CN108513318A - A kind of customer service queuing optimization method based on edge calculations - Google Patents
A kind of customer service queuing optimization method based on edge calculations Download PDFInfo
- Publication number
- CN108513318A CN108513318A CN201810132806.4A CN201810132806A CN108513318A CN 108513318 A CN108513318 A CN 108513318A CN 201810132806 A CN201810132806 A CN 201810132806A CN 108513318 A CN108513318 A CN 108513318A
- Authority
- CN
- China
- Prior art keywords
- edge calculations
- business
- node
- batch
- calculations node
- 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
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/02—Traffic management, e.g. flow control or congestion control
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/90—Buffering arrangements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W52/00—Power management, e.g. TPC [Transmission Power Control], power saving or power classes
- H04W52/02—Power saving arrangements
- H04W52/0203—Power saving arrangements in the radio access network or backbone network of wireless communication networks
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The present invention relates to a kind of, and the customer service based on edge calculations is lined up optimization method, and this approach includes the following steps：The different business for using packet classifier to generate user first, classifies according to customer identification number and QoS requirement, and the business classified then is put into corresponding sub- caching management module；Next, which is established Batch Arrival model and is reached to business using Batch Arrival model, carries out probability Estimation；The service parameter of edge calculations node is obtained again according to Batch Arrival model, and fringe node is then selected according to the service parameter of edge calculations node；Finally setting time threshold is T, the opening and closing of edge calculations node is controlled according to time threshold T, with energy saving.Since the quantity of 5G environment lower edge equipment is more, the portfolio of outburst is also more, and the present invention selects edge calculations node to have very great help business as needed using the Batch Arrival model in queueing theory, can significantly improve the service quality of business.
Description
Technical field
The present invention relates to a kind of, and the multiple dimensioned frequency spectrum access scheme based on business demand, especially one kind being based on edge meter
The customer service of calculation is lined up optimization method, belongs to mobile communication technology field.
Background technology
In recent years, mobile device develops rapidly, and user volume also rises sharply therewith, with Internet of Things fast development and
The birth of internet 5G technologies, society have come into the epoch of all things on earth interconnection, and network edge device quantity increases sharply,
So that data caused by such equipment have reached damp byte rank.The epoch are handled by the big data of core of cloud computing model,
Its key technology has been unable to data caused by efficient process network edge device, and mobile network is faced with unprecedented
Challenge.In view of the above-mentioned problems, edge calculations model comes into being, edge calculations model can preferably solve the Internet of things era
The above problem in the presence of big data processing.Edge calculations refer to that one kind of task computation is executed on network edge device
" edge " of novel computation model, edge calculations refers to from data source caused by network edge device to cloud computing center number
According to the arbitrary computing resource and Internet resources between path, basic concept is by calculating task in the calculating close to data source
It is run in resource.
Queueing theory as it is a kind of based on the mathematical tool that theory of random processes grows up be often used in system modelling and
Performance evaluation.When early utilization queueing theory models and assesses communication network, it is contemplated that the arrival of customer has randomness, can use
Probability distribution indicates arrival process, and in research starting stage of queueing theory, the arrival process for mainly having studied customer is most simple
The queuing system singly flowed.With the development of network, the model for describing network flow becomes increasingly complex.In various queuing models
In, Batch Arrival queuing model with stronger applicability due to being widely studied.
Under the following 5G environment, it is contemplated that super-intensive scene, D2D scenes etc., edge device are even more to high density direction
Development.Since edge calculations node inherently has the function of to calculate, store etc., edge calculations can be utilized to generate user
Mass data is handled, and in addition edge calculations need not pass through cloud computing center close to user terminal processing data, reduce use
The time delay of family business processing.The arrival of customer service has prodigious randomness, therefore with queue theory model to the clothes of user
Certain optimization is done in business, can optimize what edge calculations node distributed customer service, improves efficiency.
Invention content
It is an object of the invention to：In view of the defects existing in the prior art, a kind of user's industry based on edge calculations is proposed
Business is lined up optimization method, on the one hand this method can effectively improve customer service order QoS, on the other hand has very greatly to energy saving
It helps, there is good application prospect.
In order to reach object above, the present invention provides a kind of, and the customer service based on edge calculations is lined up optimization method,
This approach includes the following steps：
The first step, the different business for being generated user using packet classifier, according to customer identification number and service quality
(QoS) demand is classified, and the business classified then is put into corresponding sub- caching management module；
Second step establishes Batch Arrival model and reaches progress probability Estimation to business using Batch Arrival model；
Third step, the service parameter that edge calculations node is obtained according to Batch Arrival model, then according to edge calculations section
The service parameter of point selects fringe node；
4th step, setting time threshold are T, the opening and closing of edge calculations node are controlled according to time threshold T, to save energy
Source.
The present invention can utilize Batch Arrival model in queueing theory, and probability analysis is carried out to the business reached at random, according to
Parameter carries out node selection optimization, and it is few when the task quantity for being asked in the unit interval when it is more, due to edge device number
There are many amount, and considerable energy can be expended by being constantly in service state, and the present invention has also carried out certain improvement to this problem,
On the basis of above, the present invention proposes a kind of new customer service queuing prioritization scheme based on edge calculations, Neng Gou
The customer service queuing based on edge calculations is optimized using queueing theory under 5G environment.The customer service of the present invention is lined up
Prioritization scheme first classifies to packet customer service according to different QoS of survice and customer identification number, then in queueing theory
Batch Arrival model the arrival probability of customer service is predicted, and calculate relevant parameter, carry out performance estimation,
To promote the flexibility ratio and service quality of customer service service on this basis.In short, the method for the present invention is simple, it is real to be easy to
It is existing, there is good application prospect.
Preferably, in the first step, sub- caching management module meets claimed below：
1.1 assume a user for carrying sub- caching management module, cache size hj, cache and j-th stage needed to take
Business, then the bandwidth needed for j-th stage service is cjRb；
1.2 assume that the total activation capacity of a certain edge calculations node is bandwidth C, and the sub- caching management module of user has m
A, then when m sub- caching management modules provide service by the same edge calculations node, required total bandwidth is c, according to (1) formula
C values are calculated,
1.3 set the total number of above-mentioned edge calculations node serve user as N, and N values are calculated according to (2) formula,
Illustrate that above-mentioned edge calculations node can service N number of user simultaneously,
If above-mentioned edge calculations node supports that the total number of packet is n, n values are calculated according to (3) formula,
Wherein, cjFor the packet number of j-th stage business, illustrate that above-mentioned edge calculations node can support n packet simultaneously.
It is further preferred that when the number of user or packet exceeds the bearing capacity of edge calculations node, if being continuing with
The edge calculations node provides service, then will appear packet loss phenomenon, thus when the edge calculations node at full capacity when, should broadcast
Information removes other edge calculations nodes to shunt business.
Still further preferably, the traffic handing capacity of different edge calculations nodes is different, the industry serviced as needed
Business bandwidth and each edge calculations node for receiving of each port send out whether fully loaded information, from the end of business generation place
Port address selects edge calculations node, to reduce time delay and reduce packet loss from the near to the distant.
Preferably, in the second step, establishing Batch Arrival model, the specific method is as follows：
2.1 assume that has an edge calculations node for h classification caching, while servicing N number of user, in the edge calculations
When a certain business A of node serve, wrapped if business A Batch Arrival intensity is k, and 1≤k≤n≤c+h, if that each wraps is averaged
Rate is Rb, the inflow parameter for each packet is the exponential distribution of μ, flows out the exponential distribution that parameter is λ, service regulation is
First Come First Served establishes queue theory model MAP using Markov ChainX/ M/c/c+h, wherein n are maximum batch intensity, MAP
For batch Markov；
2.2 set markovian state space as S={ 1,2, L, m }, and j states are transferred to from i states, wherein 1≤
I, j≤m≤n, if the probability that batch event reaches is Pij, when there is the event that batch intensity is k to reach, probability pij
(k), when there is no event arrival and i ≠ j, probability pij(0), it can thus be concluded that only batch event reaches, state can just be made
I returns to itself；
2.3 set the state-transition matrix of queuing system as Q, then
Wherein, D is Batch Arrival rate, with Dmin(m,n)For, it is the maximum Batch Arrival rate of arrival state m, DkFor m+
The batch rate of inflow of k-state enables And 1≤i, j≤m≤n, μiIt is each
Packet stream enters the parameter of the exponential distribution of obedience.
In addition, inflow (m+1) μ and Batch Arrival D of the inflow of state m by m+1 stateskComposition, the outflow of state m by
The outflow m μ and Batch Arrival D of m-1 stateskComposition.Due to m<N, the maximum value for reaching inflow state m are Dm。
It is further preferred that pij(0) it is not have the probability that packet stream enters, p from i states to j statesij(k) be from i states to
J states have the probability that k packet stream enters.
Preferably, in the third step, edge calculations node serve parameter tool is obtained according to the state-transition matrix of system
Body method is as follows：3.1 set the probability of stability of queuing system as πi, the probability of stability of queuing system is solved according to (4) formula,
3.2 obtain the service delay of queuing system according to (5) formula,
3.3 obtain the bandwidth availability ratio of queuing system according to (6) formula,
Preferably, the method for selecting fringe node is as follows：The business that user generates is according to the demand of its service quality QoS
Corresponding edge calculations node is selected, if certain business demand packet loss is small, the packet loss of respective edges calculate node, which is more than, to be used
Packet loss required by the business of family meets selection and requires.
Preferably, in the second step, if the probability that business reaches is π, business is reached according to (7 formula) and carries out probability
Estimation,
π=(π1,Lπi)。
Preferably, in the 4th step, there is certain randomness since business reaches, in order to reduce edge calculations section
A time threshold T is arranged at edge calculations node, is referred to as switching delay phase, edge calculations node for the energy waste of point
After having serviced corresponding service, judge whether there is business arrival in T time, it, will if being reached without business in T time
The service channel of the edge calculations node is closed, and is then turned on when there is business arrival, if there is business arrival in T time,
The service channel of the edge calculations node is kept to open.
Time threshold is set in the present invention, be in order to avoid edge calculations node frequently switch cause energy consumption waste and
Delay.In addition, since the QoS that different business requires under 5G environment is different, can according to the performance parameter solved in above-mentioned steps,
Some screenings are done to edge calculate node, such as the high business of delay requirement, distance may be selected closely, and the edge meter that W is smaller
Operator node.
Advantages of the present invention is as follows：
1. considering the randomness that business reaches, customer service is reached using the Batch Arrival model in queueing theory and is carried out
Probability Estimation, and then the selection of edge calculations node is carried out, to ensure that the business of different demands can be in edge calculations section appropriate
Point carries out business processing, effectively reduces time delay, improves the QoS of mobile communication system；
2. for the sake of energy saving, the switch of a time threshold control edge calculations node is introduced.
In a word since the quantity of 5G environment lower edge equipment is more, the portfolio of outburst is also more, utilizes the batch in queueing theory
Arrival mode selects edge calculations node to have very great help business as needed, can significantly improve the service quality of business.
Description of the drawings
The present invention will be further described below with reference to the drawings.
Fig. 1 is the business classification chart of the present invention.
Fig. 2 is the inflow and outflow figure of state M in the present invention.
Fig. 3 is the schematic diagram that time threshold T is introduced in the present invention.
Fig. 4 is the flow chart of the present invention.
Specific implementation mode
Embodiment one
A kind of customer service queuing optimization method based on edge calculations is present embodiments provided, as shown in figure 4, the party
Method includes the following steps：
The first step, business classification：
The different business for being generated user using packet classifier, according to customer identification number and service quality (QoS) demand into
Then the business classified is put into corresponding sub- caching management module by row classification.
Assuming that a user for carrying sub- caching management module, operator can be cached big with its cache size of dynamic control
Small is hj, which needs j-th stage service, then the bandwidth needed for j-th stage service is cjRb, as shown in Figure 1.
Assuming that the total activation capacity of a certain edge calculations node is bandwidth C, the sub- caching management module of user has m, then
When m sub- caching management modules provide service by the same edge calculations node, required total bandwidth is c, and c is calculated according to (1) formula
Value,
Assuming that the total number of above-mentioned edge calculations node serve user is N, N values are calculated according to (2) formula,
Illustrate that above-mentioned edge calculations node can service N number of user simultaneously.Assuming that above-mentioned edge calculations node supports the total of packet
Number is n, and n values are calculated according to (3) formula,
Wherein, cjFor the packet number of j-th stage business, illustrate that above-mentioned edge calculations node can support n packet simultaneously.
When the number of user or packet exceeds the bearing capacity of edge calculations node, if being continuing with the edge calculations section
Point offer service, then will appear packet loss phenomenon, thus when the edge calculations node at full capacity when, answer broadcast message to shunt industry
Other edge calculations nodes are removed in business.
The traffic handing capacity of different edge calculations nodes is different, the service bandwidth serviced as needed and each end
Each edge calculations node for receiving of mouth send out whether fully loaded information, from the port address of business generation place, from the near to the distant
Edge calculations node is selected, to reduce time delay and reduce packet loss.
Second step establishes Batch Arrival model and reaches progress probability Estimation to business using Batch Arrival model：
Assuming that one has the edge calculations node of h classification caching, while N number of user is serviced, in the edge calculations node
When servicing a certain business A, if business A Batch Arrival intensity is k packet, and 1≤k≤n≤c+h, if the Mean Speed each wrapped
For Rb, the exponential distribution that the inflow parameter for each packet is μ, the exponential distribution that outflow parameter is λ, service regulation is to arrive first
It first services, based on this, queue theory model MAP is established using Markov ChainX/ M/c/c+h, wherein n are that maximum batch is strong
Degree, MAP are batch Markov.
If markovian state space is S={ 1,2, L, m }, j states are transferred to from i states, wherein 1≤i, j
≤ m≤n, for whether thering is batch event to reach there are two types of situation, if there is the event that batch intensity is k to reach, probability pij
(k), if when without event arrival and i ≠ j, probability pij(0).Wherein pij(0) it is not have packet stream from i states to j states
The probability entered, pij(k) it is to have the probability that k packet stream enters from i states to j states.Thus, only batch event reaches,
Just state i can be made to return to itself.It can be obtained by the relationship
Wherein, 1≤i≤m≤n.It enablesWherein1≤i, j≤m≤n,
μiEnter the parameter of the exponential distribution of obedience for each packet stream.
The inflow of state m by m+1 states inflow (m+1) μ and Batch Arrival DkComposition, the outflow of state m is by m-1 shapes
The outflow m μ and Batch Arrival D of statekComposition, the inflow and outflow figure of state m are as shown in Figure 2.Due to m<N reaches inflow state m
Maximum value be Dm.If the state-transition matrix of queuing system is Q, then the state-transition matrix Q of system
Wherein, D is Batch Arrival rate, with Dmin(m,n)For, it is the maximum Batch Arrival rate of arrival state m, DkFor m+
The batch rate of inflow of k-state, if the probability of stability of queuing system is πi, the stable state that queuing system can be solved according to (4) formula is general
Rate,
If the probability that business reaches is π, business is reached according to (7 formula) and carries out probability Estimation,
π=(π1,Lπi)(7)。
The service delay of queuing system is obtained according to (5) formula,
The bandwidth availability ratio of queuing system is obtained according to (6) formula,
Third step, the service parameter that edge calculations node is obtained according to Batch Arrival model, then according to edge calculations section
The service parameter of point selects fringe node.
The method for selecting fringe node is as follows：The business that user generates selects corresponding according to the demand of its service quality QoS
Edge calculations node, if certain business demand packet loss is small, the packet loss of respective edges calculate node is more than customer service institute
It is required that packet loss, meet selection require.
4th step, setting time threshold are T, the opening and closing of edge calculations node are controlled according to time threshold T, to save energy
Source.
As shown in figure 3, there is certain randomness since business reaches, in order to reduce the energy wave of edge calculations node
Take, a time threshold T is set at edge calculations node, is referred to as the switching delay phase, edge calculations node is servicing phase
After answering business, judge whether there is business arrival in T time, if being reached without business in T time, by the edge calculations
The service channel of node is closed, and is then turned on when there is business arrival, if there is business arrival in T time, is kept the edge
The service channel of calculate node is opened.
In addition to the implementation, the present invention can also have other embodiment.It is all to use equivalent substitution or equivalent transformation shape
At technical solution, fall within the scope of protection required by the present invention.
Claims (10)
1. a kind of customer service based on edge calculations is lined up optimization method, which is characterized in that include the following steps：
The first step, the different business for being generated user using packet classifier are carried out according to customer identification number and QoS requirement
Then the business classified is put into corresponding sub- caching management module by classification；
Second step establishes Batch Arrival model and reaches progress probability Estimation to business using Batch Arrival model；
Third step, the service parameter that edge calculations node is obtained according to Batch Arrival model, then according to edge calculations node
Service parameter selects fringe node；
4th step, setting time threshold are T, the opening and closing of edge calculations node are controlled according to time threshold T, with energy saving.
2. a kind of customer service based on edge calculations is lined up optimization method according to claim 1, which is characterized in that described
In the first step, sub- caching management module meets claimed below：
1.1 assume a user for carrying sub- caching management module, cache size hj, caching needs j-th stage service, then and the
Bandwidth needed for j grades of services is cjRb；
1.2 assume that the total activation capacity of a certain edge calculations node is bandwidth C, and the sub- caching management module of user has m, then m
When a sub- caching management module provides service by the same edge calculations node, required total bandwidth is c, and c is calculated according to (1) formula
Value,
1.3 set the total number of edge calculations node serve user as N, and N values are calculated according to (2) formula,
Illustrate that edge calculations node can service N number of user simultaneously,
If edge calculations node supports that the total number of packet is n, n values are calculated according to (3) formula,
Wherein, cjFor the packet number of j-th stage business, illustrate that edge calculations node can support n packet simultaneously.
3. being lined up optimization method according to a kind of customer service based on edge calculations described in claim 2, it is characterised in that：Work as user
It, can if being continuing with the edge calculations node provides service or when bearing capacity of the number of packet beyond edge calculations node
There is packet loss phenomenon, answers broadcast message to shunt business and remove other edge calculations nodes.
4. being lined up optimization method according to a kind of customer service based on edge calculations described in claim 3, it is characterised in that：According to need
Each edge calculations node that the service bandwidth and each port to be serviced receive send out whether fully loaded information, produced from business
Port address at raw selects edge calculations node, to reduce time delay and reduce packet loss from the near to the distant.
5. a kind of customer service based on edge calculations is lined up optimization method according to claim 4, which is characterized in that described
In second step, establishing Batch Arrival model, the specific method is as follows：
2.1 assume that has an edge calculations node for h classification caching, while servicing N number of user, in the edge calculations node
When servicing a certain business A, if business A Batch Arrival intensity is k packet, and 1≤k≤n≤c+h, if the Mean Speed each wrapped
For Rb, the exponential distribution that the inflow parameter for each packet is μ, the exponential distribution that outflow parameter is λ, service regulation is to arrive first elder generation
Service, queue theory model MAP is established using Markov ChainX/ M/c/c+h, wherein n are maximum batch intensity, and MAP is batch horse
Er Kefu；
2.2 set markovian state space as S={ 1,2, L, m }, j states are transferred to from i states, wherein 1≤i, j
≤ m≤n, if the probability that batch event reaches is Pij, when there is the event that batch intensity is k to reach, probability pij(k), when
When there is no event arrival and i ≠ j, probability pij(0)；
2.3 set the state-transition matrix of queuing system as Q, then
Wherein, D is Batch Arrival rate, Dmin(m,n)To reach the maximum Batch Arrival rate of state m, DkFor the bulk stream of m+k states
Enter rate, enablesAnd 1≤i, j≤m≤n, μiEnter obedience for each packet stream
The parameter of exponential distribution.
6. a kind of customer service based on edge calculations is lined up optimization method according to claim 5, it is characterised in that：pij
(0) it is not have the probability that packet stream enters, p from i states to j statesij(k) it is to have the probability that k packet stream enters from i states to j states.
7. a kind of customer service based on edge calculations is lined up optimization method according to claim 6, which is characterized in that described
In third step, obtaining edge calculations node serve parameter according to the state-transition matrix of system, the specific method is as follows：3.1 set queuing
The probability of stability of system is πi, the probability of stability of queuing system is solved according to (4) formula,
3.2 obtain the service delay of queuing system according to (5) formula,
3.3 obtain the bandwidth availability ratio of queuing system according to (6) formula,
8. a kind of customer service based on edge calculations is lined up optimization method according to claim 7, which is characterized in that selection
The method of fringe node is as follows：The business that user generates selects corresponding edge calculations section according to the demand of its service quality QoS
Point, if certain business demand packet loss is small, the packet loss of respective edges calculate node is more than the packet loss required by customer service,
Meet selection to require.
9. a kind of customer service based on edge calculations is lined up optimization method according to claim 8, which is characterized in that described
In second step, if the probability that business reaches is π, business is reached according to (7 formula) and carries out probability Estimation,
π=(π1,L πi)。
10. a kind of customer service based on edge calculations is lined up optimization method according to claim 9, which is characterized in that institute
State in the 4th step, at edge calculations node be arranged a time threshold T, edge calculations node after having serviced corresponding service,
Judge whether there is business arrival in T time, if being reached without business in T time, the business of the edge calculations node is led to
Road is closed, and is then turned on when there is business arrival, if there is business arrival in T time, is kept the business of the edge calculations node
It opens in channel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810132806.4A CN108513318B (en) | 2018-02-09 | 2018-02-09 | User service queuing optimization method based on edge calculation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810132806.4A CN108513318B (en) | 2018-02-09 | 2018-02-09 | User service queuing optimization method based on edge calculation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108513318A true CN108513318A (en) | 2018-09-07 |
| CN108513318B CN108513318B (en) | 2021-04-06 |
Family
ID=63375645
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810132806.4A Active CN108513318B (en) | 2018-02-09 | 2018-02-09 | User service queuing optimization method based on edge calculation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108513318B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109597680A (en) * | 2018-10-22 | 2019-04-09 | 阿里巴巴集团控股有限公司 | Task queue's response parameter evaluation method and device |
| CN111082997A (en) * | 2019-12-30 | 2020-04-28 | 西安电子科技大学 | Network function arrangement method based on service identification in mobile edge computing platform |
| CN114095869A (en) * | 2022-01-07 | 2022-02-25 | 广东海洋大学 | Method, device and system for scheduling multi-access edge computing nodes for terminal |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103632209A (en) * | 2013-11-28 | 2014-03-12 | 国家电网公司 | Queuing theory-based data transmission bandwidth prediction method in intelligent power distribution and consumption business |
| CN106790092A (en) * | 2016-12-23 | 2017-05-31 | 北京奇艺世纪科技有限公司 | Remote procedure call services end control system and method |
| CN106936645A (en) * | 2017-04-19 | 2017-07-07 | 西安电子科技大学 | The optimization method of the tree network topology structure based on queueing theory |
| CN107450982A (en) * | 2017-06-07 | 2017-12-08 | 上海交通大学 | A kind of method for scheduling task based on system mode |
-
2018
- 2018-02-09 CN CN201810132806.4A patent/CN108513318B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103632209A (en) * | 2013-11-28 | 2014-03-12 | 国家电网公司 | Queuing theory-based data transmission bandwidth prediction method in intelligent power distribution and consumption business |
| CN106790092A (en) * | 2016-12-23 | 2017-05-31 | 北京奇艺世纪科技有限公司 | Remote procedure call services end control system and method |
| CN106936645A (en) * | 2017-04-19 | 2017-07-07 | 西安电子科技大学 | The optimization method of the tree network topology structure based on queueing theory |
| CN107450982A (en) * | 2017-06-07 | 2017-12-08 | 上海交通大学 | A kind of method for scheduling task based on system mode |
Non-Patent Citations (1)
| Title |
|---|
| XU CHEN等: "Efficient Multi-User Computation Offloading for Mobile-Edge Cloud Computing", 《IEEE/ACM TRANSACTIONS ON NETWORKING》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109597680A (en) * | 2018-10-22 | 2019-04-09 | 阿里巴巴集团控股有限公司 | Task queue's response parameter evaluation method and device |
| CN111082997A (en) * | 2019-12-30 | 2020-04-28 | 西安电子科技大学 | Network function arrangement method based on service identification in mobile edge computing platform |
| CN111082997B (en) * | 2019-12-30 | 2021-05-14 | 西安电子科技大学 | Network function arrangement method based on service identification in mobile edge computing platform |
| CN114095869A (en) * | 2022-01-07 | 2022-02-25 | 广东海洋大学 | Method, device and system for scheduling multi-access edge computing nodes for terminal |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108513318B (en) | 2021-04-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102915254B (en) | task management method and device | |
| CN113138860B (en) | Message queue management method and device | |
| CN108513318A (en) | A kind of customer service queuing optimization method based on edge calculations | |
| CN115665227B (en) | Universal heterogeneous integrated computing network resource intelligent adaptation network architecture and method | |
| CN107404443A (en) | Queue cache resources control method and device, server and storage medium | |
| CN107113251A (en) | Dispatched using the dynamic bandwidth of transmission net | |
| CN109151077A (en) | One kind being based on goal-oriented calculating discharging method | |
| CN110474842A (en) | A kind of communication means, customer terminal equipment and transmission network | |
| CN117762597A (en) | Yun Bian-based cooperative intelligent scheduling algorithm | |
| CN100466593C (en) | Method of implementing integrated queue scheduling for supporting multi service | |
| CN114710571A (en) | Data packet processing system | |
| CN117439655B (en) | Space terahertz information center network lightweight caching method and device | |
| CN107659505A (en) | The route selecting method and SDN controllers of a kind of SDN | |
| CN111405614B (en) | Method for calculating APP load sharing at mobile edge | |
| CN117201408A (en) | Large-scale internet of things equipment load balancing solution method based on group game | |
| Wang et al. | Analytical modeling of an integrated priority and WFQ scheduling scheme in multi-service networks | |
| Chousainov et al. | An analytical framework of a C-RAN supporting random, quasi-random and bursty traffic | |
| Zhao et al. | A research of task-offloading algorithm for distributed vehicles | |
| CN115442313A (en) | Wide-area deterministic service flow online scheduling system | |
| Mahi et al. | A New Unified Communication Approach to comply Bandwidth Optimization Technique using Dynamic Channel Allocation. | |
| Hasnain et al. | Coflow scheduling with performance guarantees for data center applications | |
| CN100442760C (en) | Packet equity dispatching method and apparatus | |
| Chen et al. | Caching with statistical request delay information | |
| Szabó et al. | Worst-case deterministic delay bounds for arbitrary weighted generalized processor sharing schedulers | |
| Chervenets et al. | Quality increasing of traffic flow transferring in multiservice 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 |