WO2017193850A1 - 一种终端的接入调度方法、装置及计算机存储介质 - Google Patents

一种终端的接入调度方法、装置及计算机存储介质 Download PDF

Info

Publication number
WO2017193850A1
WO2017193850A1 PCT/CN2017/082938 CN2017082938W WO2017193850A1 WO 2017193850 A1 WO2017193850 A1 WO 2017193850A1 CN 2017082938 W CN2017082938 W CN 2017082938W WO 2017193850 A1 WO2017193850 A1 WO 2017193850A1
Authority
WO
WIPO (PCT)
Prior art keywords
cluster
terminal
resource
machine
access
Prior art date
Application number
PCT/CN2017/082938
Other languages
English (en)
French (fr)
Inventor
吕阳阳
宋应湃
郑泽栋
杨阳
Original Assignee
腾讯科技（深圳）有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 腾讯科技（深圳）有限公司 filed Critical 腾讯科技（深圳）有限公司
Priority to EP17795468.2A priority Critical patent/EP3451727B1/en
Publication of WO2017193850A1 publication Critical patent/WO2017193850A1/zh
Priority to US16/047,402 priority patent/US10652360B2/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/12—Wireless traffic scheduling
    • H04W72/121—Wireless traffic scheduling for groups of terminals or users
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
    • H04L67/1004—Server selection for load balancing
    • H04L67/101—Server selection for load balancing based on network conditions
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
    • H04L67/61—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources taking into account QoS or priority requirements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
    • H04W16/02—Resource partitioning among network components, e.g. reuse partitioning
    • H04W16/10—Dynamic resource partitioning
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
    • H04W16/18—Network planning tools
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/06—Selective distribution of broadcast services, e.g. multimedia broadcast multicast service [MBMS]; Services to user groups; One-way selective calling services
    • H04W4/08—User group management
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/50—Allocation or scheduling criteria for wireless resources
    • H04W72/54—Allocation or scheduling criteria for wireless resources based on quality criteria

Definitions

  • the present invention relates to the field of communications technologies, and in particular, to a terminal access scheduling method, apparatus, and computer storage medium.
  • the access scheduling of the terminal is an important link, and the access scheduling is a control that specifies which servers access a certain terminal and provides services for the terminal. logic.
  • a commonly used access scheduling solution is generally scheduled based on location, specifically, acquiring geographic location information of the terminal (which may be acquired through IP or directly), and then, correspondingly, according to the geographical location information, corresponding A machine (such as a server) provides services for the terminal, that is, the corresponding machine accesses the terminal to provide services for the terminal; for example, after acquiring the geographical location information of the terminal, assigning a machine that is close to or closest to the geographic location of the terminal is The terminal provides a service, and the access scheduling at this time may be referred to as a nearby access scheduling; if the IP address of the terminal is obtained, the location of the IP address corresponding to the IP needs to be searched in the preset IP library, and then allocated and A terminal with a similar geographical location or the nearest machine serves the terminal.
  • the inventors of the present invention have found that the existing access scheduling scheme only performs access scheduling based on the geographical location of the terminal, and the obtained in the scheme IP library, IP or geographic location information may be inaccurate, and as a result, the access scheduling of the terminal is less accurate, especially when faced with massive services.
  • the embodiments of the present invention provide a method and device for access scheduling of a terminal and a computer storage medium, which can improve the accuracy of access scheduling of the terminal.
  • An embodiment of the present invention provides an access scheduling method for a terminal, including:
  • Clustering terminals to obtain multiple terminal clusters
  • Clustering the machines that provide resources to get multiple machine clusters
  • An embodiment of the present invention further provides an access scheduling apparatus for a terminal, including:
  • a first dividing unit configured to perform clustering on the terminal to obtain multiple terminal clusters
  • a second dividing unit configured to perform clustering on a machine that provides resources to obtain a plurality of machine clusters
  • a first acquiring unit configured to acquire resource requirements of the terminal cluster, available resources of the machine cluster, and access quality between the terminal cluster and the machine cluster;
  • a second acquiring unit configured to acquire, according to a resource requirement of the terminal cluster, an available resource of the machine cluster, and the access quality, a proportion of the cluster that covers the terminal cluster;
  • the scheduling unit is configured to perform access scheduling on the terminal according to the ratio.
  • the embodiment of the present invention further provides a computer storage medium, where the computer storage medium stores computer executable instructions, and the computer executable instructions are used to execute the embodiment of the present invention.
  • the access scheduling method is not limited to:
  • the terminal is clustered to obtain a plurality of terminal clusters, and the machines providing the resources are divided to obtain a plurality of machine clusters, and then the resource requirements of the terminal cluster, available resources of the machine cluster, and And the access quality between the terminal cluster and the machine cluster, according to resource requirements of the terminal cluster, available resources of the cluster, and the access quality, obtaining a proportion of the cluster of the terminal covering the terminal cluster, according to the ratio
  • the terminal performs access scheduling; the solution can perform access scheduling based on the available resources of the cluster, the resource requirements of the terminal cluster, and the access quality between the terminal cluster and the machine cluster. It can be seen that the scheduling scheme takes into account the access capacity. And access quality, therefore, the access scheduling of the terminal can be optimized, and the accuracy of the access scheduling of the terminal is improved relative to the prior art.
  • FIG. 1 is a flowchart of a method for access scheduling of a terminal according to Embodiment 1 of the present invention
  • FIG. 2 is a flowchart of a method for access scheduling of a terminal according to Embodiment 2 of the present invention
  • FIG. 3 is a schematic structural diagram of an access scheduling apparatus of a terminal according to Embodiment 3 of the present invention.
  • FIG. 3b is a schematic structural diagram of another terminal access scheduling apparatus according to Embodiment 3 of the present invention.
  • the embodiment of the invention provides a method and device for access scheduling of a terminal. The details will be described separately below.
  • a terminal access scheduling method includes: clustering a terminal to obtain a plurality of terminal clusters, dividing a machine providing resources to obtain a plurality of machine clusters, and then acquiring resource requirements of the terminal cluster, Obtaining the available resources of the machine cluster, and the access quality between the terminal cluster and the machine cluster, obtaining the machine cluster covering the terminal cluster according to resource requirements of the terminal cluster, available resources of the machine cluster, and the access quality.
  • the ratio is based on the ratio of the terminal to perform access scheduling.
  • the specific process of the access scheduling method of the terminal can be as follows:
  • Step 101 Perform cluster division on the terminal to obtain multiple terminal clusters.
  • the division of the terminal may be performed in multiple ways.
  • the terminal may be divided according to the geographic location of the terminal, that is, the clustering of the terminal may include:
  • the terminal is clustered according to the geographical location of the terminal.
  • the geographic location of the terminal may include the current geographic location of the terminal, or the network registration attribution of the terminal; for example, the cluster may be divided according to the current location of the terminal or the area (region, city, province, country, etc.) registered by the network, specifically, The same terminal in the area where the current network or the network is registered is classified into the same terminal cluster.
  • the clustering may be performed based on the network address (such as an IP address) of the terminal, that is, the clustering of the terminal may include:
  • the terminal is clustered according to the network address of the terminal.
  • the clustering of the terminal may include: dividing the terminal according to the IP address segment to which the IP address of the terminal belongs, for example, dividing the IP address of the terminal into the same preset IP address segment into a terminal cluster.
  • the member in the terminal cluster may be a single terminal or a terminal combination, and may be set according to actual requirements.
  • the terminal may be a device in the computer network that is at the outermost periphery of the network, that is, the terminal may be a user device, for example, the terminal may be a personal computer, a mobile phone, a tablet computer, or other user equipment.
  • Step 102 Perform clustering on a machine that provides resources to obtain multiple machine clusters.
  • the machine providing the resource may be a server or other device that can provide resources.
  • the manner in which the machine is divided may be various.
  • the device may be divided according to the geographic location of the machine, that is, the clustering of the machine providing the resources may include:
  • the machine is clustered according to the geographic location of the machine that provides the resource.
  • the clustering may be performed according to the area (region, city, province, country, etc.) where the machine is located, for example, dividing a machine located in a certain province into a machine cluster.
  • step 101 may be before step 102, or step 102 may be before step 101.
  • step 101 and step 102 may be performed simultaneously. Wait.
  • Step 103 Obtain resource requirements of the terminal cluster, available resources of the machine cluster, and access quality between the terminal cluster and the machine cluster.
  • the method for obtaining the resource requirements of the terminal cluster may be multiple, for example, may be obtained based on the historical request quantity of the terminal cluster, that is, the resource requirements of the acquiring the terminal cluster may include:
  • the historical request amount of each terminal can be separately obtained, and then the historical request amount of each terminal is analyzed to obtain the resource requirements of each terminal, and the resources of the 10 terminals are respectively used. The requirements are added together to get the resource requirements of the cluster.
  • the resource requirement of the terminal cluster may be acquired by using a supervisory evaluation or the like, that is, the resource requirement of the terminal cluster may be included: receiving a resource requirement of the terminal cluster input by the user.
  • the available resources of the machine cluster may be obtained in various manners, for example, by using a stress test method, specifically, performing stress tests on the machines in the machine cluster to obtain available resources of the machine. Then, the available resources of the machine cluster are obtained based on the available resources of the machines in the machine cluster. In the actual test, in order to improve the acquisition speed, the efficiency of the access scheduling can be improved based on the type of the machine, and then the available resources of the machine cluster can be obtained by combining the number of different types of machines; that is, the process of obtaining the available resources of the machine cluster can be include:
  • the available resources of the machine cluster are obtained according to the available resources of the different types of machines and the number of different types of machines.
  • the pressure test can be used to obtain the available resources, but also
  • the available resource amount may be obtained based on the historical operation information of the machine; that is, the process of obtaining the available resources of the machine cluster may include:
  • the historical operation information may include at least one of historical resource consumption information and historical service processing amount.
  • the embodiment may obtain the available resources of the machine cluster based on the type and quantity of the machine;
  • the available resources of the machine cluster are obtained according to the available resources of different types of machines and the number of different types of machines.
  • the access quality between the terminal cluster and the machine cluster may be the network service quality between the terminal cluster and the machine cluster; wherein the network service quality (ie, access quality) between the terminal and the machine may be based on the terminal. Obtained from at least one of a network performance parameter between the machine, historical operation information between the terminal and the machine, a physical distance between the terminal and the machine, a network address of the terminal, and a network topology diagram of the network system of the terminal and the machine to which the machine belongs.
  • the network performance parameters may include: network delay, packet loss rate, and delay jitter.
  • Step 104 Obtain a proportion of the machine cluster accessing the terminal cluster according to resource requirements of the terminal cluster, available resources of the machine cluster, and the access quality.
  • the proportion of the cluster that covers the terminal cluster is the percentage of the resources that the cluster provides for the terminal cluster to the resource requirements of the terminal cluster; for example, the cluster 1 covers 30% of the terminal cluster 1, which indicates that the cluster 1 is the terminal.
  • the resources provided by cluster 1 account for 30% of the resource requirements of the terminal cluster 1; the cluster 1 covers 50% of the terminal clusters 2, which indicates that the resources provided by the cluster 1 for the terminal cluster 2 account for 50 of the resource requirements of the terminal cluster 2.
  • the ratio may be employed n ij denotes the j n ij represents a cluster of machines covering ratio of the terminal cluster i, the i and j are positive integers greater than 1.
  • the manner in which the machine cluster covers the proportion of the terminal cluster according to the resource requirement, the available resource, and the access quality may be multiple.
  • the total resource requirement of the terminal cluster to the machine cluster may be less than the available of the machine cluster.
  • the resource is a constraint, and then, according to the constraint condition and the access quality between the terminal cluster and the machine cluster, the proportion of the cluster that covers the terminal cluster is obtained; that is, according to the resource requirement of the terminal cluster, the available of the cluster is
  • the resource and the access quality, and the proportion of the machine cluster covering the terminal cluster may include:
  • the preset restriction condition includes: the sum of the resource requirements of the terminal cluster for the machine cluster is less than or equal to the available resources of the machine cluster; wherein the sum of the resource requirements is obtained according to the resource requirements of the terminal cluster and the proportion to be acquired.
  • the sum of the resource requirements of the terminal cluster to the machine cluster is the sum of the resources that the machine cluster needs to provide for all the terminal clusters; for example, for the machine cluster 1, if there are m terminal clusters, the terminal clusters the cluster of the machine.
  • the sum of the resource requirements of 1 can be the resource that the machine cluster 1 needs to provide for the terminal cluster 1, the resources provided for the terminal cluster 2, the sum of the resources provided by the terminal cluster m; the sum of the resource requirements of the terminal cluster to the machine cluster can be based on The resource requirements of the terminal cluster and the proportion of the clusters covering the terminal clusters are obtained; for example, the sum of the resource requirements of the terminal cluster for the machine cluster 1 can be based on the resource requirement u1 of the terminal cluster 1 and the ratio n 11 of the cluster cluster 1 covering the terminal cluster 1
  • a resource requirement set, an available resource set, and an access quality set may be obtained, where the resource requirement set includes resource requirements of multiple terminal clusters, and the available resource collection package
  • the resource requirement u of each terminal cluster can be obtained to obtain a resource requirement set U, which includes: u1, u2, u3, ... ui, um, where ui Indicates the resource requirement size of the terminal cluster i, the i is smaller than m; obtain the available resources a of each machine cluster to obtain the available resource set A, and the A includes: a1, a2, a3, ...
  • aj...an where aj Representing the available resource size of the machine cluster j, where j is less than n; obtaining the access quality d between each terminal cluster and the machine cluster to obtain an access quality set D, where D includes: d 11 , d 12 , d 13 ...d 1n , d 21 , d 22 ... d 2n ..., d m1 , d m2 ..., d mn , where d ij represents the quality of access between the terminal cluster i and the machine cluster j; For n ij , the n ij represents the proportion of the machine cluster j covering the terminal cluster i.
  • the ratio of the total access quality in the access quality set may be obtained, that is, the terminal cluster and the machine cluster are used by default in order to improve the resource utilization and service quality of the system.
  • the optimal scheduling is performed.
  • the proportion of the machine cluster covering the terminal cluster may be:
  • the preset constraint includes: the sum of the access qualities in the access quality set is the largest.
  • n ij is obtained according to the preset constraint condition and the access quality set D to obtain the proportional set N.
  • the matrix operation can be used to calculate the proportional set, that is, the problem that the scale set of the machine cluster overlay terminal is converted into the solution matrix; for example, it can be connected Generating a corresponding matrix into the quality set, and then obtaining a corresponding weight matrix according to the preset constraint condition, the matrix and the preset constraint condition, wherein the matrix element in the weight matrix represents the proportion of the machine cluster covering the terminal cluster; That is, the obtaining the ratio set according to the preset constraint condition, the access quality set, and the preset constraint condition may include:
  • performing the access scheduling on the terminal according to the ratio in the following step 105 may include: performing access scheduling on the terminal according to the weight matrix.
  • a corresponding resource requirement matrix U m also referred to as a user vector U m , may be created according to the resource requirement set U, (u i represents the resource requirement size of the terminal cluster i):
  • a corresponding resource requirement matrix A n is generated according to the available resource set A, (a j represents the available resource size of the cluster j):
  • a n [a 1 ,a 2 ,...a n ];
  • Creating a corresponding access quality matrix D m ⁇ n according to the access quality set may also be referred to as an access quality vector D m ⁇ n (d ij represents the access quality between the terminal cluster i and the machine cluster j):
  • d ij can be a logical distance between the terminal i and the machine cluster j, the logical distance indicating the access quality between the terminal cluster i and the machine cluster j, and the greater the logical distance, the greater the access quality (ie, the better) The smaller the logical distance, the smaller the access quality (ie, the worse).
  • the weight matrix to be solved is N m ⁇ n , where n ij represents the proportion of the machine cluster j covering the terminal cluster i, and the resource restriction condition may be increased to ensure that the allocated resources do not exceed the available resources.
  • the total resource requirement of the terminal cluster for the machine cluster is less than or equal to the available resources of the machine cluster, that is, U m ⁇ N m ⁇ n ⁇ A n .
  • the sum of the access quality between the terminal cluster and the machine cluster can be maximized as the optimal scheduling.
  • the terminal i and the machine cluster j can be used.
  • the logical distance between the logical distances is the smallest and the optimal scheduling is used, that is, the calculation of N m ⁇ n increases the maximum constraint of the sum of the access quality between the terminal cluster and the machine cluster, wherein the terminal cluster and the machine cluster
  • the sum of the access quality between the maximum can be represented by the access quality matrix D m ⁇ n and the weight matrix N m ⁇ n to be obtained, as can be passed
  • the minimum or maximum value indicates that the sum of the access qualities is the largest; specifically, when the larger d ij indicates that the access quality is larger (ie, the better), the preset constraint may be Maximum, when d ij is larger, indicating that the access quality is smaller (ie, worse), the preset constraint may be The smallest.
  • Step 105 Perform access scheduling on the terminal according to the ratio.
  • the terminal may perform access scheduling according to the proportional set.
  • the performing the access scheduling on the terminal according to the ratio may include:
  • a target machine that the terminal needs to access in the terminal cluster is determined from the target machine cluster.
  • the ratio n 11 of the terminal cluster 1 is covered by the machine cluster 1 and the ratio n 12 of the terminal cluster 1 is covered by the machine cluster 2.
  • the machine cluster n covers the ratio n 1n of the terminal cluster 1 to acquire the target of the terminal cluster 1 access. Machine cluster.
  • the preset algorithm may be used to calculate the ratio to obtain the operation result, and the target machine cluster accessed by the terminal cluster is determined according to the operation result; wherein the preset algorithm may be set according to actual requirements, such as the pre- Let the algorithm be a weighting algorithm and so on.
  • the embodiment may further adjust the foregoing resource requirements, available resources, and access quality, specifically, to reduce the pressure of the machine cluster and improve the service quality of the service. It is generally desirable to reserve resources, that is, do not want the resource utilization rate of the machine cluster to be 100%; for example, to prevent resource waste and improve resource utilization, it is hoped that the resource utilization rate of some machine clusters can be higher, such as a certain For overseas machine clusters, only 60% is too wasteful, and it is hoped that 90% can be used. In this case, the resource utilization rate can be used to process the available resources of the machine cluster, that is, according to the resource requirements of the terminal cluster.
  • the available resources of the cluster of the machine and the access quality, and the proportion of the cluster of the machine that covers the terminal cluster may include:
  • the preset upper limit resource usage rate represents a resource usage upper limit of the machine cluster.
  • the available resources of the cluster of the machine may be multiplied by the preset upper limit resource usage rate of the machine cluster to obtain the upper limit available resources of the machine cluster, and the assumed upper limit resource usage rate of the machine cluster n is r n .
  • the resource usage vector R n when solving the scale by matrix operation, the resource usage vector R n can be increased:
  • R n [r 1 ,r 2 ,...r n ], where r j is the preset upper limit resource usage rate of the machine cluster j.
  • the process of obtaining the proportion according to the resource requirement, the upper limit available resource, and the access quality is similar to the above step 104.
  • the only difference is that one is an available resource, and the other is an upper limit available resource, and the available resources of the above machine cluster are replaced with the upper limit available resource.
  • the resource usage upper limit is set, that is, the size of the available resources is strictly limited. Therefore, there may be a failure in the proportion acquisition (that is, there is no solution to the resource allocation, such as no solution to the weight matrix) or the scheduling accuracy is changed. Poor problems, such as a large fleet, stable operating demand is about 60% of the overall resource consumption, but when the demand for this point increases, it is necessary that the carrying capacity of the point temporarily exceeds 60% of the resources, instead of There is no solution to the allocation due to insufficient resources.
  • the method in this embodiment may also increase the preset absolute upper limit resource usage rate of the machine cluster, and process the upper limit available resources of the machine cluster through the usage rate. Then, based on the processed upper limit, the ratio can be obtained by the available resources, the access quality, and the resource requirement; that is, the method in this embodiment may further include:
  • the upper limit available resources of the machine cluster are processed according to the preset absolute upper limit resource usage rate to obtain an absolute upper limit available resource of the machine cluster, where the preset is Absolute cap resource usage is greater than the default cap resource Usage rate
  • the proportion of the cluster that covers the terminal cluster is obtained.
  • the preset absolute upper limit resource usage rate is an absolute upper limit of the resource usage rate, which is greater than the upper limit of the resource usage rate.
  • the upper limit available resource of the machine cluster can be multiplied by the preset absolute upper limit resource usage of the machine cluster and then divided by the preset upper limit resource to obtain the absolute upper limit available resources of the machine cluster, that is, the available resources of the machine cluster are multiplied.
  • the preset absolute upper limit resource of the machine cluster it is assumed that the preset upper limit resource usage rate of the machine cluster n is r n , and the preset absolute upper limit resource usage rate is r n ', and when the acquisition ratio fails, the r n 'pair is available.
  • the upper limit of the machine cluster can be processed by resources, such as the absolute upper limit available resources of the machine cluster n Then re-acquire the ratio based on the absolute upper limit available resources, access quality, and resource requirements of the machine cluster.
  • the failure to obtain the proportion may be met.
  • the failure to meet the preset restriction condition may result in the failure of the acquisition ratio, that is, the sum of the resource requirements of the terminal cluster to the machine cluster is greater than the upper limit available resources of the machine cluster.
  • the upper limit available resources of the machine cluster can be processed by using the preset absolute upper limit resource usage rate of the machine cluster to obtain the absolute upper limit available resources of the machine cluster, and then, according to the absolute upper limit available resources, resource requirements, and access quality. Regain the scale.
  • the resource usage rate vector R n ' may be increased, where r' j represents a preset absolute upper limit resource usage rate of the machine cluster j;
  • r j ' can be used to process a j ', for example, by formula: Calculate the absolute upper limit of the machine cluster j available resources a j ”, and then solve the weight matrix.
  • the (r' j -r j ) is the capacity expansion requirement of the machine cluster j.
  • the capacity expansion requirement can be recorded, so that the capacity expansion is performed based on the capacity expansion requirement.
  • the process according to the resource requirement, the absolute upper limit available resource, and the access quality acquisition ratio is similar to the foregoing step 104, except that one is an available resource, and one is an absolute upper limit available resource, and the available resources of the foregoing machine cluster may be used. Replace with the upper limit available resources to get the scale.
  • the Guangdong terminal obtains resources preferentially compared to Guangxi;
  • Obtain a priority, and set a corresponding resource acquisition priority for each terminal cluster, and the resource acquisition priority may also be taken into account in the access scheduling, that is, according to the preset restriction condition and the terminal cluster and the machine
  • the access quality between the clusters, and the proportion of the cluster accessing the cluster to the terminal may include:
  • obtaining the cluster of the machine to cover the terminal cluster may include:
  • the resource acquisition priority of the machine cluster can be obtained according to the attributes of the machine cluster. Specifically, the resource acquisition priority of the machine cluster is higher than that of the Jiangxi machine cluster.
  • the obtaining the proportional set according to the preset constraint condition, the access quality set, the priority set, and the preset constraint condition may include:
  • a priority vector P m may be added, where p i represents a resource acquisition priority of the terminal cluster i, and at this time, it may be solved by The lowest weight matrix N m ⁇ n to get the ratio.
  • the problem of conflict between automatic scheduling and manual adjustment is solved, for example, all resources of the United States need to be used for covering the United States, or 30% of Guangdong Telecom users are required to access the Shenzhen Telecom fleet.
  • the method in this embodiment may also preferentially ensure manual adjustment; that is, the preset constraint condition in this embodiment may further include:
  • the proportion of the proportion in the proportional set is a predetermined ratio.
  • the ratio n 11 of the machine cluster 1 covering the terminal cluster 2 is equal to 30%, and the ratio n 21 of the cluster 1 covering the terminal cluster 2 is equal to 20% and the like.
  • the partial value of the solution vector N m ⁇ n can be initialized before the calculation.
  • n 11 30% is set, wherein the n11 can represent the proportion of the Shenzhen telecommunications fleet covering the Guangdong telecom terminal cluster, and then the subsequent calculation of the weight matrix is performed.
  • the method in this embodiment may further include:
  • the system to which the machine cluster belongs is expanded according to the resource gap.
  • the resource demand vector U m can be adjusted; specifically, the resource demand can be processed according to the expected growth amount of the terminal, that is, the resource requirement according to the terminal cluster, the available resources of the machine cluster, and the access quality.
  • Obtaining the cluster of the machine to access the terminal cluster may include:
  • the resource requirements of the terminal cluster are processed according to the expected growth amount of the terminal corresponding to the terminal cluster, to obtain an expected resource requirement of the terminal cluster;
  • the proportion of the cluster that covers the terminal cluster is obtained.
  • the ratio at this time is to take into account the ratio of the expected growth of the terminal.
  • the resource gap can be obtained according to the ratio and the resource demand of the terminal cluster, and the capacity gap can be expanded based on the resource gap; thus, the expected growth of the terminal can be obtained.
  • the capacity of the cluster is expanded to provide a reference for capacity expansion and capacity expansion based on the capacity expansion. This improves the accuracy of resource expansion and resource utilization.
  • the process according to the expected resource requirement, the available resource, and the access quality acquisition ratio is similar to the foregoing step 104.
  • the only difference is that one resource requirement and one expected resource requirement, and the resource requirements of the terminal cluster can be replaced with Expect resource requirements to get a percentage.
  • the terminal expected growth amount may be the terminal growth ratio.
  • the expected growth amount may be multiplied by the resource demand sent by the terminal, and the expected resource requirement of the terminal cluster may be obtained, for example, for the terminal cluster i, the terminal cluster
  • the resources of a specific cluster can only be applied for.
  • resources of certain points need to be abolished, and some points can only be applied for.
  • resource constraints must be considered in the expansion, that is, the resource growth is limited.
  • the resources can be constrained.
  • the ratio of the available resources of the cluster of the machine to be adjusted according to the resource requirements of the terminal cluster, the available resources of the cluster of the machine, and the access quality, and the proportion of the machine cluster covering the terminal cluster may include:
  • the available resources of the machine cluster are processed according to the preset resource growth amount corresponding to the machine cluster to obtain the target available resources of the machine cluster;
  • the resource growth amount may be a ratio of resource growth.
  • the method may include: a preset upper limit resource growth amount and a preset absolute upper limit resource growth amount; the preset upper limit resource growth amount is The upper limit of the growth of the cluster resource of the machine cluster, the absolute upper limit resource growth amount is the absolute upper limit of the resource growth of the machine cluster; at this time, the available resources of the machine cluster can be based on the preset upper limit resource growth amount and the preset absolute upper limit resource growth amount.
  • the method of the embodiment can obtain the ratio in the case of the resource increase limit, and then, according to the ratio and the terminal cluster Resource requirements to obtain the sum of resource requirements of the terminal cluster to the machine cluster, and obtain the sum of the resource requirements and the machine cluster
  • the resource gap between the available resources so that the capacity of the device cluster can be expanded in the case of the resource growth limit, and the parameters are provided for the capacity expansion.
  • the system can be expanded based on the capacity expansion requirement, which can improve the accuracy of the expansion. sexuality, improve the utilization of resources.
  • the implementation method may be applicable to the volume reduction assessment.
  • the preset upper limit resource growth ratio or the preset absolute upper limit resource growth ratio may be less than 1 or may be a negative number.
  • the terminal is clustered to obtain a plurality of terminal clusters, and the machines providing the resources are divided to obtain a plurality of machine clusters, and then the resource requirements of the terminal cluster are acquired, and the machine cluster is obtained.
  • the available resources, and the access quality between the terminal cluster and the machine cluster according to the resource requirements of the terminal cluster, the available resources of the machine cluster, and the access quality, obtaining the proportion of the cluster that covers the terminal cluster.
  • Access scheduling is performed on the terminal; the solution can perform access scheduling based on available resources of the cluster, resource requirements of the terminal cluster, and access quality between the terminal cluster and the machine cluster, and the scheduling scheme can also take into account the access. Capacity and access quality, therefore, the access scheduling of the terminal can be optimized, and the accuracy of the access scheduling of the terminal is improved relative to the prior art.
  • the embodiment of the present invention further increases the access quality constraint condition to obtain the ratio, ensures that the overall access quality of the scheduling scheme is optimal, improves resource utilization, and improves service quality after access, thereby improving users.
  • the embodiment of the present invention can also obtain the resource gap between the available resources and the resource requirements, and expand the capacity according to the resource gap, and provide a reference for the capacity expansion, and automatically output the capacity expansion scheme, thereby improving the speed and accuracy of the system expansion. .
  • the embodiment of the present invention can also adjust resource requirements, available resources, and the like according to actual operation requirements (resource usage requirements, manual intervention requirements, capacity expansion requirements, etc.), so that the access scheduling method of the embodiment can satisfy various types.
  • resource usage requirements resource usage requirements, manual intervention requirements, capacity expansion requirements, etc.
  • this method has the advantages of wide application, flexibility, and easy expansion.
  • the access scheduling device is specifically integrated in the scheduling server, and the device providing the resource is taken as a server for detailed description.
  • a terminal access scheduling method As shown in FIG. 2, a terminal access scheduling method, the specific process can be as follows:
  • Step 201 The scheduling server divides the terminal to be scheduled into multiple terminal clusters, and obtains resource requirements of each terminal cluster to obtain a resource requirement set.
  • the resource requirement set includes the resource demand of each terminal cluster.
  • the terminal may be a device in the computer network that is at the outermost periphery of the network, that is, the terminal may be a user device, for example, the terminal may be a personal computer, a mobile phone, a tablet computer, or other user equipment.
  • the scheduling server may perform clustering on the terminal according to the geographic location or the network address of the terminal. For example, the scheduling server performs cluster division on the terminal according to the province where the terminal is located.
  • the resource requirement of the terminal cluster in this embodiment may be obtained according to the resource requirement of the terminal in the terminal cluster, for example, the sum of the resource requirements of the terminals in the cluster; the resource requirements of the terminal may be obtained according to the historical request amount; the new service Can be obtained according to subjective assessment and other means.
  • the scheduling server divides the terminal into m terminal clusters, and then acquires the resource demand quantity u1 of the terminal cluster 1, the resource demand quantity u2 of the terminal cluster 2, the resource demand quantity i of the terminal cluster i, and the resource requirement of the terminal cluster m.
  • resource requirements may be converted into a corresponding set of resource requirements matrix, e.g., a collection of resources u converted into matrix U m, U m may also be referred to as the Resource Vector:
  • U m [u 1 , u 2 , ... u m ], where u i represents the resource demand of the terminal cluster i.
  • the resource requirement of the terminal cluster may be adjusted to meet the personalized requirement of the expansion. Specifically, according to the expected growth of the terminal corresponding to the terminal cluster, The resource demand of the terminal cluster is adjusted; for example, when the expected growth of the terminal is the expected growth rate of the terminal, the resource demand of the terminal cluster may be multiplied by the expected growth ratio corresponding to the cluster.
  • Step 202 The scheduling server divides the server to be scheduled into multiple service clusters, and obtains the available resources of each service cluster to obtain a set of available resources.
  • the set of available resources includes the amount of resources available for each service cluster.
  • the scheduling server may divide the server according to the attributes of the server, for example, The server can be divided according to the area or geographic location to which the server belongs.
  • the amount of available resources of the service cluster can be obtained according to the available resources of the server in the service cluster, for example, the sum of available resources of the server in the cluster, wherein the available resources of the server can be obtained in various ways, for example, by
  • the server performs a stress test to obtain the available resources of the server, and can also analyze the historical operation information of the server to obtain the available resources of the server, and the like.
  • the set of available resources can be converted into a corresponding available resource matrix, for example, the set A is converted into a resource matrix A n , which can also be called a vector A n :
  • a n [a 1 , a 2 , ... a n ], where aj represents the amount of available resources of the service cluster j.
  • the resource usage rate of all the service clusters is about 70%, the resource utilization rate of some service clusters is 80%, and the like;
  • the available resources of the service cluster are adjusted.
  • the available resources of the service cluster can be adjusted according to the upper limit resource usage rate of the service cluster.
  • the available resources can be initialized according to the upper limit resource usage rate, where the upper limit resource usage rate is The resource usage limit of the service cluster.
  • the upper limit available resource amount of j, and the available resource set A is the upper limit available resource set A at this time.
  • the embodiment may add a vector Rn, where rj expresses the upper limit resource usage rate of the service cluster j.
  • the amount of available resources of the service cluster may also be adjusted. Specifically, the amount of available resources of the service cluster may be adjusted according to the resource growth amount corresponding to the service cluster, where The growth of the resource may include an increase in the upper limit resource and an increase in the absolute upper limit resource; the increase may be a growth ratio.
  • Step 203 The scheduling server acquires access quality between the terminal cluster and the service cluster to obtain an access quality set.
  • the access quality set includes: access quality between each terminal cluster and each service cluster.
  • the access quality between the terminal cluster and the service cluster may be the network service quality between the terminal cluster and the service cluster; wherein the network service quality (ie, access quality) between the terminal and the server may be based on the terminal and the service.
  • the network performance parameter, the historical operation information between the terminal and the service, the physical distance between the terminal and the service, the network address of the terminal, and the network topology diagram of the network system to which the terminal belongs and the service are obtained.
  • the network performance parameters may include: network delay, packet loss rate, and delay jitter.
  • the access quality d 11 between the terminal cluster 1 and the service cluster 1 and the access quality d 12 between the terminal cluster 1 and the service cluster 2 can be obtained.
  • the access quality d mn between the terminal cluster m and the service cluster n, at which point the access quality set D (d 11 , d 12 , d 13 ... d 1n , d 21 , d 22 ... d 2n ... ..., d m1 , d m2 ..., d mn ), where d ij represents the access quality between the terminal cluster i and the service cluster j.
  • the quality of the access can be converted into a corresponding set of access mass matrix, e.g., D-converted into a collection of access quality matrix D m ⁇ n, D m ⁇ n which may also be called an access Into the mass vector:
  • the d ij represents the access quality between the terminal cluster i and the service cluster j.
  • the embodiment may also use the logical distance between the terminal cluster and the service cluster to express the access quality.
  • the access quality matrix may be a logical distance matrix.
  • D m ⁇ n may be a logical distance.
  • d ij is a logical distance between the terminal cluster i and the service and its j
  • the logical distance indicating the access quality between the terminal cluster i and the service cluster j
  • the smaller the logical distance the greater the access quality or The better.
  • Step 204 The scheduling server acquires a proportional set according to the access quality set, the resource requirement set, and the available resource set, where the proportional set includes a proportion of the service cluster covering the terminal cluster.
  • the proportion of the service cluster covering the terminal cluster is the percentage of the resource provided by the service cluster for the terminal cluster as a percentage of the resource requirements of the terminal cluster; for example, the service cluster 1 covers 30% of the terminal cluster 1 and indicates that the service cluster 1 is the terminal.
  • the resources provided by cluster 1 account for 30% of the resource requirements of the terminal cluster 1, and both i and j are positive integers greater than one.
  • the resource requirement set according to the access quality set, the resource requirement set, and the available resource set.
  • the total resource requirement of the terminal cluster to the machine cluster is less than the resource limit of the machine cluster, and then the ratio set is obtained according to the constraint condition and the access quality set; That is, the scheduling server obtains the proportional set according to the access quality set, the resource requirement set, and the available resource set, and may include:
  • the scheduling server acquires a corresponding proportional set according to the resource restriction condition and the access quality set;
  • the resource restriction condition includes: the sum of the resource requirements of the terminal cluster to the service cluster is less than or equal to the available resources of the service cluster, and the sum of the resource requirements is obtained according to the resource requirement set and the proportion set to be acquired.
  • the sum of resource requirements of the terminal cluster for the service cluster is the sum of resources that the service cluster needs to provide for all terminal clusters.
  • the weight matrix when using the model matrix, can obtain n N m ⁇ n matrix according to the access mass D m ⁇ n, a set of resource requirements and available resources U m matrix A, wherein the weight matrix elements N m ⁇ n n ij represents the proportion of the service cluster j covering the terminal cluster i; specifically, the weight matrix N m ⁇ can be solved according to the resource restriction condition: U m ⁇ N m ⁇ n ⁇ A n and the access quality matrix D m ⁇ n n to get the corresponding scale set.
  • the sum of the access quality between the terminal cluster and the machine cluster may be the optimal scheduling in this embodiment;
  • the scheduling server obtains the corresponding proportional set according to the resource restriction condition and the access quality set, and may include:
  • the scheduling server acquires a corresponding proportional set according to the resource restriction condition, the quality constraint condition, and the access quality set; wherein the quality constraint condition includes: the sum of the access quality in the access quality set is the largest.
  • the quality constraint may be represented or obtained by a set of access qualities and a set of ratios to be obtained.
  • the access quality matrix D m ⁇ n the resource demand set U m , and the available resource matrix A n are obtained , and then the resource restriction condition is obtained: U m ⁇ N m ⁇ n ⁇ A n ,
  • the maximum quality constraint of the sum of the access qualities may be represented by the access quality matrix D m ⁇ n and the weight matrix N m ⁇ n to be obtained, if The minimum or maximum value indicates that the sum of access quality is the largest, and only needs to be solved when A scale set is obtained by the weight matrix N m ⁇ n at the maximum or minimum.
  • the scheduling server in the embodiment may obtain the corresponding proportional set according to the resource restriction condition, the quality constraint condition, and the access quality set, and may include:
  • the scheduling server acquires a corresponding proportional set according to the resource restriction condition, the priority set, the quality constraint condition, and the access quality set, where the priority set includes the resource acquisition priority of the terminal cluster.
  • the priority set P can be converted into a matrix P m , that is, a priority vector, where p i represents the resource acquisition priority of the terminal cluster i, and at this time, the solution can be solved.
  • the minimum or maximum weight matrix N m ⁇ n to obtain a proportional set.
  • the method of the embodiment can also preferentially ensure the manual adjustment; that is, the method of the embodiment can also increase the proportional constraint condition, the ratio
  • the constraint may include: the proportion of the proportion in the proportional set is a predetermined ratio; the scheduling server may obtain the corresponding proportional set according to the resource restriction condition, the quality constraint condition, and the access quality set, where the scheduling server may include: scheduling the server root A corresponding set of proportions is obtained according to resource constraints, quality constraints, proportional constraints, and access quality sets.
  • the partial value of the solution vector N m ⁇ n can be initialized before the calculation.
  • n 31 40% is set, wherein the n31 can indicate that the proportion of the service cluster 3 covering the terminal cluster is 40%.
  • the embodiment may also adjust the available resource amount according to the upper limit resource usage rate, and if the scheduling service fails to obtain the proportional set, the available resource set according to the absolute upper limit resource usage rate. The amount of available resources in the adjustment is greater than the upper limit resource usage.
  • the available resource amount aj of the service cluster j can be initialized by using the rj to meet the resource usage requirement.
  • aj is the upper limit available resource of the service cluster j, after which the scheduling server can obtain the proportional set N according to the access quality set D, the resource requirement set U, and the available resource set A, if obtained If the proportional set fails, the upper limit available resource amount aj in the available resource set A may be initialized according to the absolute upper limit resource usage rate rj', and the initialized aj is the absolute upper limit available resource amount, specifically, may be made After the initialization of all the upper limit available resources in the available resource set A is completed, the scheduling server may acquire the proportional set N again according to the access quality set D, the resource requirement set U, and the available resource set A.
  • Step 205 The scheduling server allocates a corresponding target service cluster to the terminal cluster according to the proportional set.
  • the terminal cluster may be allocated a corresponding target service cluster by using a preset algorithm and the proportional set.
  • Step 206 The scheduling server allocates a target server that needs to be accessed to the terminal in the terminal cluster in the target service cluster.
  • the target server may be allocated to the terminal according to the available resources of the server in the target service cluster, or the geographical location.
  • the scheduling server may be a separate entity, or may include multiple entities, for example, may include multiple service devices.
  • the scheduling server acquires a proportional set according to the access quality set, the resource requirement set, and the available resource set, and then the scheduling server allocates a corresponding target service cluster to the terminal cluster according to the proportional set, where the scheduling server is configured.
  • the target service cluster allocates the target server to be accessed to the terminal in the terminal cluster; the solution can be accessed based on the available resources of the service cluster, the resource requirements of the terminal cluster, and the access quality between the terminal cluster and the service cluster.
  • Scheduling it can be seen that the scheduling scheme takes into account the access capacity and the access quality. Therefore, the access scheduling of the terminal can be optimized, and the accuracy of the access scheduling of the terminal is improved compared with the prior art.
  • the embodiment of the present invention can calculate the weight matrix through the matrix vector model, thereby quickly obtaining the proportion of the service cluster covering the terminal cluster, and improving the efficiency of the access scheduling.
  • the scheduling scheme provided by the embodiment of the present invention may also be optimal scheduling when the default access quality is the best, which may improve the utilization of system resources and improve the service quality after access; and the scheduling scheme also
  • the resource requirements, available resources, and the like can be adjusted according to the actual operation requirements (resource usage requirements, manual intervention requirements, capacity expansion requirements, etc.), so that the access scheduling method of the embodiment can meet various personalized requirements, and the method has Wide application, flexibility, and easy to expand.
  • the embodiment of the present invention further provides an access scheduling apparatus for a terminal.
  • the access scheduling apparatus may further include a first dividing unit 301, a second dividing unit 302, and a An obtaining unit 303, a second obtaining unit 304, and a scheduling unit 305 are as follows:
  • the first dividing unit 301 is configured to perform clustering on the terminal to obtain a plurality of terminal clusters.
  • the first dividing unit 301 can be configured to perform clustering on the terminal according to the geographic location of the terminal.
  • the geographic location of the terminal may include the current geographic location of the terminal, or the network registration attribution of the terminal; for example, the cluster may be divided according to the current location of the terminal or the area (region, city, province, country, etc.) registered by the network, specifically, The same terminal in the area where the current network or the network is registered is classified into the same terminal cluster.
  • the second dividing unit 302 is configured to perform clustering on the machines providing resources to obtain a plurality of machine clusters.
  • the second partitioning unit 302 can be configured to cluster the machine according to the geographic location of the machine providing the resources.
  • the cluster can be divided according to the region (region, city, province, country, etc.) where the machine is located, for example, a machine located in a certain province is divided into a machine cluster.
  • the first obtaining unit 303 is configured to acquire resource requirements of the terminal cluster, available resources of the machine cluster, and access quality between the terminal cluster and the machine cluster.
  • the method for obtaining the resource requirements of the terminal cluster may be multiple.
  • the first acquisition unit 303 may be configured to:
  • the available resources of the machine cluster may be obtained in various manners, for example, by using a stress test method, specifically, performing stress tests on the machines in the machine cluster to obtain available resources of the machine. Then, the available resources of the machine cluster are obtained based on the available resources of the machines in the machine cluster. In order to improve the acquisition speed and improve the efficiency of access scheduling in actual testing, the pressure can be measured based on the type of machine, and then the available resources of the machine cluster can be obtained by combining the number of different types of machines.
  • the access quality between the terminal cluster and the machine cluster may be the network service quality between the terminal cluster and the machine cluster; wherein the network service quality (ie, access quality) between the terminal and the machine may be based on the terminal. Obtained from at least one of a network performance parameter between the machine, historical operation information between the terminal and the machine, a physical distance between the terminal and the machine, a network address of the terminal, and a network topology diagram of the network system of the terminal and the machine to which the machine belongs.
  • the network performance parameters may include: network delay, packet loss rate, and delay jitter.
  • the second obtaining unit 304 is configured to obtain, according to resource requirements of the terminal cluster, available resources of the machine cluster, and the access quality, a proportion of the cluster accessing the terminal cluster.
  • the second obtaining unit 304 may be configured to: obtain, according to a preset restriction condition and an access quality between the terminal cluster and the machine cluster, a proportion of the cluster of the system covering the terminal cluster;
  • the preset restriction condition includes: the sum of the resource requirements of the terminal cluster for the machine cluster is less than or equal to the available resources of the machine cluster; wherein the sum of the resource requirements is obtained according to the resource requirements of the terminal cluster and the proportion to be acquired.
  • the second obtaining unit 304 may be configured to: acquire a proportional set according to a preset constraint condition, an access quality set, and the preset restriction condition, where the access quality set includes multiple the terminal cluster and the machine cluster Inter-access quality, the scale set includes the machine cluster coverage The proportion of the terminal cluster;
  • the preset constraint includes: the sum of the access qualities in the access quality set is the largest.
  • the second obtaining unit 304 may include: a matrix creation subunit and a ratio acquisition subunit;
  • the matrix creation subunit is configured to create a corresponding quality matrix according to the access quality set, and the matrix element in the quality matrix represents the access quality between the terminal cluster and the machine cluster;
  • the ratio acquisition subunit is configured to obtain a corresponding weight matrix according to the preset constraint condition, the quality matrix, and the preset constraint condition, where a matrix element in the weight matrix indicates a proportion of the machine cluster covering the terminal cluster;
  • the scheduling unit is specifically configured to perform access scheduling on the terminal according to the weight matrix.
  • the second obtaining unit 304 may include: a processing subunit and a proportional acquisition subunit;
  • the processing subunit is configured to process the available resources of the machine cluster according to the preset upper limit resource usage rate of the machine cluster to obtain an upper limit available resource of the machine cluster;
  • the ratio acquisition subunit is configured to obtain, according to the resource requirement of the terminal cluster, the upper limit available resources of the machine cluster, and the access quality, the proportion of the cluster that covers the terminal cluster.
  • the processing sub-unit is further configured to: when the proportion of the machine cluster accessing the terminal cluster fails, the upper limit available resources of the machine cluster are processed according to a preset absolute upper limit resource usage rate to obtain the machine cluster.
  • An absolute upper limit available resource wherein the preset absolute upper limit resource usage rate is greater than a preset upper limit resource usage rate;
  • the ratio acquisition sub-unit is further configured to obtain, according to resource requirements of the terminal cluster, an absolute upper limit available resource of the machine cluster, and the access quality, a proportion of the cluster of the terminal that covers the terminal cluster.
  • the second acquiring unit in this embodiment may be configured as:
  • the access scheduling apparatus of the embodiment further includes: a third obtaining unit 306, a resource gap obtaining unit 307, and a expansion unit 308;
  • the third obtaining unit 306 is configured to obtain, according to the ratio and the resource requirement of the terminal cluster, a sum of resource requirements of the terminal cluster for the machine cluster;
  • the resource gap obtaining unit 307 is configured to acquire a resource gap between the sum of the resource requirements and the available resources of the machine cluster;
  • the expansion unit 308 is configured to expand the system to which the machine cluster belongs according to the resource gap.
  • the second obtaining unit 304 may be configured to:
  • the resource requirements of the terminal cluster are processed according to the expected growth amount of the terminal corresponding to the terminal cluster, to obtain an expected resource requirement of the terminal cluster;
  • the proportion of the cluster that covers the terminal cluster is obtained.
  • the second obtaining unit 304 can be configured to:
  • the available resources of the machine cluster are processed according to the preset resource growth amount corresponding to the machine cluster to obtain the target available resources of the machine cluster;
  • the scheduling unit 305 is configured to compare the proportional set or the weight matrix according to the ratio.
  • the scheduling unit 305 may be configured to be based on a proportional set or a weight matrix scale set or a weight matrix.
  • the scheduling unit 305 can be configured to:
  • a target machine that the terminal needs to access in the terminal cluster is determined from the target machine cluster.
  • the preset algorithm may be used to calculate the ratio to obtain the operation result, and the target machine cluster accessed by the terminal cluster is determined according to the operation result; wherein the preset algorithm may be set according to actual requirements, such as the preset algorithm.
  • the foregoing units may be implemented as a separate entity, or may be implemented in any combination, and may be implemented as the same or a plurality of entities.
  • the foregoing method embodiments and details are not described herein.
  • the access scheduling device may be specifically integrated in a server or other device that needs to perform access scheduling.
  • the access scheduling apparatus in this embodiment uses the first dividing unit 301 to perform cluster partitioning on the terminal to obtain a plurality of terminal clusters, and the second dividing unit 302 divides the devices that provide resources to obtain multiple machine clusters. Then, the first acquisition unit 303 acquires the resource requirement of the terminal cluster, the available resources of the machine cluster, and the access quality between the terminal cluster and the machine cluster, and the second obtaining unit 304 is configured according to the terminal cluster.
  • the resource requirement, the available resources of the cluster of the machine, and the access quality obtain the proportion of the cluster that covers the terminal cluster, and the scheduling unit 305 performs access scheduling on the terminal according to the ratio; the solution may be based on available resources of the cluster
  • the resource requirements of the terminal cluster and the access quality between the terminal cluster and the machine cluster are used for access scheduling. It can be seen that the scheduling scheme takes into account both the access capacity and the access quality. Therefore, the access scheduling of the terminal can be optimized. In the prior art, the accuracy of the access scheduling of the terminal is improved.
  • the disclosed method and apparatus can be implemented in other ways.
  • the device embodiments described above are merely illustrative.
  • the division of the modules is only a logical function division.
  • there may be another division manner for example, multiple modules or components may be combined, or Can be integrated into another system, or some features can be ignored or not executed.
  • the communication connections between the various components shown or discussed may be indirect coupling or communication connections through some interfaces, devices or modules, and may be electrical, mechanical or otherwise.
  • the modules described above as separate components may or may not be physically separated.
  • the components displayed as modules may or may not be physical modules, that is, may be located in one place or distributed to multiple network modules; Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may be separately used as one module, or two or more modules may be integrated into one module;
  • the module can be implemented in the form of hardware or in the form of hardware plus software function modules.
  • the foregoing program may be stored in a computer readable storage medium, and when executed, the program includes The foregoing steps of the method embodiment; and the foregoing storage medium includes: a removable storage device, a read-only memory (ROM), a magnetic disk or an optical disk, and the like, which can store program codes.
  • ROM read-only memory
  • the above-described integrated module of the embodiment of the present invention may be stored in a computer readable storage medium if it is implemented in the form of a software function module and sold or used as a stand-alone product.
  • the technical solution of the embodiments of the present invention may be embodied in the form of a software product in essence or in the form of a software product stored in a storage medium, including a plurality of instructions. Enabling a computer device (which may be a personal computer, server, or network device, etc.) to perform the methods of the various embodiments of the present invention All or part.
  • the foregoing storage medium includes various media that can store program codes, such as a mobile storage device, a ROM, a magnetic disk, or an optical disk.
  • the technical solution of the embodiment of the present invention performs access scheduling based on the available resources of the machine cluster, the resource requirements of the terminal cluster, and the access quality between the terminal cluster and the machine cluster. It can be seen that the scheduling scheme takes into account both access capacity and access. Quality, therefore, the access scheduling of the terminal can be optimized, and the accuracy of the access scheduling of the terminal is improved relative to the prior art.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Multimedia (AREA)
  • Quality & Reliability (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本发明实施例公开了一种终端的接入调度方法、装置及计算机存储介质；所述方法包括：对终端进行集群划分，以得到多个终端集群；对提供资源的机器进行划分，以得到多个机器集群；获取该终端集群的资源需求、该机器集群的可用资源、以及该终端集群与该机器集群之间的接入质量，根据该终端集群的资源需求、该机器集群的可用资源以及该接入质量，获取该机器集群覆盖该终端集群的比例，根据该比例对该终端进行接入调度。

Description

一种终端的接入调度方法、装置及计算机存储介质
相关申请的交叉引用
本申请基于申请号为201610301528.1、申请日为2016年05月09日的中国专利申请提出，并要求该中国专利申请的优先权，该中国专利申请的全部内容在此引入本申请作为参考。
技术领域
本发明涉及通信技术领域，具体涉及一种终端的接入调度方法、装置及计算机存储介质。
背景技术
对于客户机/服务器(C/S)架构的通信***，终端的接入调度是及其重要的一个环节，该接入调度为指定由哪些服务器接入某个终端并为该终端提供服务的控制逻辑。
在现有技术中，常用的接入调度方案一般是基于位置来调度，具体地，获取终端的地理位置信息(可以通过IP获取，也可以直接获取)，然后，根据该地理位置信息分配相应的机器(如服务器)为该终端提供服务，也即使得相应的机器接入该终端，以为该终端提供服务；比如，在获取终端的地理位置信息之后，分配与终端地理位置相近或者最近的机器为该终端提供服务，此时的接入调度可以称为就近接入调度；该方案如果获取的是终端的IP，则需要在预设IP库中查找与该IP对应的地理位置，然后，分配与终端地理位置相近或者最近的机器为该终端提供服务。
在对现有技术的研究和实践过程中，本发明的发明人发现，现有的接入调度方案仅仅基于终端的地理位置进行接入调度，并且该方案中获取的 IP库、IP或者地理位置信息可能会不准确，因此，会导致终端的接入调度的精确性较低，尤其是在面临海量服务时。
发明内容
本发明实施例提供一种终端的接入调度方法、装置及计算机存储介质，可以提高终端的接入调度的精确性。
本发明实施例提供一种终端的接入调度方法，包括：
对终端进行集群划分，以得到多个终端集群；
对提供资源的机器进行集群划分，以得到多个机器集群；
获取所述终端集群的资源需求、所述机器集群的可用资源、以及所述终端集群与所述机器集群之间的接入质量；
根据所述终端集群的资源需求、所述机器集群的可用资源以及所述接入质量，获取所述机器集群覆盖所述终端集群的比例；
根据所述比例对所述终端进行接入调度。
本发明实施例还提供一种终端的接入调度装置，包括：
第一划分单元，配置为对终端进行集群划分，以得到多个终端集群；
第二划分单元，配置为对提供资源的机器进行集群划分，以得到多个机器集群；
第一获取单元，配置为获取所述终端集群的资源需求、所述机器集群的可用资源、以及所述终端集群与所述机器集群之间的接入质量；
第二获取单元，配置为根据所述终端集群的资源需求、所述机器集群的可用资源以及所述接入质量，获取所述机器集群覆盖所述终端集群的比例；
调度单元，配置为根据所述比例对所述终端进行接入调度。
本发明实施例还提供了一种计算机存储介质，所述计算机存储介质中存储有计算机可执行指令，所述计算机可执行指令用于执行本发明实施例 所述的接入调度方法。
本发明实施例采用对终端进行集群划分，以得到多个终端集群，对提供资源的机器进行划分，以得到多个机器集群，然后，获取该终端集群的资源需求、该机器集群的可用资源、以及该终端集群与该机器集群之间的接入质量，根据该终端集群的资源需求、该机器集群的可用资源以及该接入质量，获取该机器集群覆盖该终端集群的比例，根据该比例对该终端进行接入调度；该方案可以基于机器集群的可用资源、终端集群的资源需求以及终端集群与机器集群之间的接入质量来进行接入调度，可见该调度方案同时兼顾了接入容量和接入质量，因此，可以优化终端的接入调度，相对于现有技术而言，提高了终端的接入调度的精确性。
附图说明
为了更清楚地说明本发明实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是本发明实施例一提供的一种终端的接入调度方法的流程图；
图2是本发明实施例二提供的一种终端的接入调度方法的流程图；
图3a是本发明实施例三提供的一种终端的接入调度装置的结构示意图；
图3b是本发明实施例三提供的另一种终端的接入调度装置的结构示意图。
具体实施方式
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例， 而不是全部的实施例。基于本发明中的实施例，本领域技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
本发明实施例提供一种终端的接入调度方法和装置。以下将分别进行详细说明。
本实施例将从接入调度装置的角度进行描述，该接入调度装置具体可以集成在服务器或其他需要进行接入调度的设备中。
一种终端的接入调度方法，包括：对终端进行集群划分，以得到多个终端集群，对提供资源的机器进行划分，以得到多个机器集群，然后，获取该终端集群的资源需求、该机器集群的可用资源、以及该终端集群与该机器集群之间的接入质量，根据该终端集群的资源需求、该机器集群的可用资源以及该接入质量，获取该机器集群覆盖该终端集群的比例，根据该比例对该终端进行接入调度。
如图1所示，该终端的接入调度方法的具体流程可以如下：
步骤101、对终端进行集群划分，以得到多个终端集群。
其中，对终端的划分方式可以有多种，比如，可以基于终端的地理位置进行划分，也即所述对终端进行集群划分可以包括：
根据终端的地理位置，对所述终端进行集群划分。
其中，终端的地理位置可以包括终端当前的地理位置，或者终端的网络注册归属地；例如，可以按照终端当前所在或者网络注册的区域(地区、市、省份、国家等)进行集群划分，具体地，将当前所在或者网络注册的区域相同的终端归为同一个终端集群。
又比如，本实施例还可以基于终端的网络地址(如IP地址)进行集群划分，也即所述对终端进行集群划分可以包括：
根据终端的网络地址，对该终端进行集群划分。
例如，可以按照IP地址段来划分，也即所述根据终端的网络地址，对该 终端进行集群划分可以包括：根据终端的IP地址所属的IP地址段，对该终端进行集群划分，比如可以将终端的IP地址属于同一预设IP地址段划分成一个终端集群等等。
为适用于各种场景的调度，以满足个性化的需求，本实施例中终端集群中成员可以为单个终端，也可以为终端组合，具体地可以根据实际需求设定。
在本实施例中，终端可以为计算机网络中处于网络最***的设备，也即终端可以为用户设备，比如终端可以为个人计算机、手机、平板电脑、或者其他用户设备。
步骤102、对提供资源的机器进行集群划分，以得到多个机器集群。
其中，提供资源的机器可以为服务器或者其他可以提供资源的设备。
本实施例中，对机器进行划分的方式可以有多种，比如，可以根据机器的地理位置进行划分，也即所述对提供资源的机器进行集群划分可以包括：
根据提供资源的机器的地理位置，对该机器进行集群划分。
具体地，可以根据机器所在的区域(地区、市、省份、国家等)进行集群划分，比如，将位于某个省份的机器划分成一个机器集群。
本实施例中，步骤101和步骤102之间的时序不受限制，比如，步骤101可以在步骤102之前，也可以是步骤102在步骤101之前，当然，也可以同时执行步骤101和步骤102等等。
步骤103、获取该终端集群的资源需求、该机器集群的可用资源、以及该终端集群与该机器集群之间的接入质量。
其中，获取终端集群的资源需求的方式可以有多种，比如，可以基于终端集群的历史请求量获取，也即所述获取终端集群的资源需求可以包括：
获取终端集群中终端的历史请求量；
根据该历史请求量获取该终端集群中终端的资源需求；
根据终端集群中终端的资源需求，获取该终端集群的资源需求。
例如，对于10个终端的集群，可以分别获取每个终端的历史请求量，然后，对该每个终端的历史请求量分析，以得到每个终端的资源需求，将这10个终端各自的资源需求进行相加，以得到该集群的资源需求。
又比如，在本实施例还可以采用主管评估等方式来获取终端集群的资源需求，也即所述获取该终端集群的资源需求可以包括：接收用户输入的终端集群的资源需求。
同样，本实施例中获取机器集群的可用资源的方式也可以有多种，比如，可以通过压力测试方式得到，具体地，对机器集群中的机器进行压力测试，以得到该机器的可用资源，然后，根据机器集群中机器的可用资源得到机器集群的可用资源。在实际测试时为了提高获取速度，提高接入调度的效率可以基于机器的类型进行压测，然后结合不同类型的机器的数量得到机器集群的可用资源；也即获取机器集群的可用资源的过程可以包括：
对该机器集群中不同类型的机器进行压测，以得到不同类型的机器的可用资源；
根据该不同类型的机器的可用资源、以及不同类型的机器的数量得到该机器集群的可用资源。
例如，当机器集群包括：3个机器b、4个机器c、5个机器f时，本实施例可以先对一个机器b、机器c、机器f进行压测，以得到机器b的可用资源量b1、机器c的可用资源量c1、机器f的可用资源量f1，然后，分别将可用资源量b1乘以机器b的数量3、可用资源量c1乘以机器c的数量4，可用资源量f1乘以机器f的数量5，最后，将这些乘积相加即得到该机器集群的可用资源量，也即机器集群的可用资源量＝b1×3+c1×4+f1×5。
在本实施例中，不仅可用采用压力测试的方式来获取可用资源量，还 可以基于机器的历史运营信息来获取该可用资源量；也即获取机器集群的可用资源的过程可以包括：
根据机器集群中机器的历史运营信息获取该机器的可用资源；
根据机器集群中机器的可用资源获取该机器集群的可用资源。
其中，该历史运营信息可以包括：历史资源消耗信息、历史业务处理量中的至少一种。
在实际应用中，为了提高资源获取速度，提高接入调度的效率，本实施例可以基于机器的类型和数量获取机器集群的可用资源；包括：
根据机器集群中不同类型的机器的历史运营信息获取不同类型机器的可用资源；
根据不同类型机器的可用资源以及不同类型机器的数量获取该机器集群的可用资源。
本实施例中，终端集群与机器集群之间的接入质量可以为终端集群到机器集群之间的网络服务质量；其中，终端与机器之间的网络服务质量(即接入质量)可以根据终端与机器之间的网络性能参数、终端与机器之间的历史运营信息、终端与机器之间的物理距离、终端的网络地址、终端和机器所属网络***的网络拓扑图中的至少一种得到。其中，网络性能参数可以包括：网络延时、丢包率和时延抖动等等。
步骤104、根据该终端集群的资源需求、该机器集群的可用资源以及该接入质量，获取该机器集群接入该终端集群的比例。
其中，机器集群覆盖终端集群的比例为该机器集群为该终端集群提供的资源占该终端集群的资源需求的百分比；比如，机器集群1覆盖30％的终端集群1，其表示机器集群1为终端集群1提供的资源占该终端集群1的资源需求的30％；机器集群1覆盖50％的终端集群2，其表示机器集群1为终端集群2提供的资源占该终端集群2的资源需求的50％，该比例可以采用nij表示， 该nij表示机器集群j覆盖终端集群i的比例，该i和j均为大于1的正整数。
在本实施例中，根据资源需求、可用资源以及接入质量获取机器集群覆盖终端集群的比例的方式可以有多种，比如，可以以终端集群对机器集群的资源需求总和小于该机器集群的可用资源为限制条件，然后，根据该限制条件和终端集群与机器集群之间的接入质量来获取机器集群覆盖终端集群的比例；也即所述根据该终端集群的资源需求、该机器集群的可用资源以及该接入质量，获取该机器集群覆盖该终端集群的比例可以包括：
根据预设限制条件和该终端集群与该机器集群之间的接入质量，获取该机器集群覆盖该终端集群的比例；
其中，该预设限制条件包括：该终端集群对该机器集群的资源需求总和小于或等于该机器集群的可用资源；其中，该资源需求总和根据该终端集群的资源需求和欲获取的比例获得。
本实施例中，终端集群对机器集群的资源需求总和为机器集群需要为其所有终端集群提供的资源之和；比如，对于机器集群1，假如有m个终端集群，那么终端集群对该机器集群1的资源需求总和可以为机器集群1需要为终端集群1提供的资源、为终端集群2提供的资源……为终端集群m提供的资源之和；该终端集群对机器集群的资源需求总和可以根据终端集群的资源需求以及机器集群覆盖终端集群的比例得到；比如，终端集群对该机器集群1的资源需求总和可以根据终端集群1的资源需求u1和机器集群1覆盖终端集群1的比例n11、终端集群2的资源需求u2和机器集群1覆盖终端集群2的比例n21……终端集群m的资源需求um和机器集群1覆盖终端集群m的比例nm1得到，具体地该终端集群对该机器a1的资源需求总和＝u1×n11+u2×n21+……+um×nm1，其中，m和n均为大于1的正整数。
具体地，可以获取资源需求集合、可用资源集合以及接入质量集合，其中，该资源需求集合包括多个终端集群的资源需求，该可用资源集合包 括多个机器集群的可用资源，该接入质量集合包括多个终端集群与机器集群之间的接入质量，然后，根据预设限制条件和该接入质量集合获取比例集合，该比例集合包括多个机器集群覆盖该终端集群的比例；其中，该预设限制条件包括：终端集群对该机器集群的资源需求总和小于该机器集群的可用资源。
例如，假设有m个终端集群，n个机器集群；可以获取每个终端集群的资源需求u，以得到资源需求集合U，该U包括：u1、u2、u3……ui……um，其中ui表示终端集群i的资源需求大小，该i小于m；获取每个机器集群的可用资源a，以得到可用资源集合A，该A包括：a1、a2、a3……aj……an，其中，aj表示机器集群j的可用资源大小，其中j小于n；获取每个终端集群与机器集群之间的接入质量d，以得到接入质量集合D，该D包括：d11、d12、d13……d1n、d21、d22……d2n……、dm1、dm2……、dmn，其中，dij表示终端集群i与机器集群j之间的接入质量；假设欲求比例为nij，该nij表示机器集群j覆盖终端集群i的比例，此时，预设限制条件包括：Qi≤ai，其中Qi表示终端集群对机器集群i的资源需求总和，该Qi＝u1×n1i+u2×n2i+……+um×nmi；然后，可以根据预设限制条件和接入质量集合D获取nij，以得到比例集合N，该N包括：n11、n12、n13……n1n、n21、n22……n2n……、nm1、nm2……、nmn。
作为一种实施方式，为了能够提升***的资源利用率以及业务服务质量，本实施例可以获取在接入质量集合中接入质量总和最大情况下的比例，也即默认使用终端集群与机器集群之间的接入质量总和最大时为最优调度；此时所述根据预设限制条件和该终端集群与该机器集群之间的接入质量，获取该机器集群覆盖该终端集群的比例可以包括：
根据预设约束条件、接入质量集合以及该预设限制条件获取比例集合，该接入质量集合包括多个该终端集群与该机器集群之间的接入质量，该比例集合包括该机器集群覆盖该终端集群的比例；
其中，该预设约束条件包括：该接入质量集合中接入质量的总和最大。
比如，在接入质量集合D中各接入质量dij的总和为最大时，根据预设限制条件和接入质量集合D获取nij，以得到比例集合N。
在实际应用中，为了方便快捷地获取机器集群覆盖终端的比例，可以采用矩阵运算的方式来计算比例集合，也即将获取机器集群覆盖终端的比例集合转化成求解矩阵的问题；比如，可以根据接入质量集合创建相应的矩阵，然后，根据预设约束条件、该矩阵和预设限制条件获取相应的权值矩阵，该权值矩阵中的矩阵元素表示该机器集群覆盖该终端集群的比例；也即所述根据预设约束条件、接入质量集合以及该预设限制条件获取比例集合可以包括：
根据该接入质量集合创建相应的质量矩阵，该质量矩阵中的矩阵元素表示该终端集群与该机器集群之间的接入质量；
在该预设约束条件下，根据该质量矩阵以及该预设限制条件获取相应的权值矩阵，该权值矩阵中的矩阵元素表示该机器集群覆盖该终端集群的比例；
此时以下步骤105所述的根据该比例对该终端进行接入调度可以包括：根据该权值矩阵对该终端进行接入调度。
例如，可以根据资源需求集合U创建相应的资源需求矩阵Um，也称为用户向量Um，(ui表示终端集群i的资源需求大小)：
Um＝[u1,u2,……um]；
根据可用资源集合A创建相应的资源需求矩阵An,也称为资源向量An，(aj表示机群j的可用资源大小)：
An＝[a1,a2,……an]；
根据接入质量集合创建相应的接入质量矩阵Dm×n，也可称为接入质量向量Dm×n(dij表示终端集群i与机器集群j之间的接入质量)：
Figure PCTCN2017082938-appb-000001
其中，dij可以为终端i与机器集群j之间的逻辑距离，该逻辑距离表示终端集群i与机器集群j之间的接入质量，逻辑距离越大表示接入质量越大(即越好)，逻辑距离越小表示接入质量越小(即越差)。
假设，欲求解的权值矩阵为Nm×n，其中，nij表示机器集群j覆盖终端集群i的比例，为保证分配的资源不超过可用资源，可以增加资源限制条件，该资源限制条件包括：对于每个机器集群，终端集群对该机器集群的资源需求总和小于或等于该机器集群的可用资源，也即：Um×Nm×n≤An。
为了能够提升业务服务质量，保证接入调度的合理性，本实施例可以使用终端集群与机器集群之间的接入质量总和最大为最优调度，比如在dij可以为终端i与机器集群j之间的逻辑距离时使用逻辑距离总和最小为最优调度，也即在计算Nm×n增加终端集群与机器集群之间的接入质量总和最大的约束条件，其中，该终端集群与机器集群之间的接入质量总和最大可以由接入质量矩阵Dm×n和欲获取的权值矩阵Nm×n表示，如可以通过
Figure PCTCN2017082938-appb-000002
最小或者最大时表示接入质量的总和最大；具体地，当dij越大表示接入质量越大(即越好)时，该预设约束条件可以为
Figure PCTCN2017082938-appb-000003
最大，当dij越大表示接入质量越小(即越差)时，该预设约束条件可以为
Figure PCTCN2017082938-appb-000004
最小。
在获取接入质量向量Dm×n，资源限制条件Um×Nm×n≤An，以及约束条件即
Figure PCTCN2017082938-appb-000005
最大或者最小之后，就可以求解当
Figure PCTCN2017082938-appb-000006
最大或者最小时的权值矩阵Nm×n，也即得到机器集群覆盖终端集群的比例。
步骤105、根据该比例对该终端进行接入调度。
本实施例中，在步骤104获取比例集合的情况下，可以根据比例集合对终端进行接入调度。
具体地，所述根据该比例对该终端进行接入调度可以包括：
根据该比例确定该终端集群接入的目标机器集群；
从该目标机器集群中确定该终端集群中终端需要接入的目标机器。
例如，可以根据机器集群1覆盖终端集群1的比例n11、机器集群2覆盖终端集群1的比例n12……机器集群n覆盖终端集群1的比例n1n，来获取终端集群1接入的目标机器集群。
本实施例中，可以采用预设算法对比例进行运算，以得到运算结果，根据该运算结果确定终端集群接入的目标机器集群；其中，该预设算法可以根据实际需求设定，比如该预设算法可以为加权算法等等。
作为一种实施方式，为了能够满足个性化的运营需求，本实施例还可以对上述资源需求、可用资源以及接入质量等进行调整，具体地，为减少机器集群的压力，提高业务服务质量，一般希望对资源有保留，即不希望机器集群的资源使用率为100％；又比如，为防止资源浪费，提升资源利用率，希望某些机器集群的资源使用率可以更高点，如某个海外机器集群，仅用60％太浪费，希望能用到90％；此时，本实施例可以利用资源使用率对机器集群的可用资源进行处理，也即所述根据该终端集群的资源需求、该机器集群的可用资源以及该接入质量，获取该机器集群覆盖该终端集群的比例可以包括：
根据该机器集群的预设上限资源使用率对该机器集群的可用资源进行处理，以得到该机器集群的上限可用资源；
根据该终端集群的资源需求、该机器集群的上限可用资源以及该接入质量，获取该机器集群覆盖该终端集群的比例。
其中，预设上限资源使用率表示机器集群的资源使用率上限。
比如，可以将机器集群的可用资源乘以该机器集群的预设上限资源使用率，得到该机器集群的上限可用资源，假设机器集群n的预设上限资源使用率为rn，此时该机器集群n的上限可用资源an'＝an×rn。
又比如，在通过矩阵运算求解比例时，可以增加资源使用率向量Rn：
Rn＝[r1,r2,……rn]，其中，rj为机器集群j的预设上限资源使用率。
在求解权值矩阵之前，可以对An进行预处理，具体地将An中每个机器集群的可用资源乘以相应的预设上限资源使用率，以得到上限可以资源也即an'＝an×rn，之后，将an替换成an’再计算权值矩阵。
其中，根据资源需求、上限可用资源以及接入质量获取比例的过程与上述步骤104类似，区别仅在于一个是可用资源，一个是上限可用资源，将上述机器集群的可用资源替换为上限可用资源。
在本实施例中，设置了资源使用率上限，也即严格限制了可用资源的大小，因此，可能会存在比例获取失败(即资源分配无解，如权值矩阵无解)或者调度精确性变差的问题，例如某个大机群，稳定运营需求是整体资源消耗在60％左右，但当对这个点的需求量增加时，需要的是该点的承载量临时超过资源的60％，而不是由于资源不足导致分配无解。
为能够克服上述问题，并充分地利用机器集群的资源完成调度，本实施例方法还可以增加机器集群的预设绝对上限资源使用率，并通过该使用率对机器集群的上限可用资源进行处理，然后，基于该处理后的上限可可用资源、接入质量以及资源需求来获取比例；也即本实施例方法还可以包括：
在获取该机器集群接入该终端集群的比例失败时，根据预设绝对上限资源使用率对该机器集群的上限可用资源进行处理，以得到该机器集群的绝对上限可用资源，其中，该预设绝对上限资源使用率大于预设上限资源 使用率；
根据该终端集群的资源需求、该机器集群的绝对上限可用资源以及该接入质量，获取该机器集群覆盖该终端集群的比例。
该预设绝对上限资源使用率为资源使用率的绝对上限，其大于资源使用率的上限。
比如，可以将机器集群的上限可用资源乘以该机器集群的预设绝对上限资源使用率然后除以预设上限资源，以得到该机器集群的绝对上限可用资源，也即将机器集群的可用资源乘以该机器集群的预设绝对上限资源，假设机器集群n的预设上限资源使用率为rn，预设绝对上限资源使用率为rn’,在获取比例失败时，可用通过rn’对机器集群的上限可用资源进行处理，如该机器集群n的绝对上限可用资源
Figure PCTCN2017082938-appb-000007
之后再根据机器集群的绝对上限可用资源、接入质量以及资源需求重新获取比例。
其中，获取比例失败可以有多种情况，比如，在获取比例的过程中，发现满足不了预设限制条件导致获取比例失败，即终端集群对机器集群的资源需求总和大于该机器集群的上限可用资源，此时，可以利用机器集群的预设绝对上限资源使用率对上述机器集群的上限可用资源进行处理得到机器集群的绝对上限可用资源，然后，根据该绝对上限可用资源、资源需求以及接入质量重新获取比例。
例如，在通过矩阵运算求解比例时，可以增加资源使用率向量Rn’,其中r′j表示机器集群j的预设绝对上限资源使用率；
当在求解权值矩阵时，发现对于机器集群j，
Figure PCTCN2017082938-appb-000008
无法满足时，即该机器集群j的上限可以资源满足终端的资源需求时，则可以采用rj’对aj’进行处理，比如，通过公式：
Figure PCTCN2017082938-appb-000009
计算出机器集群j的绝对上限可用资 源aj”，然后，再行求解权值矩阵。
其中，(r′j-rj)为机器集群j的扩容需求，本实施例可以记录该扩容需求，以便后续基于该扩容需求进行扩容。
本实施例中，根据资源需求、绝对上限可用资源以及接入质量获取比例的过程与上述步骤104类似，区别仅在于一个是可用资源，一个是绝对上限可用资源，可以将上述机器集群的可用资源替换为上限可用资源来获取比例。
作为一种实施方式，为保证某些终端集群可以优选获取资源，比如，在根据省份对终端进行机器划分时，希望广东的终端相比于广西优先获得资源；此时，本实施例可以引入资源获取优先级，可以为每个终端集群设置相应的资源获取优先级，在接入调度时可以将该资源获取优先级也考虑进去，也即所述根据预设限制条件和该终端集群与该机器集群之间的接入质量，获取该机器集群接入该终端集群的比例可以包括：
根据该终端集群的资源获取优先级、预设限制条件和该终端集群与该机器集群之间的接入质量，获取该机器集群覆盖该终端集群。
本实施例中，所述根据该终端集群的资源获取优先级、预设限制条件和该终端集群与该机器集群之间的接入质量，获取该机器集群覆盖该终端集群可以包括：
根据预设约束条件、接入质量集合、该优先级集合以及该预设限制条件获取比例集合，该接入质量集合包括多个该终端集群与该机器集群之间的接入质量，该比例集合包括该机器集群覆盖该终端集群的比例，该优先级集合包括多个机器集群的资源获取优先级。
其中，机器集群的资源获取优先级可以根据机器集群的属性获取，具体地，可以根据机器的地理位置获取，比如，广东机器集群的资源获取优先级高于江西机器集群的资源获取优先级。
在采用矩阵计算比例的时，所述根据预设约束条件、接入质量集合、该优先级集合以及该预设限制条件获取比例集合可以包括：
根据该接入质量集合创建相应的质量矩阵，该质量矩阵中的矩阵元素表示该终端集群与该机器集群之间的接入质量；
根据该优先级集合创建相应的优先级矩阵，该优先级矩阵中的矩阵元素表示终端集群对应的资源获取优先级；
根据预设约束条件、质量矩阵、优先级矩阵以及预设限制条件获取相应的权值矩阵。
例如，在利用矩阵计算比例(即权值矩阵)的过程中，可以增加优先级向量Pm，其中pi表示终端集群i的资源获取优先级，此时，可以通过求解
Figure PCTCN2017082938-appb-000010
最小时的权值矩阵Nm×n，以得到比例。
作为一种实施方式，为满足人工干预调度的需求，解决自动化调度与人工调整冲突的问题，例如需要让美国资源全部用于覆盖美国本地，或者需要广东电信用户30％接入到深圳电信机群，本实施例方法还可以优先保证人工调整；也即本实施例的预设约束条件还可以包括：
该比例集合中部分该比例为预定比例。
比如，限制机器集群1覆盖终端集群2的比例n11等于30％，机器集群1覆盖终端集群2的比例n21等于20％等等。
例如，在通过矩阵计算比例的过程中，可以在计算前，对求解向量Nm×n的部分值进行初始化。例如设置n11＝30％，其中，该n11可以表示深圳电信机群覆盖广东电信终端集群的比例，然后再进行后续求解权值矩阵的计算。
作为一种实施方式，为能够满足扩容需求以及精确地对***扩容，在获取该机器集群覆盖该终端集群的比例之后，本实施例方法还可以包括：
根据该比例和该终端集群的资源需求获取终端集群对该机器集群的资源需求总和；
获取该资源需求总和与该机器集群的可用资源之间的资源差距；
根据该资源差距对该机器集群所属的***进行扩容。
比如，假设Qi表示终端集群对机器集群i的资源需求总和，该Qi＝u1×n1i+u2×n2i+……+um×nmi，其中ui为终端集群i的资源需求，nji为机器集群i覆盖终端集群的比例；此时，可以获取终端集群对机器集群i的需求总和Qi与该机器集群i之间的资源差距，即ci＝ai-Qi，其中ci表示机器集群i对应的资源差距；同理可以其他机器集群对应的资源差距，比如机器集群1对应的资源差距c1、机器集群2对应的资源差距c2……机器集群n对应的资源差距cn，然后，基于机器集群对应的资源差距对***进行扩容，例如，获取所有资源差距之和即c＝c1+c2……+cn，然后基于该c进行扩容等等。
在实际扩容时，有两种情况，一种是无用户增长预期的扩容，另一种是有用户增长预期的扩容，这都可以归纳为对终端集群的资源需求进行处理，比如，在矩阵运算时，可以对资源需求向量Um的调整；具体地，可以根据终端预期增长量对资源需求进行处理，也即所述根据该终端集群的资源需求、该机器集群的可用资源以及该接入质量，获取该机器集群接入该终端集群可以包括：
根据该终端集群对应的终端预期增长量对该终端集群的资源需求进行处理，以得到该终端集群的预期资源需求；
根据该终端集群的预期资源需求、该机器集群的可用资源以及该接入质量，获取该机器集群覆盖该终端集群的比例。
此时的比例，是考虑到终端预期增长量下的比例，然后，可以根据该比例和终端集群的资源需求获取资源差距，并基于该资源差距进行扩容；这样既可获取在有终端预期增长的情况下机群集群的扩容需求，为扩容提供参考，并基于该扩容需求进行扩容，可以提高扩容的准确性以及资源的利用率。
本实施例中，根据预期资源需求、可用资源以及接入质量获取比例的过程与上述步骤104类似，区别仅在于一个是资源需求，一个是预期资源需求，可以将上述终端集群的资源需求替换为预期资源需求来获取比例。
其中，终端预期增长量可以为终端增长比例，此时，可以将该预期增长量乘以终端寄去的资源需求，既可得到终端集群的预期资源需求，比如，对于终端集群i，该终端集群i的预期资源需求ui”＝ui×ui’，其中ui’为终端预期增长量，ui为终端集群i的可用资源。
例如，在通过矩阵运算求解比例时，可以增加向量Ui’,其中ui’表示终端集群i的终端预期增长量；在求解权值矩阵之前，利用ui’对ui进行处理，以得到终端集群i的预期资源需求ui”，该ui”＝ui×ui’，然后，将该ui替换成ui”之后，再求解权值矩阵。
作为一种实施方式，实际扩容中，一般***扩容是有限制的，即某些情况下只能申请到特定机群的资源，甚至扩容时，某些点的资源需要裁撤，某些点只能申请到固定数量的机器，而某些点的资源很充裕，几乎完全满足需求；因此，在扩容时还需要考虑到资源的约束，即资源增长量受限制，本实施例中可以将扩容资源的约束，归纳为对机器集群的可用资源进行调整，也即所述根据该终端集群的资源需求、该机器集群的可用资源以及该接入质量，获取该机器集群覆盖该终端集群的比例可以包括：
根据该机器集群对应的预设资源增长量对该机器集群的可用资源进行处理，以得到该机器集群的目标可用资源；
根据该终端集群的资源需求、该机器集群的目标可用资源以及该接入质量，获取该机器集群覆盖该终端集群的比例。
其中，资源增长量可以为资源增长比例，作为一种是实施方式，该预设资源增长量可以包括：预设上限资源增长量和预设绝对上限资源增长量；该预设上限资源增长量为机器集群资源增长的上限，该预设绝对上限资源 增长量为该机器集群资源增长的绝对上限；此时，可以根据预设上限资源增长量和预设绝对上限资源增长量对机器集群的可用资源进行处理，以得到机器集群的目标可用资源(假设gj为预设上限资源增长比例，hj为预设绝对上限资源增长比例，此时，机器集群的目标可用资源aj”’＝aj×gj+hj),然后，基于该目标可用资源、资源需求以及接入质量获取比例；这样本实施例方法可以获取在资源增加量限制的情况下比例，然后，根据该比例和终端集群的资源需求获取终端集群对机器集群的资源需求总和，并获取该资源需求总和与该机器集群的可用资源之间的资源差距，这样本实施例即可获取为在资源增长量限制的情况机器集群的扩容需求，为扩容提供参数，之后可以基于该扩容需求对***进行扩容，可以提高扩容的准确性，提高资源的利用率。
例如，在通过矩阵运算求解比例时，可以增加向量Gn＝[g1、g2、……gn]和向量Hn＝[h1、h2、……hn]，其中，Gn中元素gj代表机器集群j的预设上限资源增长比例，Hn中元素hj代表机器集群j的预设绝对上限资源增长比例，h1＝1时代表该机器集群无资源增加需求；在求解权值矩阵之前，利用gj和hj对aj进行处理，以得到机器集群j的目标可用资源aj”’，该aj”’＝aj×gj+hj，然后，将向量An中aj替换成aj”’之后，再求解权值矩阵。
作为一种实施方式，本实施方法可以适用于减容评估，此时，预设上限资源增长比例或者预设绝对上限资源增长比例可以小于1，也可以为负数。
由上可知，本发明实施例采用对终端进行集群划分，以得到多个终端集群，对提供资源的机器进行划分，以得到多个机器集群，然后，获取该终端集群的资源需求、该机器集群的可用资源、以及该终端集群与该机器集群之间的接入质量，根据该终端集群的资源需求、该机器集群的可用资源以及该接入质量，获取该机器集群覆盖该终端集群的比例，根据该比例 对该终端进行接入调度；该方案可以基于机器集群的可用资源、终端集群的资源需求以及终端集群与机器集群之间的接入质量来进行接入调度，可见该调度方案同时兼顾了接入容量和接入质量，因此，可以优化终端的接入调度，相对于现有技术而言，提高了终端的接入调度的精确性。
另外，本发明实施例还增加了接入质量的约束条件来获取比例，保证了调度方案的整体接入质量为最优，提升了资源利用率以及接入后的业务服务质量，进而提升了用户体验，此外，本发明实施例还可以获取可用资源与资源需求之间的资源差距，并根据该资源差距进行扩容，可以为扩容提供参考，自动输出扩容方案，进而提高***扩容的速度和精确性。
进一步地，本发明实施例还可以根据实际运营需求(资源使用需求、人工干预需求，扩容需求等等)对资源需求、可用资源等进行调整，使得本实施例的接入调度方法可以满足各种个性化的需求，该方法具有应用广泛、灵活性强、易于扩展等优点。
根据实施例一所描述的方法，以下将举例作进一步详细说明。
在本实施例中，将以该接入调度装置具体集成在调度服务器，且提供资源的机器为服务器为例进行详细说明。
如图2所示，一种终端的接入调度方法，具体流程可以如下：
步骤201、调度服务器将待调度的终端划分成多个终端集群，并获取每个终端集群的资源需求量，以得到资源需求集合。
其中，资源需求集合包括每个终端集群的资源需求量。在本实施例中，终端可以为计算机网络中处于网络最***的设备，也即终端可以为用户设备，比如终端可以为个人计算机、手机、平板电脑、或者其他用户设备。
具体地，调度服务器可以根据终端的地理位置或者网络地址对终端进行集群划分，比如，调度服务器根据终端所在的省份对终端进行集群划分。
本实施例中终端集群的资源需求量可以根据终端集群中终端的资源需求量获取，比如，为集群内终端的资源需求量的总和；该终端的资源需求可以根据历史请求量分析获得；新业务可根据主观评估等途径获得。
比如，调度服务器将终端划分成m个终端集群，然后，获取终端集群1的资源需求量u1、终端集群2的资源需求量u2……终端集群i的资源需求量i、终端集群m的资源需求量um，此时，可以得到集合u＝(u1、u2……ui……um)。
在采用矩阵模型计算比例集合时，可以将资源需求集合转换成相应的资源需求矩阵，如，将集合u转换成资源矩阵Um，该Um也可称为资源向量：
Um＝[u1,u2,……um]，其中，ui表示终端集群i的资源需求量。
作为一种实施方式，在有终端预期增长的扩容场景下，为满足扩容的个性化需求，还可以对终端集群的资源需求量进行调整，具体地，根据终端集群对应的终端预期增长量对该终端集群的资源需求量进行调整；如当终端预期增长量为终端预期增长比例时，可以将终端集群的资源需求量乘以该集群对应的预期增长比例。
例如，假设终端集群i的终端预期增长比例为ui’，如无增长，则u'i＝1资源需求量为ui时，本实施例可以利用该ui’对ui进行初始化，具体地使ui＝ui×u′i，此时ui即为终端集群i的预期可用资源。
又比如，在矩阵模型中，本实施例可以增加向量Um’，其中，uj’表达终端集群j的终端预期增长比例，此时，可通过公式ui＝ui×u'i对向量Um中每个元素进行初始化。
步骤202、调度服务器将待调度的服务器划分成多个服务集群，并获取每个服务集群的可用资源量，以得到可用资源集合。
其中，可用资源集合包括每个服务集群的可用资源量。
具体地，调度服务器可以根据服务器的属性对服务器进行划分，比如， 可以根据服务器所属的区域或者地理位置对服务器进行划分。
该服务集群的可用资源量可以根据服务集群中服务器的可用资源量得到，比如为集群中服务器的可用资源之和，其中，该服务器的可用资源量的获取方式可以有多种，比如，通过对服务器进行压力测试得到服务器的可用资源量，也可以对服务器的历史运营信息进行分析得到服务器的可用资源量等等。
例如，调度服务器将服务器划分成n个服务集群，然后，调度服务器获取每个服务器集群的可用资源量，具体地，获取服务集群1的可用资源量a1、服务集群2的可用资源量a2、……服务集群j的可用资源量aj……服务集群n的可用资源量an，此时，可以得到可以资源集合A＝(a1、a2……aj……am)。
在采用矩阵模型计算比例集合时，可以将可用资源集合转换成相应的可用资源矩阵，如，将集合A转换成资源矩阵An，该资源矩阵A也可称为向量An：
An＝[a1,a2,……an]，其中，aj表示服务集群j的可用资源量。
作为一种实施方式，为满足限制服务集群的资源使用需求，比如，所有服务集群的资源使用率在70％左右，某些服务集群的资源使用率达到80％等等；本实施例还可以对服务集群的可用资源量进行调整，具体地，可以根据服务集群的上限资源使用率对该服务集群的可用资源进行调整，比如可以根据上限资源使用率对可用资源进行初始化，其中上限资源使用率为服务集群的资源使用率上限。
例如，假设服务集群j的上限资源使用率为rj时，可以利用该rj对服务集群j的可用资源量aj进行初始化，具体地，可以使得aj＝aj×rj，此时，aj即为服务集群j的上限可用资源量，可用资源集合A此时为上限可用资源集合A。
又比如，在矩阵模型中，本实施例可以增加向量Rn，其中，rj表达服务集群j的上限资源使用率，此时，可通过公式得aj＝aj×rj对向量An中每个元 素进行初始化。
作为一种实施方式，在有资源约束的扩容场景下，也可以服务集群的可用资源量进行调整，具体地，可以根据服务集群对应的资源增长量对该服务集群的可用资源量进行调整，其中，该资源增长量可以包括上限资源增长量和绝对上限资源增长量；该增长量可以为增长比例。
例如，假设服务集群i的上限资源增长比例为gi，绝对上限资源增长比例为hi，此时，可以利用gi和hi对服务集群i的可用资源量ai进行初始化，具体地，使ai＝ai×gi+hi，此时ai即为服务集群i的目标可用资源。
又比如，在矩阵模型中，可以增加向量Gn＝[g1、g2、……gn]和向量Hn＝[h1、h2、……hn]，其中，Gn中元素gj代表服务集群j的预设上限资源增长比例，Hn中元素hj代表服务集群j的预设绝对上限资源增长比例，h1＝1时代表该服务集群无资源增加需求；通过公式ai＝ai×gi+hi对向量An中每个元素进行初始化。
步骤203、调度服务器获取终端集群与服务集群之间的接入质量，以得到接入质量集合。
该接入质量集合包括：每个终端集群与各服务集群之间的接入质量。
其中，终端集群与服务集群之间的接入质量可以为终端集群到服务集群之间的网络服务质量；其中，终端与服务器之间的网络服务质量(即接入质量)可以根据终端与服务之间的网络性能参数、终端与服务之间的历史运营信息、终端与服务之间的物理距离、终端的网络地址、终端和服务所属网络***的网络拓扑图中的至少一种得到。其中，网络性能参数可以包括：网络延时、丢包率和时延抖动等等。
例如，当有m个终端集群以及n个服务集群时，可以获取终端集群1与服务集群1之间的接入质量d11、终端集群1与服务集群2之间的接入质量d12……终端集群m与服务集群n之间的接入质量dmn，此时可以得到接入质量集合D＝ (d11、d12、d13……d1n、d21、d22……d2n……、dm1、dm2……、dmn)，其中，dij表示终端集群i与服务集群j之间的接入质量。
在采用矩阵模型计算比例集合时，可以将接入质量集合转换成相应的接入质量矩阵，如，将集合D转换成接入质量矩阵Dm×n，该Dm×n也可称为接入质量向量：
Figure PCTCN2017082938-appb-000011
其中，该dij表示终端集群i与服务集群j之间的接入质量。
为方便计算，本实施例还可以使用终端集群与服务集群之间的逻辑距离来表达接入质量，此时，接入质量矩阵可以为逻辑距离矩阵，比如，此时Dm×n可以逻辑距离矩阵，其中，dij为终端集群i与服务及其j之间的逻辑距离，该逻辑距离表示终端集群i与服务集群j之间的接入质量，逻辑距离越小表示接入质量越大或者越好。
步骤204、调度服务器根据接入质量集合、资源需求集合以及可用资源集合，获取比例集合，该比例集合包括服务集群覆盖终端集群的比例。
其中，服务集群覆盖终端集群的比例为该服务集群为该终端集群提供的资源占该终端集群的资源需求的百分比；比如，服务集群1覆盖30％的终端集群1，其表示服务集群1为终端集群1提供的资源占该终端集群1的资源需求的30％，该i和j均为大于1的正整数。
比如，调度服务器根据接入质量集合D、资源需求集合U以及可用资源集合A获取比例集合N，该比例集合N＝(n11、n12、n13……n1n、n21、n22……n2n……、nm1、nm2……、nmn)，其中，nij表示服务集群j覆盖终端集群i的比例。
在本实施例，根据接入质量集合、资源需求集合以及可用资源集合获 取比例集合的方式有多种，比如，以终端集群对机器集群的资源需求总和小于该机器集群的可用资源为资源限制条件，然后，根据该限制条件和接入质量集合来获取比例集合；也即所述调度服务器根据接入质量集合、资源需求集合以及可用资源集合，获取比例集合可以包括：
调度服务器根据资源限制条件和接入质量集合获取相应的比例集合；
其中，该资源限制条件包括：终端集群对服务集群的资源需求总和小于或等于该服务集群的可用资源，资源需求总和根据资源需求集合和欲获取的比例集合获得。
本实施例中，终端集群对服务集群的资源需求总和为服务集群需要为其所有终端集群提供的资源之和。
例如，在采用矩阵模型时，可以根据接入质量矩阵Dm×n、资源需求集合Um以及可用资源矩阵An获取权值矩阵Nm×n，其中，权值矩阵Nm×n中元素nij表示服务集群j覆盖终端集群i的比例；具体地，可以根据资源限制条件：Um×Nm×n≤An，以及接入质量矩阵Dm×n，求解权值矩阵Nm×n，从而得到相应的比例集合。
作为一种实施方式，为了能够提升业务服务质量，保证接入调度的合理性，本实施例可以使用终端集群与机器集群之间的接入质量总和最大为最优调度；即本实施例获取在接入质量总和最大或者好情况下的比例集合；此时所述调度服务器根据资源限制条件和接入质量集合获取相应的比例集合可以包括：
调度服务器根据资源限制条件、质量约束条件以及接入质量集合获取相应的比例集合；其中，质量约束条件包括：接入质量集合中接入质量的总和最大。作为一种实施方式，该质量约束条件可以由接入质量集合和欲获取的比例集合表示或者获取。
例如，在采用矩阵模型时，首先，获取接入质量矩阵Dm×n、资源需求 集合Um以及可用资源矩阵An，然后，获取资源限制条件：Um×Nm×n≤An，其中，接入质量的总和最大的质量约束条件可以由接入质量矩阵Dm×n和欲获取的权值矩阵Nm×n表示，如可以通过
Figure PCTCN2017082938-appb-000012
最小或者最大时表示接入质量的总和最大，此时只需求解当
Figure PCTCN2017082938-appb-000013
最大或者最小时的权值矩阵Nm×n即可得到比例集合。在实际应用中，如果Dm×n中的dij为终端集群i与服务集群j之间的逻辑距离(逻辑距离越小表示的接入质量越大)，那么此时只需根据资源限制条件求解当
Figure PCTCN2017082938-appb-000014
最小时的权值矩阵Nm×n，即可得所需的比例集合，也即最优的调度。
作为一种实施方式，为保证某些终端集群可以优选获取资源，比如，在根据国家对终端进行机器划分时，希望中国的终端相比于英国优先获得资源；本实施例设置终端集群的资源获取优先级，在获取比例时可将资源获取优先级也考虑进去，此时，本实施例中所述调度服务器根据资源限制条件、质量约束条件以及接入质量集合获取相应的比例集合可以包括：
调度服务器根据资源限制条件、优先级集合、质量约束条件以及接入质量集合获取相应的比例集合，其中，优先级集合包括终端集群的资源获取优先级。
例如，在采用矩阵模型计算比例是，可以将优先级集合P转换成矩阵Pm，即优先级向量，其中pi表示终端集群i的资源获取优先级，此时，可以通过求解
Figure PCTCN2017082938-appb-000015
最小或者最大时的权值矩阵Nm×n，以得到比例集合。
作为一种实施方式，为满足人工干预调度的需求，解决自动化调度与人工调整冲突的问题，本实施例方法还可以优先保证人工调整；也即本实施例方法还可以增加比例约束条件，该比例约束条件可以包括：比例集合中部分该比例为预定比例；此时所述调度服务器根据资源限制条件、质量约束条件以及接入质量集合获取相应的比例集合可以包括：调度服务器根 据资源限制条件、质量约束条件、比例约束条件以及接入质量集合获取相应的比例集合。
例如，在通过矩阵模型计算比例的过程中，可以在计算前，对求解向量Nm×n的部分值进行初始化。例如设置n31＝40％，其中，该n31可以表示服务集群3覆盖终端集群的比例为40％。
由于一旦限制了服务集群的资源使用率，就严格限制了服务集群的资源大小，也即采用预设上限资源使用率对服务集群的可用资源进行调整后，此时，可能会存在无法获取比例进而导致无法进行调度，比如，当某个终端集群的需求增加的情况下，可能会导致终端集群对服务集群的资源需求总和大于该服务集群的上限可用资源量，此时无法获取满足资源限制条件，因此，无法获取比例集合；为克服该问题，本实施例还可以在根据上限资源使用率对可用资源量调整，且调度服务获取比例集合失败的情况下，根据绝对上限资源使用率对可用资源集合中的可用资源量进行调整，该绝对上限资源使用率大于上限资源使用率。
例如，假设服务集群j的上限资源使用率为rj、绝对上限资源使用率为rj’时，为满足资源使用需求，可以利用该rj对服务集群j的可用资源量aj进行初始化，具体地，可以使得aj＝aj×rj，此时，aj即为服务集群j的上限可用资源量，之后，调度服务器可以根据接入质量集合D、资源需求集合U以及可用资源集合A获取比例集合N，如果获取比例集合失败，则可以根据绝对上限资源使用率为rj’对可用资源集合A中上限可用资源量aj进行初始化，初始化的aj即为绝对上限可用资源量，具体地，可以使得
Figure PCTCN2017082938-appb-000016
在对可用资源集合A中所有上限可用资源量初始化完成之后，调度服务器可以再次根据接入质量集合D、资源需求集合U以及可用资源集合A获取比例集合N。
步骤205、调度服务器根据该比例集合为终端集群分配相应的目标服务集群。
具体地，可以通过预设算法和该比例集合为终端集群分配相应的目标服务集群。
步骤206、调度服务器在该目标服务集群中为该终端集群中的终端分配需要接入的目标服务器。
具体地分配时，可以根据目标服务集群中服务器的可用资源、或者地理位置等为终端分配目标服务器。
需说明的是，具体实施时，该调度服务器可以为一独立的实体，也可以包括多个实体，比如，可以包括多个服务设备。
由上可知，本实施例采用调度服务器根据接入质量集合、资源需求集合以及可用资源集合，获取比例集合，然后，调度服务器根据该比例集合为终端集群分配相应的目标服务集群，调度服务器在该目标服务集群中为该终端集群中的终端分配需要接入的目标服务器；该方案可以基于服务集群的可用资源、终端集群的资源需求以及终端集群与服务集群之间的接入质量来进行接入调度，可见该调度方案同时兼顾了接入容量和接入质量，因此，可以优化终端的接入调度，相对于现有技术而言，提高了终端的接入调度的精确性。
此外，本发明实施例可以通过矩阵向量模型来计算权值矩阵，从而快速得到服务集群覆盖终端集群的比例，提高了接入调度的效率。
另一方面，本发明实施例提供的调度方案还可以默认接入质量最好时为最优调度，可以提高***资源的利用率，并提高了接入后的业务服务质量；并且该调度方案还可根据实际运营需求(资源使用需求、人工干预需求，扩容需求等等)对资源需求、可用资源等进行调整，使得本实施例的接入调度方法可以满足各种个性化的需求，该方法具有应用广泛、灵活性强、易于扩展等优点。
为了更好地实施以上方法，本发明实施例还提供一种终端的接入调度装置，如图3a所示，该接入调度装置还可以包括第一划分单元301、第二划分单元302、第一获取单元303、第二获取单元304和调度单元305，如下：
(1)第一划分单元301；
第一划分单元301，配置为对终端进行集群划分，以得到多个终端集群。
比如，第一划分单元301可以配置为根据终端的地理位置，对该终端进行集群划分。
其中，终端的地理位置可以包括终端当前的地理位置，或者终端的网络注册归属地；例如，可以按照终端当前所在或者网络注册的区域(地区、市、省份、国家等)进行集群划分，具体地，将当前所在或者网络注册的区域相同的终端归为同一个终端集群。
(2)第二划分单元302；
第二划分单元302，配置为对提供资源的机器进行集群划分，以得到多个机器集群。
比如，第二划分单元302可以配置为根据提供资源的机器的地理位置，对该机器进行集群划分。例如可以根据机器所在的区域(地区、市、省份、国家等)进行集群划分，比如，将位于某个省份的机器划分成一个机器集群。
(3)第一获取单元303；
第一获取单元303，配置为获取该终端集群的资源需求、该机器集群的可用资源、以及该终端集群与该机器集群之间的接入质量。
其中，获取终端集群的资源需求的方式可以有多种；比如可以基于终端集群的历史请求量获取，即第一获取单元303可以配置为：
获取终端集群中终端的历史请求量；
根据该历史请求量获取该终端集群中终端的资源需求；
根据终端集群中终端的资源需求，获取该终端集群的资源需求。
同样，本实施例中获取机器集群的可用资源的方式也可以有多种，比如，可以通过压力测试方式得到，具体地，对机器集群中的机器进行压力测试，以得到该机器的可用资源，然后，根据机器集群中机器的可用资源得到机器集群的可用资源。在实际测试时为了提高获取速度，提高接入调度的效率可以基于机器的类型进行压测，然后结合不同类型的机器的数量得到机器集群的可用资源。
本实施例中，终端集群与机器集群之间的接入质量可以为终端集群到机器集群之间的网络服务质量；其中，终端与机器之间的网络服务质量(即接入质量)可以根据终端与机器之间的网络性能参数、终端与机器之间的历史运营信息、终端与机器之间的物理距离、终端的网络地址、终端和机器所属网络***的网络拓扑图中的至少一种得到。其中，网络性能参数可以包括：网络延时、丢包率和时延抖动等等。
(4)第二获取单元304；
第二获取单元304，配置为根据该终端集群的资源需求、该机器集群的可用资源以及该接入质量，获取该机器集群接入该终端集群的比例。
比如，该第二获取单元304，可以配置为：根据预设限制条件和该终端集群与该机器集群之间的接入质量，获取该机器集群覆盖该终端集群的比例；
其中，该预设限制条件包括：终端集群对该机器集群的资源需求总和小于或等于该机器集群的可用资源；其中，该资源需求总和根据该终端集群的资源需求和欲获取的比例获得。
又比如，该第二获取单元304，可以配置为：根据预设约束条件、接入质量集合以及该预设限制条件获取比例集合，该接入质量集合包括多个该终端集群与该机器集群之间的接入质量，该比例集合包括该机器集群覆盖 该终端集群的比例；
其中，该预设约束条件包括：该接入质量集合中接入质量的总和最大。
具体地，该第二获取单元304可以包括：矩阵创建子单元和比例获取子单元；
该矩阵创建子单元，配置为根据该接入质量集合创建相应的质量矩阵，该质量矩阵中的矩阵元素表示该终端集群与该机器集群之间的接入质量；
该比例获取子单元，配置为根据该预设约束条件、该质量矩阵以及该预设限制条件获取相应的权值矩阵，该权值矩阵中的矩阵元素表示该机器集群覆盖该终端集群的比例；
该调度单元，具体配置为根据该权值矩阵对该终端进行接入调度。
作为一种实施方式，为满足资源限制需求，该第二获取单元304可以包括：处理子单元和比例获取子单元；
该处理子单元，配置为根据该机器集群的预设上限资源使用率对该机器集群的可用资源进行处理，以得到该机器集群的上限可用资源；
该比例获取子单元，配置为根据该终端集群的资源需求、该机器集群的上限可用资源以及该接入质量，获取该机器集群覆盖该终端集群的比例。
所述处理子单元，还配置为：当获取该机器集群接入该终端集群的比例失败时，根据预设绝对上限资源使用率对该机器集群的上限可用资源进行处理，以得到该机器集群的绝对上限可用资源，其中，该预设绝对上限资源使用率大于预设上限资源使用率；
所述比例获取子单元，还配置为根据该终端集群的资源需求、该机器集群的绝对上限可用资源以及该接入质量，获取该机器集群覆盖该终端集群的比例。
作为一种实施方式，为满足资源优选获取的需求，本实施例中第二获取单元，可以配置为：
根据该终端集群的资源获取优先级、预设限制条件和该终端集群与该机器集群之间的接入质量，获取该机器集群覆盖该终端集群。
作为一种实施方式，为满足扩容的个性化需求，参考图3b，本实施例接入调度装置还包括：第三获取单元306、资源差距获取单元307以及扩容单元308；
所述第三获取单元306，配置为根据该比例和该终端集群的资源需求获取终端集群对该机器集群的资源需求总和；
所述资源差距获取单元307，配置为获取该资源需求总和与该机器集群的可用资源之间的资源差距；
所述扩容单元308，配置为根据该资源差距对该机器集群所属的***进行扩容。
作为一种实施方式，在图3b所示的接入装置基础上，为能够实现有用户预期增长的扩容，该第二获取单元304，可以配置为：
根据该终端集群对应的终端预期增长量对该终端集群的资源需求进行处理，以得到该终端集群的预期资源需求；
根据该终端集群的预期资源需求、该机器集群的可用资源以及该接入质量，获取该机器集群覆盖该终端集群的比例。
作为一种实施方式，在图3b所示的接入装置基础上，为能够实现有资源限制的扩容，该第二获取单元304，可以配置为：
根据该机器集群对应的预设资源增长量对该机器集群的可用资源进行处理，以得到该机器集群的目标可用资源；
根据该终端集群的资源需求、该机器集群的目标可用资源以及该接入质量，获取该机器集群覆盖该终端集群的比例。
(5)调度单元305；
调度单元305，配置为根据该比例对比例集合或者权值矩阵。
比如，在获取比例集合或者权值矩阵的情况下，该调度单元305可以配置为根据比例集合或者权值矩阵比例集合或者权值矩阵。
本实施例中，该调度单元305可以配置为：
根据该比例确定该终端集群接入的目标机器集群；
从该目标机器集群中确定该终端集群中终端需要接入的目标机器。
具体地，可以采用预设算法对比例进行运算，以得到运算结果，根据该运算结果确定终端集群接入的目标机器集群；其中，该预设算法可以根据实际需求设定，比如该预设算法可以为加权算法等等。
具体实施时，以上各个单元可以作为独立的实体来实现，也可以进行任意组合，作为同一或若干个实体来实现，以上各个单元的具体实施可参见前面的方法实施例，在此不再赘述。
该接入调度装置具体可以集成在服务器或其他需要进行接入调度的设备中。
由上可知，本实施例的接入调度装置采用第一划分单元301对终端进行集群划分，以得到多个终端集群，第二划分单元302对提供资源的机器进行划分，以得到多个机器集群，然后，由第一获取单元303获取该终端集群的资源需求、该机器集群的可用资源、以及该终端集群与该机器集群之间的接入质量，由第二获取单元304根据该终端集群的资源需求、该机器集群的可用资源以及该接入质量，获取该机器集群覆盖该终端集群的比例，由调度单元305根据该比例对该终端进行接入调度；该方案可以基于机器集群的可用资源、终端集群的资源需求以及终端集群与机器集群之间的接入质量来进行接入调度，可见该调度方案同时兼顾了接入容量和接入质量，因此，可以优化终端的接入调度，相对于现有技术而言，提高了终端的接入调度的精确性。
在本发明所提供的几个实施例中，应该理解到，所揭露的方法及装置， 可以通过其他的方式实现。以上所描述的装置实施例仅仅是示意性的，例如，所述模块的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，如：多个模块或组件可以结合，或可以集成到另一个***，或一些特征可以忽略，或不执行。另外，所显示或讨论的各组成部分相互之间的通信连接可以是通过一些接口，设备或模块的间接耦合或通信连接，可以是电性的、机械的或其他形式的。
上述作为分离部件说明的模块可以是、或也可以不是物理上分开的，作为模块显示的部件可以是、或也可以不是物理模块，即可以位于一个地方，也可以分布到多个网络模块上；可以根据实际的需要选择其中的部分或全部模块来实现本实施例方案的目的。
另外，在本发明各实施例中的各功能模块可以全部集成在一个处理模块中，也可以是各模块分别单独作为一个模块，也可以两个或两个以上模块集成在一个模块中；上述集成的模块既可以采用硬件的形式实现，也可以采用硬件加软件功能模块的形式实现。
本领域普通技术人员可以理解：实现上述方法实施例的全部或部分步骤可以通过程序指令相关的硬件来完成，前述的程序可以存储于计算机可读取存储介质中，该程序在执行时，执行包括上述方法实施例的步骤；而前述的存储介质包括：移动存储设备、只读存储器(ROM，Read-Only Memory)、磁碟或者光盘等各种可以存储程序代码的介质。
或者，本发明实施例上述集成的模块如果以软件功能模块的形式实现并作为独立的产品销售或使用时，也可以存储在一个计算机可读取存储介质中。基于这样的理解，本发明实施例的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机、服务器、或者网络设备等)执行本发明各个实施例所述方法的 全部或部分。而前述的存储介质包括：移动存储设备、ROM、磁碟或者光盘等各种可以存储程序代码的介质。
本发明实施例中记载的存储器切换方法、装置只以上述实施例为例，但不仅限于此，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分或者全部技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本发明各实施例技术方案的范围。
以上所述仅为本发明的较佳实施例而已，并非用于限定本发明的保护范围。
工业实用性
本发明实施例的技术方案基于机器集群的可用资源、终端集群的资源需求以及终端集群与机器集群之间的接入质量来进行接入调度，可见该调度方案同时兼顾了接入容量和接入质量，因此，可以优化终端的接入调度，相对于现有技术而言，提高了终端的接入调度的精确性。

Claims (22)

  1. 一种终端的接入调度方法，包括：
    对终端进行集群划分，以得到多个终端集群；
    对提供资源的机器进行集群划分，以得到多个机器集群；
    获取所述终端集群的资源需求、所述机器集群的可用资源、以及所述终端集群与所述机器集群之间的接入质量；
    根据所述终端集群的资源需求、所述机器集群的可用资源以及所述接入质量，获取所述机器集群覆盖所述终端集群的比例；
    根据所述比例对所述终端进行接入调度。
  2. 如权利要求1所述的接入调度方法，其中，所述根据所述终端集群的资源需求、所述机器集群的可用资源以及所述接入质量，获取所述机器集群覆盖所述终端集群的比例的步骤包括：
    根据预设限制条件和所述终端集群与所述机器集群之间的接入质量，获取所述机器集群覆盖所述终端集群的比例；
    其中，所述预设限制条件包括：终端集群对所述机器集群的资源需求总和小于或等于所述机器集群的可用资源；其中，所述资源需求总和根据所述终端集群的资源需求和欲获取的比例获得。
  3. 如权利要求2所述的接入调度方法，其中，所述根据预设限制条件和所述终端集群与所述机器集群之间的接入质量，获取所述机器集群覆盖所述终端集群的比例的步骤包括：
    根据预设约束条件、接入质量集合以及所述预设限制条件获取比例集合，所述接入质量集合包括多个所述终端集群与所述机器集群之间的接入质量，所述比例集合包括所述机器集群覆盖所述终端集群的比例；
    其中，所述预设约束条件包括：所述接入质量集合中接入质量的总和最大。
  4. 如权利要求3所述的接入调度方法，其中，所述根据预设约束条件、接入质量集合以及所述预设限制条件获取比例集合的步骤包括：
    根据所述接入质量集合创建相应的质量矩阵，所述质量矩阵中的矩阵元素表示所述终端集群与所述机器集群之间的接入质量；
    根据所述预设约束条件、所述质量矩阵以及所述预设限制条件获取相应的权值矩阵，所述权值矩阵中的矩阵元素表示所述机器集群覆盖所述终端集群的比例；
    所述根据所述比例对所述终端进行接入调度的步骤具体包括：根据所述权值矩阵对所述终端进行接入调度。
  5. 如权利要求3所述的接入调度方法，其中，所述预设约束条件还包括：所述比例集合中部分所述比例为预定比例。
  6. 如权利要求1所述的接入调度方法，其中，根据所述终端集群的资源需求、所述机器集群的可用资源以及所述接入质量，获取所述机器集群覆盖所述终端集群的比例的步骤包括：
    根据所述机器集群的预设上限资源使用率对所述机器集群的可用资源进行处理，以得到所述机器集群的上限可用资源；
    根据所述终端集群的资源需求、所述机器集群的上限可用资源以及所述接入质量，获取所述机器集群覆盖所述终端集群的比例。
  7. 如权利要求6所述的接入调度方法，其中，还包括：
    当获取所述机器集群接入所述终端集群的比例失败时，根据预设绝对上限资源使用率对所述机器集群的上限可用资源进行处理，以得到所述机器集群的绝对上限可用资源，其中，所述预设绝对上限资源使用率大于预设上限资源使用率；
    根据所述终端集群的资源需求、所述机器集群的绝对上限可用资源以及所述接入质量，获取所述机器集群覆盖所述终端集群的比例。
  8. 如权利要求2所述的接入调度方法，其中，所述根据预设限制条件和所述终端集群与所述机器集群之间的接入质量，获取所述机器集群覆盖所述终端集群的比例的步骤具体包括：
    根据所述终端集群的资源获取优先级、预设限制条件和所述终端集群与所述机器集群之间的接入质量，获取所述机器集群覆盖所述终端集群。
  9. 如权利要求1所述的接入调度方法，其中，在获取所述机器集群覆盖所述终端集群的比例之后，所述接入调度方法还包括：
    根据所述比例和所述终端集群的资源需求获取终端集群对所述机器集群的资源需求总和；
    获取所述资源需求总和与所述机器集群的可用资源之间的资源差距；
    根据所述资源差距对所述机器集群所属的***进行扩容。
  10. 权利要求9所述的接入调度方法，其中，所述根据所述终端集群的资源需求、所述机器集群的可用资源以及所述接入质量，获取所述机器集群覆盖所述终端集群的比例的步骤包括：
    根据所述终端集群对应的终端预期增长量对所述终端集群的资源需求进行处理，以得到所述终端集群的预期资源需求；
    根据所述终端集群的预期资源需求、所述机器集群的可用资源以及所述接入质量，获取所述机器集群覆盖所述终端集群的比例。
  11. 如权利要求9所述的接入调度方法，其中，所述根据所述终端集群的资源需求、所述机器集群的可用资源以及所述接入质量，获取所述机器集群覆盖所述终端集群的比例的步骤具体包括：
    根据所述机器集群对应的预设资源增长量对所述机器集群的可用资源进行处理，以得到所述机器集群的目标可用资源；
    根据所述终端集群的资源需求、所述机器集群的目标可用资源以及所述接入质量，获取所述机器集群覆盖所述终端集群的比例。
  12. 一种终端的接入调度装置，包括：
    第一划分单元，配置为对终端进行集群划分，以得到多个终端集群；
    第二划分单元，配置为对提供资源的机器进行集群划分，以得到多个机器集群；
    第一获取单元，配置为获取所述终端集群的资源需求、所述机器集群的可用资源、以及所述终端集群与所述机器集群之间的接入质量；
    第二获取单元，配置为根据所述终端集群的资源需求、所述机器集群的可用资源以及所述接入质量，获取所述机器集群覆盖所述终端集群的比例；
    调度单元，配置为根据所述比例对所述终端进行接入调度。
  13. 如权利要求12所述的接入调度装置，其中，所述第二获取单元，配置为：
    根据预设限制条件和所述终端集群与所述机器集群之间的接入质量，获取所述机器集群覆盖所述终端集群的比例；
    其中，所述预设限制条件包括：终端集群对所述机器集群的资源需求总和小于或等于所述机器集群的可用资源；其中，所述资源需求总和根据所述终端集群的资源需求和欲获取的比例获得。
  14. 如权利要求13所述的接入调度装置，其中，所述第二获取单元，配置为：
    根据预设约束条件、接入质量集合以及所述预设限制条件获取比例集合，所述接入质量集合包括多个所述终端集群与所述机器集群之间的接入质量，所述比例集合包括所述机器集群覆盖所述终端集群的比例；
    其中，所述预设约束条件包括：所述接入质量集合中接入质量的总和最大。
  15. 如权利要求14所述的接入调度装置，其中，所述第二获取单元包 括：矩阵创建子单元和比例获取子单元；
    所述矩阵创建子单元，配置为根据所述接入质量集合创建相应的质量矩阵，所述质量矩阵中的矩阵元素表示所述终端集群与所述机器集群之间的接入质量；
    所述比例获取子单元，配置为根据所述预设约束条件、所述质量矩阵以及所述预设限制条件获取相应的权值矩阵，所述权值矩阵中的矩阵元素表示所述机器集群覆盖所述终端集群的比例；
    所述调度单元，配置为根据所述权值矩阵对所述终端进行接入调度。
  16. 如权利要求12所述的接入调度装置，其中，所述第二获取单元包括：处理子单元和比例获取子单元；
    所述处理子单元，配置为根据所述机器集群的预设上限资源使用率对所述机器集群的可用资源进行处理，以得到所述机器集群的上限可用资源；
    所述比例获取子单元，配置为根据所述终端集群的资源需求、所述机器集群的上限可用资源以及所述接入质量，获取所述机器集群覆盖所述终端集群的比例。
  17. 如权利要求16所述的接入调度装置，其中，
    所述处理子单元，还配置为：当获取所述机器集群接入所述终端集群的比例失败时，根据预设绝对上限资源使用率对所述机器集群的上限可用资源进行处理，以得到所述机器集群的绝对上限可用资源，其中，所述预设绝对上限资源使用率大于预设上限资源使用率；
    所述比例获取子单元，还配置为根据所述终端集群的资源需求、所述机器集群的绝对上限可用资源以及所述接入质量，获取所述机器集群覆盖所述终端集群的比例。
  18. 如权利要求13所述的接入调度装置，其中，所述第二获取单元，配置为：
    根据所述终端集群的资源获取优先级、预设限制条件和所述终端集群与所述机器集群之间的接入质量，获取所述机器集群覆盖所述终端集群。
  19. 如权利要求12所述的接入调度装置，其中，还包括：第三获取单元、资源差距获取单元以及扩容单元；
    所述第三获取单元，配置为根据所述比例和所述终端集群的资源需求获取终端集群对所述机器集群的资源需求总和；
    所述资源差距获取单元，配置为获取所述资源需求总和与所述机器集群的可用资源之间的资源差距；
    所述扩容单元，配置为根据所述资源差距对所述机器集群所属的***进行扩容。
  20. 如权利要求19所述的接入调度装置，其中，所述第二获取单元，配置为：
    根据所述终端集群对应的终端预期增长量对所述终端集群的资源需求进行处理，以得到所述终端集群的预期资源需求；
    根据所述终端集群的预期资源需求、所述机器集群的可用资源以及所述接入质量，获取所述机器集群覆盖所述终端集群的比例。
  21. 如权利要求19所述的接入调度装置，其中，所述第二获取单元，配置为：
    根据所述机器集群对应的预设资源增长量对所述机器集群的可用资源进行处理，以得到所述机器集群的目标可用资源；
    根据所述终端集群的资源需求、所述机器集群的目标可用资源以及所述接入质量，获取所述机器集群覆盖所述终端集群的比例。
  22. 一种计算机存储介质，所述计算机存储介质中存储有计算机可执行指令，所述计算机可执行指令用于执行权利要求1至11任一项所述的接入调度方法。
PCT/CN2017/082938 2016-05-09 2017-05-03 一种终端的接入调度方法、装置及计算机存储介质 WO2017193850A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP17795468.2A EP3451727B1 (en) 2016-05-09 2017-05-03 Access scheduling method and device for terminal, and computer storage medium
US16/047,402 US10652360B2 (en) 2016-05-09 2018-07-27 Access scheduling method and apparatus for terminal, and computer storage medium

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610301528.1A CN107360202A (zh) 2016-05-09 2016-05-09 一种终端的接入调度方法和装置
CN201610301528.1 2016-05-09

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US16/047,402 Continuation US10652360B2 (en) 2016-05-09 2018-07-27 Access scheduling method and apparatus for terminal, and computer storage medium

Publications (1)

Publication Number Publication Date
WO2017193850A1 true WO2017193850A1 (zh) 2017-11-16

Family

ID=60266226

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/082938 WO2017193850A1 (zh) 2016-05-09 2017-05-03 一种终端的接入调度方法、装置及计算机存储介质

Country Status (4)

Country Link
US (1) US10652360B2 (zh)
EP (1) EP3451727B1 (zh)
CN (1) CN107360202A (zh)
WO (1) WO2017193850A1 (zh)

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200104177A1 (en) * 2017-05-30 2020-04-02 Nec Corporation Resource allocation system, management device, method, and program
US20200090096A1 (en) * 2017-05-30 2020-03-19 Nec Corporation Resource management system, management device, method, and program
US10747785B2 (en) * 2017-11-01 2020-08-18 Mad Street Den, Inc. Method and system for efficient clustering of combined numeric and qualitative data records
CN108667654B (zh) * 2018-04-19 2021-04-20 北京奇艺世纪科技有限公司 服务器集群自动扩容方法及相关设备
CN109802997B (zh) * 2018-12-20 2021-02-09 华为技术服务有限公司 一种节点设备的选择方法及其相关设备
US20220225345A1 (en) * 2019-05-29 2022-07-14 Telefonaktiebolaget Lm Ericsson (Publ) Method, network device and network node for scheduling terminal devices
CN111049710A (zh) * 2019-11-20 2020-04-21 视联动力信息技术股份有限公司 一种接入服务器的方法和装置
CN112888076B (zh) * 2019-11-29 2023-10-24 华为技术有限公司 一种调度方法及装置
CN111211998A (zh) * 2019-12-12 2020-05-29 北京淇瑀信息科技有限公司 可弹性扩容的资源配给方法、装置和电子设备
US10860381B1 (en) * 2020-05-14 2020-12-08 Snowflake Inc. Flexible computing

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102124813A (zh) * 2008-08-15 2011-07-13 高通股份有限公司 小区间mimo***的分层级聚类架构
CN102132595A (zh) * 2008-08-15 2011-07-20 高通股份有限公司 网络 mimo ***的频率-时间中的自适应聚类架构
CN103841126A (zh) * 2012-11-20 2014-06-04 ***通信集团辽宁有限公司 一种共享资源的调配方法及装置
WO2016037660A1 (en) * 2014-09-12 2016-03-17 Huawei Technologies Co.,Ltd Method and apparatus for determining clusters of access nodes

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101583194B (zh) * 2009-06-18 2011-05-11 清华大学 基于虚拟小区的基站间协作的资源分配方法及其***
CN104461740B (zh) * 2014-12-12 2018-03-20 国家电网公司 一种跨域集群计算资源聚合和分配的方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102124813A (zh) * 2008-08-15 2011-07-13 高通股份有限公司 小区间mimo***的分层级聚类架构
CN102132595A (zh) * 2008-08-15 2011-07-20 高通股份有限公司 网络 mimo ***的频率-时间中的自适应聚类架构
CN103841126A (zh) * 2012-11-20 2014-06-04 ***通信集团辽宁有限公司 一种共享资源的调配方法及装置
WO2016037660A1 (en) * 2014-09-12 2016-03-17 Huawei Technologies Co.,Ltd Method and apparatus for determining clusters of access nodes

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3451727A4 *

Also Published As

Publication number Publication date
US20180375957A1 (en) 2018-12-27
EP3451727A4 (en) 2019-12-25
US10652360B2 (en) 2020-05-12
CN107360202A (zh) 2017-11-17
EP3451727B1 (en) 2021-08-11
EP3451727A1 (en) 2019-03-06

Similar Documents

Publication Publication Date Title
WO2017193850A1 (zh) 一种终端的接入调度方法、装置及计算机存储介质
CN109067862B (zh) API Gateway自动伸缩的方法与装置
CN109032801B (zh) 一种请求调度方法、***及电子设备和存储介质
CN107566194B (zh) 一种实现跨域虚拟网络映射的方法
CN103699445A (zh) 一种任务调度方法、装置及***
CN111817974B (zh) 基于令牌桶的接口限流方法、装置、***及可读存储介质
CN108776934A (zh) 分布式数据计算方法、装置、计算机设备及可读存储介质
US11856246B2 (en) CDN optimization platform
CN114035941A (zh) 资源调度***、方法以及计算设备
CN110661654B (zh) 一种网络带宽资源分配方法、装置、设备及可读存储介质
CN107846371B (zh) 一种多媒体业务QoE资源分配方法
CN107295090A (zh) 一种资源调度的方法和装置
CN109413147B (zh) 服务节点的管理方法、装置、设备及计算机可读存储介质
CN113348651A (zh) 切片的虚拟网络功能的动态云间放置
US10970309B2 (en) Data storage method and apparatus
CN110149377A (zh) 一种视频服务节点资源分配方法、***、装置及存储介质
CN112689007A (zh) 资源分配方法、装置、计算机设备和存储介质
Li et al. Delay-aware resource allocation for data analysis in cloud-edge system
Karakoc et al. Multi-layer decomposition of optimal resource sharing problems
CN111314234B (zh) 一种流量分配的方法、装置、存储介质及电子设备
WO2017148246A1 (zh) 一种数据配置方法和装置
Dimolitsas et al. AHP4HPA: An AHP-based Autoscaling Framework for Kubernetes Clusters at the Network Edge
CN113765966A (zh) 一种负载均衡方法和装置
CN111988388A (zh) 流量分配的方法、装置、电子设备及存储介质
CN110190988A (zh) 一种业务部署方法及装置

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17795468

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2017795468

Country of ref document: EP

Effective date: 20181130