounce path (Data Migration twice occurs) < a, v, a >, indicate that task is offloaded to node v by node a,
Then result is back to a again.Therefore, the corresponding usable probability in the path can calculate in the following way:
If stochastic variable T1And T2Respectively first jumps and the contact duration distribution of the second hop node, probability density function point
It Dui Yingyu not f (t, λ1)=Exp (t, λ1) and f (t, λ2)=Exp (t, λ2).Then, total path available duration can be expressed as T
=T1+T2, corresponding probability density function can also pass through f (t, λ1) and f (t, λ2) convolution obtain, i.e. f (t, λ1)·f(t,
λ2).The corresponding probability of availability in the path can be expressed as:
In fact, above formula calculates the path usable probability in primary contact.However, in fact, one spy of transmission
Fixed data may need to be transmitted several times, thus formula (10) does not excavate path available capability also completely.Assuming that a data exist
First jump in need to contact for m time could be by data end of transmission, and T1q~Exp (λq) it is set as q=1 in the first jump ..., a
The contact duration of secondary transmission, then it is a length of when total contact of the first jumpTheoretical proof T1Obey gamma point
Cloth, i.e. T1~Gamma (a, λ1).Similarly, T2~Gamma (a, λ2)., can be by path < a based on this, the availability of v, a > are general
Rate statement are as follows:
More generally, for the path p for having K to jump, corresponding total available duration is thenIt is corresponding
Probability density function are as follows:
Wherein, λk, (k=1 .., K) and mk, (k=1 .., K) respectively represent kth jump the probability function parameter factor and
Transmit frequency of exposure required for data.Wherein, when k is very big, f (t) is a high latitude amount, and computation complexity is very
Its approximation is reduced to single gamma stochastic variable using Satterthwaite (Satterthwaite) approximation theory by height, it may be assumed that
Wherein,
So far, the usable probability in our paths can pass through formulaIt calculates.
Due within a certain period of time, even if path is available, also can not necessarily guarantee that achievement transmits relevant data.
The application provides the probability analysis for measuring data transmission success below.Notice time of contact individually contact in contact continue
Time and be assumed to be obey Pareto distribution, since the transmission rate between node is relatively stable, be easy to get single contact
The data volume of interior transmission is also approximate to obey Pareto distribution.For node to < v, w >, if Di~Pareto (α, β) is to indicate
The stochastic variable of the data volume of i-th transmission, wherein α is form factor, and β is that scale factor (can be considered single transmission most
Small data quantity).Assuming that node within the T moment to there is c contact, then total transmitted data amount of time T interior nodes pair can indicate
ForIt is easy to get, the data that size is D probability of Successful transmissions within the T moment is P (D >=D).In view of becoming
Measure D and DiMaximum value M (i.e. M=max { the D of (i=1 .., c)i, i=1 ..., c }) order it is identical, can be with M come approximate table
Show variables D.Thus, P (D >=D) can be further simplified are as follows:
Due to successfully realizing the unloading of task in given time T, it is necessary to meet following condition: 1) path must the time
It is available in T；2) all related datas include that result data must transfer in time T along destination path.Cause
And a task n is given, outputting and inputting size of data accordingly is respectivelyWithIn its path < a,
v1,...,vK-1, the probability that a > is successfully unloaded can calculate in the following manner.
Since any node in path may all undertake task execution, for any task n, on the p of path
There is the selection of task execution in k.Equally, kth is jumped, it is contemplated that it is β that it transmits minimum transmission quantity every timek, at leastIn it is possible selection to transmit related data.Dk(n, i) represents task n when kth hop node is executed i-th
A node needs the data volume transmitted.Therefore, on the p of path the probability of successful execution task n be all possible situations probability
Summation, it may be assumed that
Wherein, fi(m1,..,mK) representing task n, respectively jump is respectively necessary for m when i-th of node executes1,..,mKSecondary contact
To transmit probability when data completely.If noticing situation m1,..,mKIt is only passed when wherein one jumps in preceding primary contact when generation
Defeated failure, i.e. m1-1,..,mKCertain in -1 jumps data transmission fails.IfIt is data in each the number of transmissions
For m1,..mk-1,mKWhen transmission failure and situation be m1,..mk,mKWhen successful probability, it may be assumed that
Wherein,Expression task n is jumped in the time that i-th of node executes with kth
It transmits the sum of time required for related data, 1I=kIt indicates that its value is 1 as i=k, is otherwise 0；Expression task n's
It executes kth item when node is i and contacts mkThe quantity that can be transmitted after secondary.Obviously, in a path, the section before node is executed
The data that point needs to transmit are the related data that task carries, i.e. data enter amountNode after execution node needs to pass
Defeated data are task executions as a result, i.e. volume of transmitted data isFormally it may be expressed as:
Wherein,It is m before kth is jumpedkThe probability of secondary transmission failure.It can analyze and be concluded that, transmission is unsuccessfully led
There are two aspects, one is the connection in path is unavailable, the second is by the non-end of transmission of data in the phase.If mk
Secondary transmission success, it is meant that mk- 1 time transmission path is centainly available.Therefore, m is causedkThe reason of -1 transmission failure
Data are essentially consisted in not transfer completely.Thus,It can be calculate by the following formula:
So, it is assumed that h task distribute into path p and set the task allocation result of path p as
Wherein,It is combined to distribute passing through to road for task,It is each task on road
Corresponding execution node in diameter p.To guarantee task execution according to priority, the application is with stack architecture arrangement distribution to road
The task of diameter p, the deletion and addition of task are by the sequence that last in, first out, in order to the adjustment of subsequent task.Fig. 3, which gives, to be appointed
Business 3 is adjusted from path u interior joint d to the example of path v interior joint c.Thus for path p, to calculate its successful execution probabilityNeed to calculate the volume of transmitted data and task execution time of every jump.If < d1,d2,...,dK>
With < t1,t2,...,tK> is respectively the transmitted data amount and task execution time of each jump.Wherein,
Indicate the total data transmission quantity that kth is jumped,It is in the pathsA node executes taskWhen kth jump need
The data volume of transmission.DkThe definition of (n, i) can refer to formula (19).Similarly, kth can be calculated and jump total execution timeWherein, Tk(n, i) is that task n kth when executing in node i jumps the time it takes,
The transmission time of execution time and data including task, can be calculated as Tk(n, i)=1I=k·wn/ck+dk/rk.Thus, it can
By the D in formula (18)k(n, i) and Tn′,i,lIt is substituted for dkAnd tk, that is, it can calculate under multi-task planning, the success of path p is held
Row probability
The application is in view of the reliability of node execution task, the energy consumption of node and other constraints as a result, specific
When design, need to follow following rule: if 1) electricity of node i is less than preset threshold (such as) when, then no longer to node i point
With task；If 2)Task n should be distributed to the node nearest from initiation node as far as possible to reduce network communication
Pressure, vice versa；3) it is more than primary that each task, which cannot be distributed to same node,.It should be understood that if node failure,
Being previously assigned to the task of the node can all fail, thus repeatedly will be unable to reach in same node standby and improve reliability
Purpose.
Based on above-mentioned analysis, the application proposes a kind of task processing method and device based on mobile device cloud system.
It should be noted that mentioning in this application for task, it can be understood as subtask n above-mentioned, that is, incited somebody to action
Initiating in task node for task is divided into multiple subtasks, and research in this application is the assignment problem to subtask.
It should also be noted that, the data volume that task carries is intended to more greatly the time more grown unloading, thus task
Lead to higher error rate in implementation procedure.As it can be seen that it is necessary to assign different priority to different task, to increase task
The whole success rate of execution.Based on this idea, the application sets the priority of task, data volume according to the size of task data
Bigger, priority is higher, and vice versa.Give a subtask, need to determine its should any paths transmitting data with
And which node on the path is selected to carry out task processing.Such as Such analysis, without loss of generality, if task distribution
It as a result is a tripleIndicate task-setIt distributes to path p and carries out data transmission, it is corresponding
Task execution node isIdeal scheme is to find a paths in all possible paths to make task execution probability most
Greatly, the probability including path usable probability and data transmission success.However, since feasible set is as the increase of number of nodes is in finger
Number increases, and feasible to be focused to find out best practice efficiency necessarily low so many.It notices in the application Such analysis, once
Path determines that task execution node can then be determined according to above-mentioned rule.
Fig. 4 is the flow chart of the task processing method based on mobile device cloud system of the embodiment of the present invention.Such as Fig. 4 institute
Show, the task processing method based on mobile device cloud system of the embodiment of the present invention, comprising the following steps:
S101: the initiation node of at least one task to be unloaded Yu task to be unloaded is obtained.
S102: for each task to be unloaded, the original allocation for executing node is carried out in mobile device cloud system.
According to one embodiment of present invention, step S102 is as shown in Figure 5, further includes:
S201: the maximum shortest path of execution task usable probability is obtained.
Wherein, shortest path does not include directapath, that is, is directed toward the path for initiating node itself.
S202: the execution node for executing task is selected according to the data volume of task in shortest path.
Further, judge whether the input data amount of task is greater than the output data quantity of task；If it is, selection
The node close with node is initiated is as execution node.
S203: when the energy that identification executes node meets execution task, task is distributed to execution node.
That is, the application first finds the path of maximum usable probability, optimal task is then selected in the paths again
Node is executed, and considers Energy Consumption Factors.Above step is repeated to all according to task priority descending after each path is assigned
Task proceeds as described above processing, until all tasks all complete just sub-distribution or without available node, this stage then stops.
For example, terminal device cloud system as shown in Figure 2, node a have a subtask n to need to be offloaded to other
In node, and the off period of the subtask is T.In initial dispensing phase, if all nodes are available execution node,
Remember enabled node setIf all set of paths of enabled node a to a areThe path for having distributed task is
Usable probability of the path k within the T moment beThe directapath for triggering node to task result receiving node from task is available
Probability is Pra′v=1.Then, being searched out using Dijsktra (Di Jiesitela) algorithm is made outside directapath
Shortest path k, subtask n are distributed to path k and are added to distribution path setIn.Path k Yi Dan distribution after,
It is then deleted from former probability graph G along the corresponding side in path.IfSubtask n should be distributed as far as possible to from initiation
The nearest node of node is to reduce network communication pressure；Otherwise, subtask n should be distributed as far as possible to farthest from node is initiated
Node.The execution node of task is selected from the enabled node of path k according to the above rule, and the node i selected needs completely
FootWherein, eiIndicate energy required for node i subtasking n.This indicates only have energy that can expire
The node of sufficient task execution could be selected.Then, to the state such as energy of each node and the current reliability of each task
It is updated.If energy is less than certain threshold value, by this node from usable setIt deletes.Above step is repeated, until
Task is all assigned in all paths, i.e.,Or all tasks are assigned and meet reliability conditions.
S103: unassigned task during original allocation is redistributed.
According to one embodiment of present invention, step S103 is as shown in Figure 6, further includes:
S301: the path that task is assigned to during original allocation is ranked up.
Wherein, sort by is currently available probability, that is, the path is also attainable available general after initial allocation
Rate.
S302: the selection maximum path of usable probability, and selection executes node in the maximum path of usable probability.
That is, after initial allocation, it is possible that task be not fully allocated also or the number of copies of task also not
The case where reaching reliability requirement, the purpose in this stage are exactly more further to ensure being fully allocated and guaranteeing reliability for task
It is required that.Most basic idea is constantly to be distributed unfinished task to the path of successful execution maximum probability by iteration
In.
Specifically, for unallocated task-setIn task, first by allocated set of pathsIn path
It is ranked up according to current usable probability, and remembers that the maximum path of corresponding usable probability is pm.Then, by subtask n points
It is assigned to path pmAnd according to aforementionedWithAnd node energy determines the execution node of task.
S104: the probability of success in each path in mobile device cloud system is obtained, and task is adjusted according to the probability of success
It is whole.
According to one embodiment of present invention, step S104 is as shown in Figure 7, further includes:
S401: the probability of success of each path in mobile device cloud is obtained.
It should be understood that mobile device cloud system experienced the distribution of two subtasks at this time, current state is lower to be needed to be unloaded
The subtask of load makes according to the performance parameter of each node (task data amount, energy etc.) all in the allocated to each node
Each path has the different probability of succesies.
S402: the successfully path of maximum probability and the smallest path of the probability of success are identified.
S403: any task in the smallest path of the probability of success is adjusted in the maximum path of the probability of success by identification
When the probability of success situation of change.
S404: it if increased at merits and demerits probability, controls the task and is adjusted to the maximum path of the probability of success.
That is, after the reallocation, since each path has increased the distribution of task newly, will affect original task
Successful execution probability.This means that the scheme after redistributing is not optimal.This stage by adjusting task distribution to mention
The probability of its high successful execution.
It should be understood that in the embodiment of the present application, usable probability is the probability value whether path is connected to, success is general
Rate is the probability value that task is successfully executed on path, wherein task, which is successfully executed, not only needs path that can be connected to,
The functions such as calculating, the transmission of task are also needed to be completed.In other words, usable probability is the premise of the probability of success.
Specifically, probability of success maximum and the smallest path u and v in mobile device cloud system are found, and probability is minimum
The task of distribution path is adjusted into maximum probability path, until successful execution probability no longer increases.
For example, it is set as function maximum probability and the smallest path is corresponding is respectivelyWithTask h is extracted from the task queue of path v, if
Then task h is added in the current path u for executing maximum probability.If the task h in the v of path is adjusted path u
Afterwards without the task of execution, that is,Then by path v from the middle deletion of set of paths P '.Repeat above step, Zhi Daoren
The adjustment of business distribution no longer improves the whole probability that runs succeeded.
In order to prove the application propose the task processing method based on mobile device cloud system effect, the application also
By (random) strategy of the present processes and Random, MaxPro (Maximum Contact probability) strategy and MaxRate (maximum transmitted
Rate) strategy compares.Wherein, Random strategy, it is randomly chosen the node that task initiation node is touched and regards
The destination node of task unloading.MaxPrO strategy, it selects the node for initiating node contact maximum probability with task to regard task
Unload object.MaxRate strategy, it selects to initiate the maximum node of node-node transmission rate with task as the unloading mesh of task
's.Obviously, this strategy generates minimal propagation delay.
As shown in Fig. 8 (a), comparative situation that PSE index changes with task deadline.With the increase of off period, remove
Random strategy is outer, and Mission Success executes probability P SE and also accordingly increases.This is because the off period is longer, node more have an opportunity with
Task initiates node contact, thus is easier within a specified time to return to implementing result.Since Random strategy randomly chooses institute
The node encountered also shows random situation of change as task execution node, corresponding curve.It notices and works as the off period
When Deadline is greater than 1500, Mission Success executes probability and is not further added by, this is because task that may be all can be 1500
It is completed in time slot, the growth of off period does not influence the successful execution of task.From this picture, the application be also found, be mentioned herein
For algorithm UNION compared to other algorithms with the variation of off period, PSE index remains highest level, thus shows institute
Propose the superiority that algorithm compares other algorithms.
As shown in Fig. 8 (b), comparative situation that PSE index changes with size of data.From figure as it can be seen that the success of task is held
Row probability constantly reduces with the growth of data volume.The application is it is thought that because data volume is smaller, within the given time
Easier transmission success, vice versa.It notices the PSE kept stable after datasize=87MB and fluctuates very little,
The application thinks the reason is that, the time of node the last time contact is long, transmits these data enough, thus to PSE shadow
Sound is smaller.But it can be predicted that further increasing with data volume, PSE can be further decreased.The application is equally seen
Observe when being compared with other strategies, UNION algorithm PSE index as the variation of data volume is always all in highest level,
This further demonstrates the superiority of this paper algorithm UNION.
As shown in Fig. 8 (c), average task number of copies with reliability requirement situation of change.According to figure as it can be seen that with reliable
Property desired increase, the number of copies of task is increasing.This is because the raising of reliability needs more copies to guarantee to appoint
The successful execution of business.Number of copies caused by this paper algorithm UNION is also observed compared to other algorithms from this result the application
It is more stable, about it is stably held in the quantity of 2 copies, it means that UNION is not reducing reliability compared to other algorithms
In the case of the advantage also more saved with resource.
As shown in Fig. 8 (d), each remaining energy of strategy compared with the situation of change of reliability requirement, it is seen that UNION
Maintain highest and most stable of dump energy.This illustrates that it is each to balance can also to optimize the distribution of task for mentioned algorithm herein
Energy consumption between node.In systems in practice, this is allowed to play for a long time as much as possible and answer to the runing time of delay mobile system
Have most important.When less as other tactful dump energies because under these strategies may a node can distribute it is multiple
Task.
In conclusion the task processing method of the mobile device cloud system of present application example, allows arbitrary size data to exist
It is transmitted several times between node, and introduces the model of dynamic evaluation mission reliability.Based on the mentioned frame of the application, realize
Task unloads decision, can guarantee the reliability of task execution while maximizing the probability that Mission Success executes.
Fig. 9 is the block diagram of the Task Processing Unit based on mobile device cloud system of the embodiment of the present invention.Such as figure
Shown in 9, the Task Processing Unit 100 based on mobile device cloud system of the embodiment of the present invention, comprising: obtain module 10, initial
Distribution module 20 redistributes module 30 and adjustment module 40.
Wherein, the initiation node that module 10 is used to obtain task and the task to be unloaded is obtained；Original allocation module
20 in the mobile device cloud system for carrying out executing the original allocation of node；Module 30 is redistributed for first
Unassigned task is redistributed in beginning assigning process；It is general for obtaining the execution Mission Success to adjust module 40
The smallest path of rate, and the task on the path is adjusted.
In order to achieve the above object, the application also proposed a kind of computer readable storage medium, be stored thereon with journey
Sequence, program, which is performed, realizes the task processing method above-mentioned based on mobile device cloud system.
It should be understood that above-mentioned specific embodiment of the invention is used only for exemplary illustration or explains the present invention
Principle, but not to limit the present invention.Therefore, it is done without departing from the spirit and scope of the present invention
Any modification, equivalent substitution, improvement and etc. should all be included in the protection scope of the present invention.In addition, right appended by the present invention
It is required that being intended to cover the whole fallen into attached claim scope and boundary or this range and the equivalent form on boundary
Change and modification.
In the above description, the technical details such as composition, the etching of each layer are not described in detail.But
It is it will be appreciated by those skilled in the art that can be by various means in the prior art, to form floor, the area of required shape
Domain etc..In addition, in order to form same structure, those skilled in the art be can be devised by and process as described above and endless
Exactly the same method.
The present invention is described above by reference to the embodiment of the present invention.But these embodiments are used for the purpose of saying
Bright purpose, and be not intended to limit the scope of the invention.The scope of the present invention is limited by appended claims and its equivalent
It is fixed.The scope of the present invention is not departed from, those skilled in the art can make a variety of substitutions and modifications, these substitutions and modifications are all
It should fall within the scope of the present invention.
Although embodiments of the present invention are described in detail, it should be understood that, without departing from of the invention
In the case where spirit and scope, embodiments of the present invention can be made with various changes, replacement and change.
Obviously, the above embodiments are merely examples for clarifying the description, and does not limit the embodiments.
For those of ordinary skill in the art, other various forms of changes can also be made on the basis of the above description
Change or changes.There is no necessity and possibility to exhaust all the enbodiments.And obvious change extended from this
Change or changes still within the protection scope of the invention.
It should be understood by those skilled in the art that, the embodiment of the present invention can provide as method, system or computer journey
Sequence product.Therefore, complete hardware embodiment, complete software embodiment or combining software and hardware aspects can be used in the present invention
The form of embodiment.Moreover, it wherein includes the calculating of computer usable program code that the present invention, which can be used in one or more,
The computer program implemented in machine usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
The form of product.
The present invention be referring to according to the method for the embodiment of the present invention, the process of equipment (system) and computer program product
Figure and/or block diagram describe.It should be understood that can be realized by computer program instructions each in flowchart and/or the block diagram
The combination of process and/or box in process and/or box and flowchart and/or the block diagram.It can provide these computers
Processor of the program instruction to general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices
To generate a machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute
For realizing the function of being specified in one or more flows of the flowchart and/or one or more blocks of the block diagram
Device.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that instruction stored in the computer readable memory generation includes
The manufacture of command device, the command device are realized in one box of one or more flows of the flowchart and/or block diagram
Or the function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that
Series of operation steps are executed on computer or other programmable devices to generate computer implemented processing, thus calculating
The instruction executed on machine or other programmable devices is provided for realizing in one or more flows of the flowchart and/or side
The step of function of being specified in block diagram one box or multiple boxes.
Claims (9)
1. a kind of task processing method based on mobile device cloud system characterized by comprising
Obtain the initiation node of at least one task to be unloaded Yu the task to be unloaded；
For each task to be unloaded, the original allocation for executing node is carried out in the mobile device cloud system；
Unassigned task during the original allocation is redistributed；
The probability of success in each path in the mobile device cloud system is obtained, and the task is carried out according to the probability of success
Adjustment.
2. the task processing method according to claim 1 based on mobile device cloud system, which is characterized in that described in institute
State the original allocation for carrying out executing node in mobile device cloud system, comprising:
It obtains and executes the maximum shortest path of task usable probability, wherein the shortest path does not include directapath；
The execution node for executing the task is selected according to the data volume of the task in the shortest path；
When identifying that the energy for executing node meets the execution task, the task is distributed to the execution node.
3. the task processing method according to claim 2 based on mobile device cloud system, which is characterized in that described in institute
It states in shortest path and the execution node for executing the task is selected according to the data volume of the task, further includes:
Judge whether the input data amount of the task is greater than the output data quantity of the task；
If it is, selecting the node close with the initiation node as the execution node.
4. the task processing method according to claim 1 based on mobile device cloud system, which is characterized in that it is described to
Unassigned task is redistributed during original allocation, comprising:
The path that task is assigned to during original allocation is ranked up, wherein sort by is currently available general
Rate；
The maximum path of the usable probability is selected, and selection executes node in the maximum path of the usable probability.
5. the task processing method according to claim 1 based on mobile device cloud system, which is characterized in that the acquisition
The probability of success in each path in the mobile device cloud system, and the task is adjusted according to the probability of success, also
Include:
Obtain the probability of success of each path in mobile device cloud；
Identify the maximum path of the probability of success and the smallest path of the probability of success；
When any task in the smallest path of the probability of success is adjusted in the maximum path of the probability of success by identification
The situation of change of the probability of success；
If described increase at merits and demerits probability, controls the task and be adjusted to the maximum path of the probability of success.
6. any task processing method based on mobile device cloud system in -5 according to claim 1, which is characterized in that
Further include:
When being allocated to the task, the energy and reliability of each node in system described in real-time update.
7. any task processing method based on mobile device cloud system in -5 according to claim 1, which is characterized in that
Further include:
After the execution node completes the task, implementing result is fed back into the initiation node.
8. a kind of Task Processing Unit based on mobile device cloud system characterized by comprising
Module is obtained, for obtaining the initiation node of task and the task to be unloaded；
Original allocation module, for carrying out executing the original allocation of node in the mobile device cloud system；
Module is redistributed, for redistributing to unassigned task during original allocation；
Module is adjusted, executes the smallest path of Probability Of Mission Success for obtaining, and carry out to the task on the path
Adjustment.
9. a kind of computer readable storage medium, is stored thereon with program, which is characterized in that described program is performed realization such as
Task processing method based on mobile device cloud system described in claim 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910417909.XA CN110377352B (en) | 2019-05-20 | 2019-05-20 | Task processing method and device based on mobile device cloud system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910417909.XA CN110377352B (en) | 2019-05-20 | 2019-05-20 | Task processing method and device based on mobile device cloud system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110377352A true CN110377352A (en) | 2019-10-25 |
| CN110377352B CN110377352B (en) | 2022-06-14 |
Family
ID=68248567
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910417909.XA Active CN110377352B (en) | 2019-05-20 | 2019-05-20 | Task processing method and device based on mobile device cloud system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110377352B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112114648A (en) * | 2020-11-23 | 2020-12-22 | 中国人民解放军国防科技大学 | Wearable device power management method and system and computer device |
| CN112181665A (en) * | 2020-10-21 | 2021-01-05 | 中国联合网络通信集团有限公司 | Task optimization method and device for low-earth-orbit satellite |
| CN112866381A (en) * | 2021-01-17 | 2021-05-28 | 湘潭大学 | Task redundancy allocation method based on facility site selection problem in edge calculation |
| CN113872850A (en) * | 2021-09-27 | 2021-12-31 | 东莞市亚太未来软件有限公司 | Real-time communication method and system |
| CN114697209A (en) * | 2022-03-30 | 2022-07-01 | 广州穗华能源科技有限公司 | Cloud edge cooperative computing resource configuration method and configuration system |
Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009076671A2 (en) * | 2007-12-13 | 2009-06-18 | Advanced Micro Devices, Inc. | Driver architecture for computing device having multiple graphics subsystems, reduced power consumption modes, software and methods |
| US20100088150A1 (en) * | 2008-10-08 | 2010-04-08 | Jamal Mazhar | Cloud computing lifecycle management for n-tier applications |
| US20120252405A1 (en) * | 2011-03-31 | 2012-10-04 | Lortz Victor B | Connecting mobile devices, internet-connected hosts, and cloud services |
| CN103595756A (en) * | 2012-08-17 | 2014-02-19 | 三星电子株式会社 | Method and apparatus for generating and utilizing a cloud service-based content shortcut object |
| US20160057230A1 (en) * | 2014-08-19 | 2016-02-25 | Hand Held Products, Inc. | Mobile computing device with data cognition software |
| US20160164986A1 (en) * | 2014-12-08 | 2016-06-09 | Google Inc. | Multi-purpose application launching interface |
| US20160285975A1 (en) * | 2014-05-21 | 2016-09-29 | Societal Innovations Ipco Limited | System and method for aggregating and acting on signals from one or more remote sources in real time using a configurable platform instance |
| CN106462542A (en) * | 2014-03-10 | 2017-02-22 | 英特尔公司 | Mobile application acceleration via fine-grain offloading to cloud computing infrastructures |
| CN106597881A (en) * | 2016-11-03 | 2017-04-26 | 深圳量旌科技有限公司 | Cloud Service Robot Based on Distributed Decision Algorithm |
| CN106936892A (en) * | 2017-01-09 | 2017-07-07 | 北京邮电大学 | A kind of self-organizing cloud multi-to-multi computation migration method and system |
| CN107087019A (en) * | 2017-03-14 | 2017-08-22 | 西安电子科技大学 | A kind of end cloud cooperated computing framework and task scheduling apparatus and method |
| US20170310564A1 (en) * | 2014-09-29 | 2017-10-26 | International Business Machines Corporation | Allocating physical nodes for processes in an execution plan |
| CN107766135A (en) * | 2017-09-29 | 2018-03-06 | 东南大学 | Method for allocating tasks based on population and simulated annealing optimization in mobile cloudlet |
| EP3355188A1 (en) * | 2017-01-31 | 2018-08-01 | OpenSynergy GmbH | Instrument display on a car dashboard by checking frames of a gui by a realtime os |
| CN108958916A (en) * | 2018-06-29 | 2018-12-07 | 杭州电子科技大学 | Workflow unloads optimization algorithm under a kind of mobile peripheral surroundings |
| CN109005211A (en) * | 2018-06-29 | 2018-12-14 | 福建师范大学 | Thin cloud deployment and scheduling user task method under a kind of wireless MAN environment |
| CN109074265A (en) * | 2016-03-28 | 2018-12-21 | 甲骨文国际公司 | The preformed instruction of mobile cloud service |
| CN109756578A (en) * | 2019-02-26 | 2019-05-14 | 上海科技大学 | A kind of low time delay method for scheduling task calculating network towards dynamic mist |
-
2019
- 2019-05-20 CN CN201910417909.XA patent/CN110377352B/en active Active
Patent Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009076671A2 (en) * | 2007-12-13 | 2009-06-18 | Advanced Micro Devices, Inc. | Driver architecture for computing device having multiple graphics subsystems, reduced power consumption modes, software and methods |
| US20100088150A1 (en) * | 2008-10-08 | 2010-04-08 | Jamal Mazhar | Cloud computing lifecycle management for n-tier applications |
| US20120252405A1 (en) * | 2011-03-31 | 2012-10-04 | Lortz Victor B | Connecting mobile devices, internet-connected hosts, and cloud services |
| CN103595756A (en) * | 2012-08-17 | 2014-02-19 | 三星电子株式会社 | Method and apparatus for generating and utilizing a cloud service-based content shortcut object |
| CN106462542A (en) * | 2014-03-10 | 2017-02-22 | 英特尔公司 | Mobile application acceleration via fine-grain offloading to cloud computing infrastructures |
| US20160285975A1 (en) * | 2014-05-21 | 2016-09-29 | Societal Innovations Ipco Limited | System and method for aggregating and acting on signals from one or more remote sources in real time using a configurable platform instance |
| US20160057230A1 (en) * | 2014-08-19 | 2016-02-25 | Hand Held Products, Inc. | Mobile computing device with data cognition software |
| US20170310564A1 (en) * | 2014-09-29 | 2017-10-26 | International Business Machines Corporation | Allocating physical nodes for processes in an execution plan |
| US20160164986A1 (en) * | 2014-12-08 | 2016-06-09 | Google Inc. | Multi-purpose application launching interface |
| CN109074265A (en) * | 2016-03-28 | 2018-12-21 | 甲骨文国际公司 | The preformed instruction of mobile cloud service |
| CN106597881A (en) * | 2016-11-03 | 2017-04-26 | 深圳量旌科技有限公司 | Cloud Service Robot Based on Distributed Decision Algorithm |
| CN106936892A (en) * | 2017-01-09 | 2017-07-07 | 北京邮电大学 | A kind of self-organizing cloud multi-to-multi computation migration method and system |
| EP3355188A1 (en) * | 2017-01-31 | 2018-08-01 | OpenSynergy GmbH | Instrument display on a car dashboard by checking frames of a gui by a realtime os |
| CN107087019A (en) * | 2017-03-14 | 2017-08-22 | 西安电子科技大学 | A kind of end cloud cooperated computing framework and task scheduling apparatus and method |
| CN107766135A (en) * | 2017-09-29 | 2018-03-06 | 东南大学 | Method for allocating tasks based on population and simulated annealing optimization in mobile cloudlet |
| CN108958916A (en) * | 2018-06-29 | 2018-12-07 | 杭州电子科技大学 | Workflow unloads optimization algorithm under a kind of mobile peripheral surroundings |
| CN109005211A (en) * | 2018-06-29 | 2018-12-14 | 福建师范大学 | Thin cloud deployment and scheduling user task method under a kind of wireless MAN environment |
| CN109756578A (en) * | 2019-02-26 | 2019-05-14 | 上海科技大学 | A kind of low time delay method for scheduling task calculating network towards dynamic mist |
Non-Patent Citations (2)
| Title |
|---|
| WEIDONG BAO 等: "ACO-BASED SOLUTION FOR COMPUTATION OFFLOADING IN MOBILE CLOUD COMPUTING", 《BIG DATA AND INFORMATION ANALYTICS (BDIA)》 * |
| 杨彬: "移动云计算环境下移动系统能源消耗情况研究", 《无线互联科技》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112181665A (en) * | 2020-10-21 | 2021-01-05 | 中国联合网络通信集团有限公司 | Task optimization method and device for low-earth-orbit satellite |
| CN112114648A (en) * | 2020-11-23 | 2020-12-22 | 中国人民解放军国防科技大学 | Wearable device power management method and system and computer device |
| CN112866381A (en) * | 2021-01-17 | 2021-05-28 | 湘潭大学 | Task redundancy allocation method based on facility site selection problem in edge calculation |
| CN112866381B (en) * | 2021-01-17 | 2022-07-01 | 湘潭大学 | Task redundancy allocation method based on facility site selection problem in edge calculation |
| CN113872850A (en) * | 2021-09-27 | 2021-12-31 | 东莞市亚太未来软件有限公司 | Real-time communication method and system |
| CN114697209A (en) * | 2022-03-30 | 2022-07-01 | 广州穗华能源科技有限公司 | Cloud edge cooperative computing resource configuration method and configuration system |
| CN114697209B (en) * | 2022-03-30 | 2023-12-22 | 广州穗华能源科技有限公司 | Cloud edge collaborative computing resource configuration method and configuration system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110377352B (en) | 2022-06-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110377352A (en) | Task processing method and device based on mobile device cloud system | |
| Liu et al. | Online computation offloading and traffic routing for UAV swarms in edge-cloud computing | |
| CN104636204B (en) | A kind of method for scheduling task and device | |
| US9154382B2 (en) | Information processing system | |
| Yang et al. | Timely-throughput optimal coded computing over cloud networks | |
| Misra et al. | Multiarmed-bandit-based decentralized computation offloading in fog-enabled IoT | |
| CN108769182A (en) | A kind of prediction executes the Combinatorial Optimization dispatching method of task execution time | |
| CN112883526B (en) | Workload distribution method under task delay and reliability constraint | |
| CN113114738B (en) | SDN-based optimization method for internet of vehicles task unloading | |
| Huang et al. | Enabling low latency edge intelligence based on multi-exit dnns in the wild | |
| Mostafa | Cooperative fog communications using a multi-level load balancing | |
| El Khatib et al. | Optimal proactive resource allocation at the extreme edge | |
| CN114760202A (en) | Reliable construction and deployment method of service function chain in network slice scene | |
| Avasalcai et al. | Adaptive management of volatile edge systems at runtime with satisfiability | |
| Zhu et al. | Deep reinforcement learning for fairness in distributed robotic multi-type resource allocation | |
| Henna et al. | Distributed and collaborative high-speed inference deep learning for mobile edge with topological dependencies | |
| Shi et al. | Workflow migration in uncertain edge computing environments based on interval many-objective evolutionary algorithm | |
| Hu et al. | Joint deployment and request routing for microservice call graphs in data centers | |
| Kang et al. | Recen: resilient manet based centralized multi robot system using mobile agent system | |
| CN116996941A (en) | Calculation force unloading method, device and system based on cooperation of cloud edge ends of distribution network | |
| Xiao et al. | Cooperative reinforcement learning in topology-based multi-agent systems | |
| Pugalia et al. | A Hybrid Algorithm for Optimized Resource Allocation under Constrained Task Schedule | |
| Johnson et al. | Convergence analysis of the hybrid information and plan consensus algorithm | |
| Yao et al. | Performance Optimization in Serverless Edge Computing Environment using DRL-Based Function Offloading | |
| Saxena et al. | Latency-aware task partitioning and resource allocation in fog networks |
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 |