WO2020143164A1 - 一种网络资源的分配方法及设备 - Google Patents

一种网络资源的分配方法及设备 Download PDF

Info

Publication number
WO2020143164A1
WO2020143164A1 PCT/CN2019/091499 CN2019091499W WO2020143164A1 WO 2020143164 A1 WO2020143164 A1 WO 2020143164A1 CN 2019091499 W CN2019091499 W CN 2019091499W WO 2020143164 A1 WO2020143164 A1 WO 2020143164A1
Authority
WO
WIPO (PCT)
Prior art keywords
resource
occupancy
actual
server
ratio
Prior art date
Application number
PCT/CN2019/091499
Other languages
English (en)
French (fr)
Inventor
易仁杰
张伟新
Original Assignee
平安科技(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 平安科技(深圳)有限公司 filed Critical 平安科技(深圳)有限公司
Publication of WO2020143164A1 publication Critical patent/WO2020143164A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44Arrangements for executing specific programs
    • G06F9/455Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines

Definitions

  • This application belongs to the field of Internet technology, and particularly relates to a method and equipment for allocating network resources.
  • cloud servers can be shared through cloud services.
  • virtual machine leasing and real-time computing services respond to user service requests through two independent server clusters. Because user needs are dynamically changing, there are fewer cloud server leasing requests at certain times. There are many real-time computing service requests at this moment, and the above two systems are independent of each other, which is not conducive to resource allocation, thereby reducing the resource utilization rate of the system and unable to meet the user's service needs.
  • the embodiments of the present application provide a method and equipment for allocating network resources to solve the existing cloud server technology, virtual machine rental and real-time computing services are to respond to user services through two independent server clusters
  • the problem is that the resource utilization rate of the system is low and cannot meet the service needs of users.
  • a first aspect of the embodiments of the present application provides a method for allocating network resources, including:
  • the actual occupation ratio is set as the initial allocation ratio, and the first resource area is adjusted based on the actual occupation ratio And the second resource area.
  • all active servers in the target service system are divided into multiple server clusters, and a corresponding resource regulator is configured for each server cluster. Then, the resource regulator can be used The hardware resources are divided. During operation, the resource regulator can collect the actual occupancy rate of each resource area, and determine the actual occupancy rate according to the first occupancy rate and the second occupancy rate, so as to determine whether the current resource allocation is reasonable. If the occupancy ratio is inconsistent with the initial allocation ratio, dynamic adjustment is performed, thereby achieving the purpose of dynamically allocating network resources.
  • the server used for virtual host rental and the server used for real-time cloud computing are deployed in the same server cluster, and even the same server can respond to two different service requests and pass resources.
  • the regulator manages the hardware resources in the server cluster and dynamically allocates the two service resource areas, which improves resource utilization.
  • FIG. 1 is an implementation flowchart of a network resource allocation method provided in the first embodiment of the present application
  • FIG. 2 is a specific implementation flowchart of a network resource allocation method S104 provided in a second embodiment of the present application;
  • FIG. 3 is a specific implementation flowchart of a network resource allocation method S101 provided in a third embodiment of the present application.
  • FIG. 4 is a flowchart of a specific implementation of a network resource allocation method provided in a fourth embodiment of the present application.
  • FIG. 5 is a flowchart of a specific implementation of a network resource allocation method provided in a fifth embodiment of the present application.
  • FIG. 6 is a structural block diagram of a network resource allocation device according to an embodiment of the present application.
  • FIG. 7 is a schematic diagram of a terminal device according to another embodiment of the present application.
  • the execution subject of the process is a terminal device.
  • the terminal device includes but is not limited to: a server, a computer, a smart phone, a tablet computer, and other devices capable of performing network resource allocation operations.
  • the network resource allocation device may be a host device of a business service system, used to manage the operation and abnormal identification of all servers in the business service system, and manage the topology structure of the business service system.
  • FIG. 1 shows an implementation flowchart of the method for allocating network resources provided by the first embodiment of the present application, which is described in detail as follows:
  • S101 a list of available servers of the target service system is obtained, all the servers in use in the list of available servers are divided into multiple server clusters, and resource regulators are configured for each of the server clusters.
  • the business service system includes a plurality of active servers.
  • the business service system can divide the active servers according to geographic blocks, that is, the user terminal sends The service request will carry the location information of the user terminal, and then the gateway device will determine the corresponding active server based on the location information, and redirect the service request again, so that it can be directed to the geographic block for processing Server in use; of course, the business service system can also distribute service requests based on the load of each server in use.
  • the business service system can obtain the current operating status of each server in a preset detection period, and add the active servers that are running normally and have idle hardware resources to the list of available servers, which means that the active servers in the available server list can be used In response to the service request initiated by the user terminal.
  • the terminal device can obtain the list of available servers from the database of the target service system, so that all the active servers are used based on the target service system's preset division method for the servers in use
  • the server is divided into multiple server clusters, and each server cluster contains at least one active server.
  • the terminal device can integrate the active servers belonging to the same geographical block or adjacent geographical blocks into the same server based on the geographical block Within the cluster. If the target service system divides the active server based on the load, the terminal device may be divided into multiple server clusters according to the current load of each active server. If the target service system distributes business requests to different servers in use based on the type of service, the terminal device can select one server for each type of response and integrate them into a server cluster, so that the server cluster can respond to all services Type of business request.
  • the terminal device can be used to manage network resources of multiple different business service systems. If a business service system meets the allocation trigger condition, the terminal device will recognize the business service system as the target service system, and The in-use server in the target service system performs network resource allocation operation. For example, if a business service system performs a capacity expansion operation and a new server in use is added, the terminal device will re-divide the network resources of the server in use in the business service system. For another example, the terminal device will be provided with a detection period, and if the terminal device detects that the time node of the preset detection period is reached at the current moment, the operation of S1 is triggered.
  • the terminal device configures a resource regulator for each server cluster.
  • the resource adjuster may be an independent physical device or a virtual device. If the resource regulator is a virtual device, the terminal device can select an active server as the target server in the server cluster, for example, according to the hardware resource parameters of each active server, determine an active server with the best processing power score As the target server. The terminal device will control the target server and create a sub-thread under the main thread of the target server, and set the sub-thread as a virtual device for running the resource adjustment process based on a preset resource adjustment program to complete the resource regulator Configuration operations.
  • the resource regulator is invoked to collect the hardware resource amount of each active server in the server cluster, and establish a hardware resource pool of the server cluster.
  • the terminal device in order to facilitate the allocation of network resources of the server cluster, the terminal device will recognize the hardware resources contained in the server cluster as a whole, so that the terminal device only needs to integrate the hardware resource pool as a unified whole Dividing resources without dividing resources on a single server can improve management efficiency. Based on this, the terminal device needs to obtain the hardware resource amount of each active server, configure a resource identification code for the server cluster for each active server hardware amount, and establish an addressing correspondence table of the resource identification code, so that The active server corresponding to the resource identification code and the resource address of the active server are determined according to the addressing correspondence table. The terminal device recognizes the hardware resource composed of all resource identification codes as a hardware resource pool about the server cluster.
  • the hardware resources include resources related to hardware processing capabilities such as memory resources, bandwidth resources, thread resources, etc. Since the above hardware resources can all be occupied online by other user terminals, that is, the availability of the business service system Internet resources.
  • the terminal device can configure corresponding hardware resource pools for different types of hardware resources.
  • the hardware resource pool is divided into a first resource area for responding to virtual host lease services and a second resource area for responding to real-time cloud computing services.
  • the virtual hosting service is mainly a rental service, that is, the user can select a target server in the target service system, create a virtual host in the target server, and set up a corresponding system, which will set resource requirements when applying , And during the lease period, the resources of the virtual host will be occupied all the time, unable to respond to other services, and it is an occupation method that cannot be time-division multiplexed.
  • the real-time cloud computing service is that the user uploads the big data analysis, program code, test examples, etc. that he needs to run to the server of the cloud service, runs the above objects through the server of the cloud service, and outputs the corresponding operation results, through OBS, EFS Or the database to store the operation results.
  • the hardware resources of the server cluster need to be divided, that is, each server cluster in the cloud service system can respond to both virtual host services and real-time cloud computing services.
  • the two types of services are physically Unified management, which facilitates the flexible allocation of resources.
  • the terminal device is configured with an initial allocation ratio, that is, the ratio between the network resources used to respond to the cloud computing service and the network resources used to respond to the virtual host lease service.
  • the initial allocation ratio can be manually configured by the user. Then, the terminal device may divide the hardware resources of the hardware resource pool according to the initial allocation ratio to obtain the first resource area and the second resource area.
  • the terminal device may obtain historical business records, count the number of virtual host services and cloud computing service services based on the historical business records, and determine the average amount of resources occupied by the virtual host service and the average amount of resources consumed by the cloud computing service, Based on the above four parameters, that is, the number of virtual host leases, the average amount of resources occupied by virtual host leases, the number of services of cloud computing services, and the average amount of resources occupied by cloud computing services, the initial ratio is calculated.
  • the terminal device may import each historical service record into the adjusted RNN neural network, and determine the initial ratio through the neural network.
  • the historical business record includes the service type, that is, it is used to determine whether the historical business record belongs to a virtual host rental service or a cloud computing service, service duration, resource consumption, etc., so that the neural network can predict the current target service system
  • the ratio of the two services improves the accuracy of the initial distribution ratio.
  • the first occupancy rate of the first resource area and the second occupancy rate of the second resource area are collected by the resource adjuster, and the actual occupancy is calculated according to the first occupancy rate and the second occupancy rate proportion.
  • the terminal device divides the hardware resource pool, that is, the type of service used by the hardware resources in the corresponding area of each server has been fixed.
  • the active server can be used to respond to cloud leasing services and cloud computing services, the specific service type of the active server response can be based on local hardware resources Determined by the properties of
  • the terminal device may receive the resource regulator feedback on the first occupancy rate of the first resource area and the second occupancy rate of the second resource area, so that it can be determined in real time whether the initial allocation ratio is reasonable, and then according to the first The occupancy rate and the second occupancy rate can determine the actual occupancy rate for responding to the two types of services. If the absolute value of the difference between the actual occupancy ratio and the initial ratio is less than or equal to the preset adjustment threshold, it means that it is closer to the preset allocation ratio and is within the acceptable floating range.
  • the actual occupation ratio is set to the initial allocation ratio, and the adjustment is performed based on the actual occupation ratio The first resource area and the second resource area.
  • the terminal device when the terminal device detects that the absolute value of the difference between the actual occupancy ratio and the initial allocation ratio in a certain server cluster is greater than the preset adjustment threshold, it is necessary to reset the hardware of the hardware resource pool in the server cluster
  • the resources are re-divided, and the operation of re-dividing is as follows: the terminal device will determine the amount of hardware resources to be adjusted according to the actual occupancy ratio, and detect the amount of idle hardware resources contained in the first resource area and the second resource area, based on the adjusted hardware For the amount of resources and the amount of idle hardware resources, adjust the amount of resources contained in the two resource areas so that the ratio between the adjusted amount of hardware resources in the first resource area and the amount of hardware resources in the second resource area meets the actual occupancy ratio.
  • the terminal device sets the actual occupation ratio obtained this time to the initial allocation ratio, thereby enabling To achieve the purpose of dynamically adjusting the initial allocation ratio.
  • a network resource allocation method provided by an embodiment of the present application divides all active servers in a target service system into multiple server clusters, and configures a corresponding resource regulator for each server cluster, and then The resource regulator can divide the hardware resources in the server cluster according to the initial allocation ratio.
  • the resource regulator can collect the actual occupancy rate of each resource area, and determine the actual occupancy rate according to the first occupancy rate and the second occupancy rate. Occupancy ratio to determine whether the current resource allocation is reasonable. If the actual occupation ratio is not consistent with the initial allocation ratio, dynamic adjustment is performed to achieve the purpose of dynamically allocating network resources.
  • the server used for virtual host rental and the server used for real-time cloud computing are deployed in the same server cluster, and even the same server can respond to two different service requests and pass resources.
  • the regulator manages the hardware resources in the server cluster and dynamically allocates the two service resource areas, which improves resource utilization.
  • FIG. 2 shows a specific implementation flowchart of a network resource allocation method S104 provided in the second embodiment of the present application.
  • a network resource allocation method S104 provided in this embodiment includes: S1041 to S1044, and specific details are as follows:
  • first occupancy rate of the first resource area and the second occupancy rate of the second resource area are collected by the resource adjuster, and the actual occupancy rate is calculated according to the first occupancy rate and the second occupancy rate Occupation ratio, including
  • the terminal device is configured with a collection condition of actual resource occupation parameters. If it is detected that the current time or the current network status of the target service system meets the preset collection condition, the relevant operation of S1041 is performed.
  • the collection condition may be a time condition, that is, the terminal device is configured with a collection period or multiple collection nodes. If it is detected that the current time meets the time condition, it will trigger the collection process of the actual resource occupancy parameter to determine whether the Adjust the distribution ratio.
  • the collection condition may be actual occupancy, that is, the terminal device sets multiple resource amount nodes according to the allocated resource amounts of the first resource area and the second resource area, if the actual occupancy of a certain resource area is detected at the current moment When the amount reaches the corresponding resource amount node, the related operation of S1041 is performed.
  • the administrator can perform the related operations of S1041 by sending manual triggers such as collection instructions.
  • the terminal device when it is determined that the current collection conditions are met, the terminal device can obtain the actual resource occupancy parameters of all active servers in its corresponding cluster through the resource regulators of each server cluster, so as to realize multi-threaded concurrent collection and improve the actual Collection efficiency of resource occupation parameters.
  • the terminal device can control the resource regulator to each active server with a preset collection frequency Acquiring multiple real-time resource occupancy parameters can improve the accuracy of real-time resource occupancy parameters and reduce the impact of instantaneous floating.
  • the resources occupied by the virtual host rental service and real-time cloud computing service are allocated based on the hardware resource pool of the entire server cluster, and the hardware resource allocation of each active server in the specific cluster is transparent to the terminal device, so In order to count the real-time resource allocation of the entire cluster, the terminal device will detect the real-time resource amount occupied by the active server with respect to the two services when collecting the actual resource amount occupied by the current time, so that all active servers in the cluster can be The actual amount of occupied resources determines the actual occupied proportion of the entire cluster.
  • the first occupancy mean value and the first occupancy standard deviation of the first resource area are calculated according to the multiple first occupancy parameters of each of the active servers.
  • the second occupancy mean value and the second occupancy standard deviation of the second resource area are calculated according to the plurality of second occupancy parameters of each of the active servers.
  • the terminal device obtains the actual resource occupancy rate of each active server in the server cluster in multiple collection cycles, so the collection can be determined according to the first occupation parameter of all active servers in the same collection cycle
  • the period refers to the total actual occupancy parameter of the first resource area.
  • the terminal device can calculate the first occupancy mean value and the first occupancy standard deviation corresponding to the first resource area based on multiple different collection periods.
  • the second resource area can also be calculated in the above manner, which will not be repeated here.
  • the first occupancy mean value, the first occupancy label difference, the second occupancy mean value, and the second occupancy standard deviation are imported into an actual occupancy rate calculation model to determine the actual occupancy ratio;
  • the actual occupancy calculation model is as follows:
  • ActualRate is the actual occupancy ratio
  • Is the first occupancy mean
  • ⁇ VM is the first occupancy standard deviation
  • Is the second occupied mean
  • ⁇ Count is the second occupied standard deviation
  • VMmax is the amount of hardware resources in the first resource area
  • Countmax is the amount of hardware resources in the second resource area.
  • the terminal device imports the four parameters calculated above into the calculation model of the actual occupancy rate, and determines the actual occupancy rate calculated for this acquisition operation. Since the actual occupancy rate not only considers the first occupancy mean and the second occupancy mean, it also introduces an occupancy standard deviation, thereby reducing the impact of the floating rate and improving the accuracy of the actual occupancy rate.
  • FIG. 3 shows a specific implementation flowchart of a network resource allocation method S101 provided in the third embodiment of the present application.
  • a network resource allocation method S101 provided in this embodiment includes: S1011 to S1013, and specific details are as follows:
  • the dividing all the servers in use in the available server list into a plurality of server clusters, and configuring a resource regulator for each of the server clusters includes:
  • each of the active servers is acquired, and each of the active servers is marked on a preset map interface according to the installation location.
  • the terminal device may divide the server cluster based on the installation location of each active server.
  • the target service system can distribute the business request to the server cluster of the geographic block according to the geographic block described by the user terminal, thereby reducing the jump of the intermediate route, thereby increasing the rate of business response.
  • the server in use can be configured with a positioning module
  • the local installation position can be obtained through the positioning module, and the installation position can be fed back to the terminal device.
  • the installation location can be determined based on the gateway address according to the gateway address where the active server is located.
  • the terminal device marks each active server on a preset map interface according to the installation location, so as to divide the active server into clusters.
  • the terminal device can call the API interface of the third-party map application, output the interface of the third-party map application on the local display module, and mark each active server, thereby eliminating the need to rewrite the map program and reducing the amount of development required.
  • a traversal frame is fetched on the map interface through a preset cluster window, and all active servers in the same cluster window are identified as belonging to the same server cluster.
  • the terminal device can frame the active server through the cluster window, so that the distance between the installation positions of the servers in the same server cluster will be less than or equal to the preset Distance threshold.
  • the terminal device can control the cluster window to slide on the map interface to implement traversal frame selection, and the active servers in the same cluster window are identified as the same server cluster.
  • the terminal device configures a resource regulator for each server cluster.
  • the terminal device divides the server cluster according to the installation location of each in-use server, so that the subsequent operation of allocating service requests can be improved, and the efficiency of allocating service requests can be improved.
  • FIG. 4 shows a specific implementation flowchart of a network resource allocation method provided in a fourth embodiment of the present application.
  • the actual occupancy ratio is calculated according to the first occupancy rate and the second occupancy rate After that, it also includes: S401 ⁇ S402, detailed as follows:
  • the initial distribution ratio and the actual occupation ratio obtained at each acquisition time are used as training samples, and the training samples are imported into a multi-layer feedforward neural network to calculate the expected distribution ratio.
  • the terminal device is preset with a multi-layer feedforward RNN neural network, which can improve the accuracy of the initial allocation ratio according to the preset initial allocation ratio and the actual occupation ratio of each cycle. Therefore, the terminal equipment will be based on The initial distribution ratio and the actual occupation ratio generate multiple training samples, and import multiple training samples into the RNN neural network to calculate the expected distribution ratio. If a new server cluster is detected or a capacity expansion request is received, the hardware resource pool can be divided according to the expected allocation ratio.
  • the expected distribution ratio is also different for different server clusters, so that it can be compared with the service request situation of the user terminal of the server cluster match.
  • the terminal device can determine the first expansion resource amount for the cloud host leasing service and the second expansion resource amount for the real-time cloud computing service according to the expected allocation ratio and the hardware resource amount of the expansion server, and based on the above Two capacity expansion resources divide the hardware resources.
  • the expected allocation ratio is predicted according to the initial allocation ratio and multiple real-time occupancy ratios, and the expansion operation period is divided based on the expected allocation ratio, thereby improving the accuracy of the division operation.
  • FIG. 5 shows a specific implementation flowchart of a network resource allocation method provided in a fifth embodiment of the present application.
  • a network resource allocation method provided in this embodiment further includes: S501-S502, which are described in detail as follows:
  • a resource report instruction is broadcast to each of the resource regulators, so that each of the resource regulators collects the corresponding resource occupancy rate of the server cluster.
  • the terminal device in order to determine whether each server cluster needs to be expanded, the terminal device sends a resource reporting instruction to each resource regulator, and collects information about resource occupancy in the server cluster to which it belongs through each resource regulator.
  • the terminal device may be set with triggering conditions, such as conditional triggering and event triggering, and if the preset triggering conditions are currently met, the relevant operation of S501 is performed.
  • the terminal device detects that the resource occupancy rate is less than or equal to the preset capacity expansion threshold, it means that the network resources of the server cluster are not in a saturated state and can continue to respond to the service request of the user terminal; otherwise, if the If the resource occupancy rate is greater than the capacity expansion threshold, it indicates that the server is saturated, and a capacity expansion reminder message is generated and sent to the administrator's terminal so that the administrator can expand the server cluster.
  • the terminal device detects the resource occupancy of the server cluster and automatically generates expansion expansion prompt information, thereby improving the response efficiency of expansion.
  • FIG. 6 shows a structural block diagram of a network resource allocation device according to an embodiment of the present application.
  • Each unit included in the network resource allocation device is used to execute each step in the embodiment corresponding to FIG. 1.
  • only parts related to this embodiment are shown.
  • the network resource allocation device includes:
  • the resource regulator configuration unit 61 is used to obtain an available server list of the target service system, divide all the servers in use in the available server list into multiple server clusters, and configure a resource regulator for each of the server clusters;
  • the hardware resource pool establishing unit 62 is configured to call the resource regulator to collect the hardware resource amount of each active server in the server cluster, and establish a hardware resource pool of the server cluster;
  • the hardware resource pool dividing unit 63 is configured to divide the hardware resource pool into a first resource area for responding to virtual host lease service and a second resource area for responding to real-time cloud computing service based on a preset initial allocation ratio ;
  • the actual occupancy ratio acquisition unit 64 is configured to collect the first occupancy rate of the first resource area and the second occupancy rate of the second resource area through the resource adjuster, according to the first occupancy rate and the second Occupancy rate to calculate the actual occupancy ratio;
  • the hardware resource adjustment unit 65 is configured to set the actual occupation ratio to the initial allocation ratio based on the actual occupation if the absolute value of the difference between the actual occupation ratio and the initial allocation ratio is greater than a preset adjustment threshold Proportionally adjust the first resource area and the second resource area.
  • the actual occupation ratio acquisition unit 64 includes:
  • the collection trigger unit is configured to obtain multiple actual resource occupation parameters of each of the active servers in the server cluster at a preset collection frequency if the preset collection conditions are met; the actual resource occupation parameters include: The first occupancy parameter in response to the virtual host lease service and the second occupancy parameter in response to the real-time cloud computing service;
  • a first occupancy parameter calculation unit configured to calculate the first occupancy mean value and the first occupancy standard deviation of the first resource area according to the plurality of first occupancy parameters of each of the active servers;
  • a second occupancy parameter calculation unit configured to calculate a second occupancy mean value and a second occupancy standard deviation of the second resource area based on the plurality of second occupancy parameters of each of the active servers;
  • An actual occupancy ratio calculation unit used to import the first occupancy mean, the first occupancy label difference, the second occupancy mean, and the second occupancy standard deviation into an actual occupancy rate calculation model to determine the actual occupancy Proportion; the actual occupancy rate calculation model is specifically:
  • ActualRate is the actual occupancy ratio
  • Is the first occupancy mean
  • ⁇ VM is the first occupancy standard deviation
  • Is the second occupied mean
  • ⁇ Count is the second occupied standard deviation
  • VMmax is the amount of hardware resources in the first resource area
  • Countmax is the amount of hardware resources in the second resource area.
  • the resource regulator configuration unit 61 includes:
  • An installation location acquiring unit configured to acquire the installation location of each of the active servers, and mark each of the active servers on a preset map interface according to the installation location;
  • a server cluster identification unit configured to perform traversal frame selection on the map interface through a preset cluster window, and identify all active servers in the same cluster window as belonging to the same server cluster;
  • a resource regulator creation unit is configured to configure the resource regulator for each server cluster.
  • the network resource allocation device further includes:
  • An expected allocation ratio calculation unit configured to use the initial allocation ratio and the actual occupancy ratio obtained at each acquisition time as training samples, and import the training samples into a multi-layer feedforward neural network to calculate the expected allocation ratio;
  • the capacity expansion response unit is used to determine the hardware resource volume of the capacity expansion server if a system capacity expansion instruction is received, and perform a hardware resource division operation on the hardware resource volume of the capacity expansion server according to the expected allocation ratio.
  • the network resource allocation device further includes:
  • a resource occupancy collection unit for broadcasting a resource reporting instruction to each of the resource regulators, so that each of the resource regulators collects the corresponding resource occupancy of the server cluster;
  • the capacity expansion prompt information sending unit is configured to send capacity expansion prompt information to the terminal of the administrator if any of the resource occupancy rates is greater than a preset capacity expansion threshold.
  • the server for virtual host lease and the server for real-time cloud computing are deployed in the same server cluster, and even the same server can respond to two different service requests
  • the resource regulator manages the hardware resources in the server cluster and dynamically allocates the resource areas of the two services, which improves the utilization rate of resources.
  • the terminal device 7 of this embodiment includes: a processor 70, a memory 71, and computer-readable instructions 72 stored in the memory 71 and executable on the processor 70, such as network resources Assignment procedures.
  • the processor 70 executes the computer-readable instructions 72, the steps in the above embodiments of the method for allocating network resources are implemented, for example, S101 to S105 shown in FIG. 1.
  • the processor 70 executes the computer-readable instructions 72, the functions of the units in the foregoing device embodiments are realized, for example, the functions of the modules 61 to 65 shown in FIG. 6.
  • the computer-readable instructions 72 may be divided into one or more units, and the one or more units are stored in the memory 71 and executed by the processor 70 to complete the application .
  • the one or more units may be a series of computer-readable instruction instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer-readable instructions 72 in the terminal device 7.
  • the computer-readable instructions 72 may be divided into a resource conditioner configuration unit, a hardware resource pool establishment unit, a hardware resource pool division unit, an actual occupation ratio acquisition unit, and a hardware resource adjustment unit. The specific functions of each unit are as described above.
  • the terminal device 7 may be a computing device such as a desktop computer, a notebook, a palmtop computer and a cloud server.
  • the terminal device may include, but is not limited to, a processor 70 and a memory 71.
  • FIG. 7 is only an example of the terminal device 7 and does not constitute a limitation on the terminal device 7, and may include more or fewer components than those illustrated, or a combination of certain components, or different components.
  • the terminal device may further include an input and output device, a network access device, a bus, and the like.
  • the so-called processor 70 may be a central processing unit (Central Processing Unit, CPU), or other general-purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), Ready-made programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • the general-purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
  • the memory 71 may be an internal storage unit of the terminal device 7, such as a hard disk or a memory of the terminal device 7.
  • the memory 71 may also be an external storage device of the terminal device 7, such as a plug-in hard disk equipped on the terminal device 7, a smart memory card (Smart, Media, Card, SMC), and a secure digital (SD) Cards, flash cards, etc. Further, the memory 71 may include both an internal storage unit of the terminal device 7 and an external storage device. The memory 71 is used to store the computer-readable instructions and other programs and data required by the terminal device. The memory 71 can also be used to temporarily store data that has been or will be output.
  • SD secure digital

Abstract

本申请适用于互联网技术领域,提供了一种网络资源的分配方法及设备,包括:获取目标服务系统的可用服务器列表,划分为多个服务器集群,为各个服务器集群配置资源调节器;调用资源调节器采集服务器集群内各个在用服务器的硬件资源量,建立服务器集群的硬件资源池;基于预设的初始分配比例,将硬件资源池划分为第一资源区域以及第二资源区域;根据第一占用率以及第二占用率计算实际占用比例;若两者之差的绝对值大于预设的调整阈值,则将实际占用比例设置为初始分配比例,调整第一资源区域以及第二资源区域。本申请通过资源调节器管理服务器集群内的硬件资源,并动态分配两个服务的资源区域,提高了资源的利用率。

Description

一种网络资源的分配方法及设备
本申请申明享有2019年01月08日递交的申请号为201910017051.8、名称为“一种网络资源的分配方法及设备”中国专利申请的优先权,该中国专利申请的整体内容以参考的方式结合在本申请中。
技术领域
本申请属于互联网技术领域,尤其涉及一种网络资源的分配方法及设备。
背景技术
随着云服务技术的不断发展,越来越多用户通过云服务器来实现存储区域的扩容以及本地运算能力补强等功能,例如通过租赁云服务器来搭建不同的系统,从而完成对应的功能测试,或者通过云服务器系统的快速实施运算能力来进行程序编译或压测等。在无需购置额外的硬件资源的情况下,通过云服务实现资源云共享。
现有的云服务器技术中,虚拟机租赁以及实时计算服务是通过两套独立的服务器集群来响应用户的服务请求,而由于用户的需求是动态变化的,部分时刻云服务器租赁请求较少,而该时刻实时计算服务的请求较多,上述两套系统相互独立的情况,则不利于资源分配,从而降低了系统的资源利用率,无法满足用户的服务需求。
技术问题
有鉴于此,本申请实施例提供了一种网络资源的分配方法及设备,以解决现有的云服务器技术中,虚拟机租赁以及实时计算服务是通过两套独立的服务器集群来响应用户的服务请求,系统的资源利用率低,无法满足用户的服务需求的问题。
技术解决方案
本申请实施例的第一方面提供了一种网络资源的分配方法,包括:
获取目标服务系统的可用服务器列表,将所述可用服务器列表内的所有在用服务器划分为多个服务器集群,为各个所述服务器集群配置资源调节器;
调用所述资源调节器采集所述服务器集群内各个所述在用服务器的硬件资源量,建立所述服务器集群的硬件资源池;
基于预设的初始分配比例,将所述硬件资源池划分为用于响应虚拟主机租赁服务的第一资源区域以及用于响应实时云计算服务的第二资源区域;
通过所述资源调节器采集所述第一资源区域的第一占用率以及第二资源区域的第二占用率,根据所述第一占用率以及所述第二占用率计算实际占用比例;
若所述实际占用比例与所述初始分配比例之差的绝对值大于预设的调整阈值,则将所述实际占用比例设置为初始分配比例,基于所述实际占用比例调整所述第一资源区域以及所述第二资源区域。
有益效果
本申请实施例通过将目标服务系统内的所有在用服务器划分为多个服务器集群,并为每个服务器集群配置对应的资源调节器,继而可以通过资源调节器根据初始分配比例对服务器集群内的硬件资源进行划分,在运行过程中,资源调节器可以采集各个资源区域的实际占用率,并根据第一占用率以及第二占用率确定实际占用比例,从而判断当前的资源分配是否合理,若实际占用比例与初始分配比例不一致,则进行动态调整,从而实现了对网络资源进行动态分配的目的。与现有的网络资源的分配方式相比,用于虚拟主机租赁的服务器以及用于实时云计算的服务器布放于同一服务器集群内,甚至同一服务器也能够响应两种不同的服务请求,通过资源调节器管理服务器集群内的硬件资源,并动态分配两个服务的资源区域,提高了资源的利用率。
附图说明
图1是本申请第一实施例提供的一种网络资源的分配方法的实现流程图;
图2是本申请第二实施例提供的一种网络资源的分配方法S104具体实现流程图;
图3是本申请第三实施例提供的一种网络资源的分配方法S101具体实现流程图;
图4是本申请第四实施例提供的一种网络资源的分配方法具体实现流程图;
图5是本申请第五实施例提供的一种网络资源的分配方法具体实现流程图;
图6是本申请一实施例提供的一种网络资源的分配设备的结构框图;
图7是本申请另一实施例提供的一种终端设备的示意图。
本发明的实施方式
在本申请实施例中,流程的执行主体为终端设备。该终端设备包括但不限于:服务器、计算机、智能手机以及平板电脑等能够执行网络资源的分配操作的设备。特别地,该网络资源的分配设备可以为一业务服务系统的上位机设备,用于管理该业务服务系统内所有服务器的运行以及异常识别等,对该业务服务系统的拓扑结构进行管理。图1示出了本申请第一实施例提供的网络资源的分配方法的实现流程图,详述如下:
在S101中,获取目标服务系统的可用服务器列表,将所述可用服务器列表内的所有在用服务器划分为多个服务器集群,为各个所述服务器集群配置资源调节器。
在本实施例中,业务服务系统内包含多个在用服务器,通过在用服务器响应各个用户 终端发起的业务请求,业务服务系统可以根据地理区块来对在用服务器进行划分,即用户终端发送业务请求时会携带有该用户终端所处的位置信息,继而网关设备会基于该位置信息确定对应的在用服务器,重新对该业务请求进行重定向,从而能够定向发送给该地理区块所处理的在用服务器;当然,业务服务系统还可以根据各个在用服务器的负载情况来分配服务请求。业务服务系统可以以预设的检测周期获取各个服务器当前的运行状态,将正常运行且存在空闲硬件资源的在用服务器添加到可用服务器列表内,即表示该可用服务器列表内的在用服务器可以用于响应用户终端发起的业务请求。
在本实施例中,终端设备,即硬件资源的分配设备,可以从目标服务系统的数据库内获取得到该可用服务器列表,从而基于目标服务系统对于在用服务器的预设划分方式,将所有在用服务器划分为多个服务器集群,每个服务器集群中至少包含一个在用服务器。如上所述,若目标服务器系统是基于地理区块对在用服务器进行划分的,则终端设备可以基于地理区块将属于同一地理区块或相邻地理区块的在用服务器集成于同一个服务器集群内。若目标服务系统是基于负载情况对在用服务器进行划分的,则终端设备可以根据各个在用服务器当前的负载情况划分为多个服务器集群。若目标服务系统是基于服务类型将业务请求分配给不同在用服务器的,则终端设备可以将用于响应不同类型的服务器各选一个,集成于一个服务器集群内,从而该服务器集群可以响应所有服务类型的业务请求。
在本实施例中,终端设备可以用于管理多个不同业务服务系统的网络资源,若某一业务服务系统满足分配触发条件,则终端设备会对该业务服务系统识别为目标服务系统,并对该目标服务系统内的在用服务器进行网络资源的分配操作。例如,若某一业务服务系统进行扩容操作,则加入了新的在用服务器,则终端设备会对该业务服务系统内的在用服务器重新进行网络资源划分。又例如,终端设备会设置有一检测周期,若终端设备检测到当前时刻到达预设的检测周期的时间节点,则触发S1的操作。
在本实施例中,为了便于对不同集群进行并发管理,从而提高网络资源的管理及时性以及准确性,终端设备会为每个服务器集群配置一个资源调节器。该资源调节器可以为一独立的实体装置,也可以为一虚拟装置。若该资源调节器为一虚拟装置,则终端设备可以在服务器集群中选取一个在用服务器作为目标服务器,例如根据各个在用服务器的硬件资源参数,确定出处理能力评分最优的一个在用服务器作为目标服务器。终端设备会控制该目标服务器,并在该目标服务器的主线程下创建一条子线程,并基于预设的资源调节程序将该子线程设置为用于运行资源调节进程的虚拟装置,完成资源调节器的配置操作。
在S102中,调用所述资源调节器采集所述服务器集群内各个所述在用服务器的硬件资源量,建立所述服务器集群的硬件资源池。
在本实施例中,终端设备在为了方便对服务器集群的网络资源进行分配,会将该服务器集群内所包含的硬件资源识别为一个整体,从而终端设备只需对硬件资源池这一统一的整体进行资源划分,而无需对单一的服务器进行资源划分,能够提高管理的效率。基于此,终端设备需要获取各个在用服务器的硬件资源量,为每个在用服务器的硬件资源量配置一个关于服务器集群的资源标识码,并建立一个资源标识码的寻址对应表,从而可以根据寻址对应表确定资源标识码所对应的在用服务器以及在该在用服务器的资源地址。终端设备将所有资源标识码所构成的硬件资源识别为关于该服务器集群的硬件资源池。
在本实施例中,该硬件资源包括有内存资源、带宽资源、线程资源等与硬件处理能力相关的资源,由于上述硬件资源均可以被其他用户终端进行在线占用,即属于该业务服务系统的可用网络资源。终端设备可以为不同类型的硬件资源配置对应的硬件资源池。
在S103中,基于预设的初始分配比例,将所述硬件资源池划分为用于响应虚拟主机租赁服务的第一资源区域以及用于响应实时云计算服务的第二资源区域。
在本实施例中,虚拟主机服务主要为租赁服务,即用户可以在目标服务系统中选取一个目标服务器,在该目标服务器中创建一个虚拟主机,并搭建相应的系统,在申请时会设置资源需求,并在租赁期内,该虚拟主机的资源将被一直占用,无法响应其他服务,是无法时分复用的占用方式。而实时云计算服务,是用户将所需运行的大数据分析、程序代码、测试实例等上传至云服务的服务器,通过云服务的服务器运行上述对象,并输出对应的运行结果,通过OBS、EFS或数据库等方式将运行结果进行存储,运行完毕后,所占用的资源将会被释放,是一个可时分复用的资源占用方式。基于上述两类不同的服务类型,需要将服务器集群的硬件资源进行划分,即云服务系统内的各个服务器集群既可响应虚拟主机服务又可以响应实时云计算服务,两类服务在物理层面上是统一管理的,从而便于实现资源的弹性分配。
在本实施例中,终端设备配置有一初始分配比例,即用于响应响应云计算服务的网络资源与用于响应虚拟主机租赁服务的网络资源之间的比值。该初始分配比例可以由用户手动进行配置。继而终端设备可以根据该初始分配比例,将硬件资源池的硬件资源进行划分,得到第一资源区域以及第二资源区域。
可选地,终端设备可以获取历史业务记录,基于历史业务记录统计虚拟主机的服务次数以及云计算服务的服务次数,并确定虚拟主机服务的平均占用资源量以及云计算服务的平均占用资源量,基于上述四个参数,即虚拟主机租赁的次数、虚拟主机租赁的平均占用资源量、云计算服务的服务次数以及云计算服务的平均占用资源量,计算初始比例。
可选地,终端设备可以将各个历史业务记录导入到调整好的RNN神经网络中,通过神 经网络确定该初始比例。其中,历史业务记录包括有服务类型,即用于确定该历史业务记录属于虚拟主机租赁服务抑或是云计算服务,服务时长、资源占用量等信息,从而神经网络可以预测到当前时刻目标服务系统中两个服务的比例,提高初始分配比例的准确性。
在S104中,通过所述资源调节器采集所述第一资源区域的第一占用率以及第二资源区域的第二占用率,根据所述第一占用率以及所述第二占用率计算实际占用比例。
在本实施例中,终端设备对硬件资源池进行划分后,即每个服务器内对应区域的硬件资源用于响应何种服务类型已经固定了,当然,若某一服务器中部分硬件资源被划分到第一占用区域,而另一部分的硬件资源被划分至第二占用区域,则该在用服务器可以用于响应云租赁服务以及云计算服务,具体在用服务器响应的服务类型可以基于本地的硬件资源的属性所决定。
在本实施例中,终端设备可以接收资源调节器反馈关于第一资源区域的第一占用率以及关于第二资源区域的第二占用率,从而可以实时确定初始分配比例是否合理,继而根据第一占用率以及第二占用率即可以确定用于响应两个类型服务的实际占用比例。若该实际占用比例与初始比例之间的差值的绝对值小于或等于预设的调整阈值,则表示与预设分配比例较为接近,在可接受的浮动范围内,此时,则无需调整两个网络资源的分配比例,等待下一个采集时刻的到达,再进行调整判断;反之,若两个之间的差值的绝对值大于预设的调整阈值,则需要对硬件资源池进行重新分配,执行S105的相关操作。
在S105中,若所述实际占用比例与所述初始分配比例之差的绝对值大于预设的调整阈值,则将所述实际占用比例设置为初始分配比例,基于所述实际占用比例调整所述第一资源区域以及所述第二资源区域。
在本实施例中,终端设备在检测到某一服务器集群中实际占用比例与初始分配比例之差的绝对值大于预设的调整阈值时,则需要重新对该服务器集群中的硬件资源池的硬件资源进行重新划分,重新划分的操作如下:终端设备会根据实际占用比例确定所需调整的硬件资源量,并检测第一资源区域以及第二资源区域中包含的空闲硬件资源量,基于调整的硬件资源量以及空闲的硬件资源量,调整两个资源区域所包含的资源量,以使调整的第一资源区域的硬件资源量与第二资源区域的硬件资源量之间的比值满足实际占用比例。
在本实施例中,为了在下一次检测周期到达时能够避免重复调整以及后续扩容操作时能够合理地进行初始资源分配,终端设备会将本次获取得到的实际占用比例设置为初始分配比例,从而能够实现动态调整初始分配比例的目的。
以上可以看出,本申请实施例提供的一种网络资源的分配方法通过将目标服务系统内的所有在用服务器划分为多个服务器集群,并为每个服务器集群配置对应的资源调节器, 继而可以通过资源调节器根据初始分配比例对服务器集群内的硬件资源进行划分,在运行过程中,资源调节器可以采集各个资源区域的实际占用率,并根据第一占用率以及第二占用率确定实际占用比例,从而判断当前的资源分配是否合理,若实际占用比例与初始分配比例不一致,则进行动态调整,从而实现了对网络资源进行动态分配的目的。与现有的网络资源的分配方式相比,用于虚拟主机租赁的服务器以及用于实时云计算的服务器布放于同一服务器集群内,甚至同一服务器也能够响应两种不同的服务请求,通过资源调节器管理服务器集群内的硬件资源,并动态分配两个服务的资源区域,提高了资源的利用率。
图2示出了本申请第二实施例提供的一种网络资源的分配方法S104的具体实现流程图。参见图2,相对于图1所述实施例,本实施例提供的一种网络资源的分配方法S104包括:S1041~S1044,具体详述如下:
进一步地,所述通过所述资源调节器采集所述第一资源区域的第一占用率以及第二资源区域的第二占用率,根据所述第一占用率以及所述第二占用率计算实际占用比例,包括
在S1041中,若满足预设的采集条件,则以预设的采集频率获取所述服务器集群内各个所述在用服务器的多个实际资源占用参数;所述实际资源占用参数包括:用于响应虚拟主机租赁服务的第一占用参量以及用于响应实时云计算服务的第二占用参量。
在本实施例中,终端设备配置有实际资源占用参数的采集条件,若检测到当前时刻或当前目标服务系统的网络状态满足该预设的采集条件,则执行S1041的相关操作。具体地,该采集条件可以为一时间条件,即终端设备配置有一采集周期或多个采集节点,若检测到当前时刻满足该时间条件,则会触发实际资源占用参数的采集流程,判断是否需要对分配比例进行调整。可选地,该采集条件可以为实际占用量,即终端设备根据第一资源区域以及第二资源区域分配的资源量设置有多个资源量节点,若检测到当前时刻某一资源区域的实际占用量到达对应的资源量节点则执行S1041的相关操作。管理员可以通过发送采集指令等手动触发的方式,来执行S1041的相关操作。
在本实施例中,终端设备在确定了当前满足采集条件时,可以通过各个服务器集群的资源调节器获取其对应集群内所有在用服务器的实际资源占用参数,从而实现多线程并发采集,提高实际资源占用参数的采集效率。另一方面,为了提高采集的准确性,特别对于实时云计算服务,其瞬时资源占用率的浮动范围较大,因此,终端设备可以控制资源调节器以预设的采集频率对每个在用服务器获取多个实时资源占用参数,从而能够提高实时资源占用参数的准确性,减少因瞬时浮动而带来的影响。
由于虚拟主机租赁服务以及实时云计算服务所占用的资源是基于整个服务器集群的硬件资源池进行分配的,而具体集群里每个在用服务器的硬件资源分配对于终端设备而言是 透明的,因此为了统计整个集群的实时资源分配情况,终端设备在采集当前时刻的实际占用资源量时,会检测该在用服务器关于两种服务分别占用的实时资源量,从而可以通过集群里所有在用服务器的实际占用资源量确定出整个集群的实际占用比例。
在S1042中,根据各个所述在用服务器的多个所述第一占用参量,计算所述第一资源区域的第一占用均值以及第一占用标准差。
在S1043中,根据各个所述在用服务器的多个所述第二占用参量,计算所述第二资源区域的第二占用均值以及第二占用标准差。
在本实施例中,终端设备获取了关于服务器集群内各个在用服务器在多个采集周期的实际资源占用率,因此可以根据所有在用服务器在同一采集周期内的第一占用参量,确定该采集周期关于第一资源区域的总的实际占用参量。同样地,对于各个采集周期都可以将所有在用服务器在该周期内的第一占用参量进行叠加,从而求出关于该周期内第一资源区域的总的实际占用参量。因此,终端设备可以计算出第一资源区域在基于多个不同的采集周期对应的第一占用均值以及第一占用标准差。对于第二资源区域也可以通过上述方式进行计算,在此不再赘述。
在S1044中,将所述第一占用均值、所述第一占用标注差、所述第二占用均值以及所述第二占用标准差导入实际占用率计算模型,确定所述实际占用比例;所述实际占用率计算模型具体为:
Figure PCTCN2019091499-appb-000001
其中,ActualRate为所述实际占用比例;
Figure PCTCN2019091499-appb-000002
为第一占用均值;ξ VM为第一占用标准差;
Figure PCTCN2019091499-appb-000003
为第二占用均值;ξ Count为第二占用标准差;VMmax为第一资源区域的硬件资源量;Countmax为第二资源区域的硬件资源量。
在本实施例中,终端设备将上述计算得到的四个参量导入到实际占用率的计算模型内,确定出关于本次采集操作计算得到实际占用比例。由于该实际占用率不仅考虑了第一占用均值以及第二占用均值,还引入了占用标准差,从而减少浮动率而带来的影响,提高实际占用率的准确性。
在本申请实施例中,在满足采集条件时,获取多个实际资源占用参量,从而能够通过多个实际资源占用参量计算出关于服务器集群的实际资源占用率,从而提高实际资源占用率的准确性。
图3示出了本申请第三实施例提供的一种网络资源的分配方法S101的具体实现流程图。 参见图3,相对于图1所述的实施例,本实施例提供的一种网络资源的分配方法S101包括:S1011~S1013,具体详述如下:
进一步地,所述将所述可用服务器列表内的所有在用服务器划分为多个服务器集群,为各个所述服务器集群配置资源调节器,包括:
在S1011中,获取各个所述在用服务器的安装位置,并根据所述安装位置在预设的地图界面上标记各个所述在用服务器。
在本实施例中,终端设备可以基于各个在用服务器的安装位置来进行服务器集群的划分。在该情况下,目标服务系统可以用户终端所述的地理区块,将业务请求分配到该地理区块的服务器集群,从而可以减少中间路由的跳转,从而提高业务响应的速率。
在本实施例中,若在用服务器可以配置有定位模块,则可以通过该定位模块获取本地的安装位置,并将该安装位置反馈给终端设备。若在用服务器并没有配置有定位模块,则可以根据该在用服务器所在的网关地址,基于网关地址确定安装位置。终端设备根据安装位置在预设的地图界面上标记出各个在用服务器,以便对在用服务器进行集群划分。特别地,终端设备可以调用第三方地图应用的API接口,在本地显示模块上输出第三方地图应用的界面,并标记出各个在用服务器,从而无需重新编写地图程序,减少所需的开发量。
在S1012中,通过预设的集群窗口在所述地图界面上进行遍历框取,将处于同一所述集群窗口内的所有在用服务器识别为隶属于同一所述服务器集群。
在本实施例中,终端设备根据预设的集群窗口,可以通过该集群窗口对在用服务器进行框取,从而同一服务器集群内的服务器所处的安装位置之间的距离会小于或等于预设的距离阈值。具体地,终端设备可以控制集群窗口在地图界面上进行滑动,实现遍历框取,而处于同一集群窗口内的在用服务器则识别为同一服务器集群。
在S1013中,为各个所述服务器集群配置所述资源调节器。
在本实施例中,为了便于对不同集群进行并发管理,从而提高网络资源的管理及时性以及准确性,终端设备会为每个服务器集群配置一个资源调节器。
在本申请实施例中,终端设备根据各个在用服务器的安装位置来实现对服务器集群的划分,从而能够提高后续业务请求的分配操作,提高业务请求的分配效率。
图4示出了本申请第四实施例提供的一种网络资源的分配方法的具体实现流程图。参见图4,相对于图1至图3所述实施例,本实施例提供的一种网络资源的分配方法中在所述根据所述第一占用率以及所述第二占用率计算实际占用比例之后,还包括:S401~S402,具体详述如下:
在S401中,将所述初始分配比例以及各个采集时刻获取得到的所述实际占用比例作为 训练样本,将所述训练样本导入多层前馈神经网络,计算出预期分配比例。
在本实施例中,终端设备预设有多层前馈RNN神经网路,可以根据预设的初始分配比例以及各个周期的实际占用比例来提高初始分配比例的准确性,因此,终端设备会基于初始分配比例以及实际占用比例生成多个训练样本,并将多个训练样本导入到该RNN神经网络内,计算出预期分配比例。若检测到新的服务器集群或接收到扩容请求时,可以根据该预期分配比例来对硬件资源池进行划分。
需要说明的是,由于各个服务器集群的业务请求的数量以及服务类型存在差异,即该预期分配比例对于不同的服务器集群而言也是不同的,从而能够与该服务器集群的用户终端的服务请求情况相匹配。
在S402中,若接收到系统扩容指令,则确定扩容服务器的硬件资源量,并根据所述预期分配比例对所述扩容服务器的硬件资源量执行硬件资源划分操作。
在本实施例中,终端设备若接收到系统扩容指令,即该服务器集群添加有新的扩容服务器,需要对该扩容服务器的硬件资源进行划分。因此,终端设备则可以根据预期分配比例以及该扩容服务器的硬件资源量,确定出用于云主机租赁服务的第一扩容资源量以及用于实时云计算服务的第二扩容资源量,并基于上述两个扩容资源量对硬件资源进行划分。
在本申请实施例中,根据初始分配比例以及多个实时占用比例预测出预期分配比例,并基于该预期分配比例对扩容服务期间进行划分操作,从而提高划分操作的准确率。
图5示出了本申请第五实施例提供的一种网络资源的分配方法的具体实现流程图。参见图5,相对于图1-图3所述实施例,本实施例提供的一种网络资源的分配方法还包括:S501~S502,具体详述如下:
在S501中,向各个所述资源调节器广播资源上报指令,以使各个所述资源调节器采集对应的所述服务器集群的资源占用率。
在本实施例中,终端设备为了判断各个服务器集群是否需要进行服务器扩容,会向各个资源调节器发送一个资源上报指令,通过各个资源调节器采集关于自身所属的服务器集群内的资源占用情况。终端设备可以设置有触发条件,例如条件触发以及事件触发,若当前满足预设的触发条件,则执行S501的相关操作。
在S502中,若任一所述资源占用率均大于预设的扩容阈值,则向管理员的终端发送扩容提示信息。
在本实施例中,终端设备若检测到资源占用率小于或等于预设的扩容阈值,则表示该服务器集群的网络资源并未处于饱和状态,可以继续响应用户终端的业务请求;反之,若该资源占用率大于扩容阈值,则表示该服务器已处于饱和状态,则生成一个扩容提示信息, 并发送给管理员的终端,以便管理员对服务器集群进行扩容。
在本申请实施例中,终端设备检测服务器集群的资源占用情况,并自动生成扩容提示信息,从而提高了扩容的响应效率。
应理解,上述实施例中各步骤的序号的大小并不意味着执行顺序的先后,各过程的执行顺序应以其功能和内在逻辑确定,而不应对本申请实施例的实施过程构成任何限定。
图6示出了本申请一实施例提供的一种网络资源的分配设备的结构框图,该网络资源的分配设备包括的各单元用于执行图1对应的实施例中的各步骤。具体请参阅图1与图1所对应的实施例中的相关描述。为了便于说明,仅示出了与本实施例相关的部分。
参见图6,所述网络资源的分配设备包括:
资源调节器配置单元61,用于获取目标服务系统的可用服务器列表,将所述可用服务器列表内的所有在用服务器划分为多个服务器集群,为各个所述服务器集群配置资源调节器;
硬件资源池建立单元62,用于调用所述资源调节器采集所述服务器集群内各个所述在用服务器的硬件资源量,建立所述服务器集群的硬件资源池;
硬件资源池划分单元63,用于基于预设的初始分配比例,将所述硬件资源池划分为用于响应虚拟主机租赁服务的第一资源区域以及用于响应实时云计算服务的第二资源区域;
实际占用比例获取单元64,用于通过所述资源调节器采集所述第一资源区域的第一占用率以及第二资源区域的第二占用率,根据所述第一占用率以及所述第二占用率计算实际占用比例;
硬件资源调整单元65,用于若所述实际占用比例与所述初始分配比例之差的绝对值大于预设的调整阈值,则将所述实际占用比例设置为初始分配比例,基于所述实际占用比例调整所述第一资源区域以及所述第二资源区域。
可选地,所述实际占用比例获取单元64包括:
采集触发单元,用于若满足预设的采集条件,则以预设的采集频率获取所述服务器集群内各个所述在用服务器的多个实际资源占用参数;所述实际资源占用参数包括:用于响应虚拟主机租赁服务的第一占用参量以及用于响应实时云计算服务的第二占用参量;
第一占用参量计算单元,用于根据各个所述在用服务器的多个所述第一占用参量,计算所述第一资源区域的第一占用均值以及第一占用标准差;
第二占用参量计算单元,用于根据各个所述在用服务器的多个所述第二占用参量,计算所述第二资源区域的第二占用均值以及第二占用标准差;
实际占用比例计算单元,用于将所述第一占用均值、所述第一占用标注差、所述第二 占用均值以及所述第二占用标准差导入实际占用率计算模型,确定所述实际占用比例;所述实际占用率计算模型具体为:
Figure PCTCN2019091499-appb-000004
其中,ActualRate为所述实际占用比例;
Figure PCTCN2019091499-appb-000005
为第一占用均值;ξ VM为第一占用标准差;
Figure PCTCN2019091499-appb-000006
为第二占用均值;ξ Count为第二占用标准差;VMmax为第一资源区域的硬件资源量;Countmax为第二资源区域的硬件资源量。
可选地,所述资源调节器配置单元61包括:
安装位置获取单元,用于获取各个所述在用服务器的安装位置,并根据所述安装位置在预设的地图界面上标记各个所述在用服务器;
服务器集群识别单元,用于通过预设的集群窗口在所述地图界面上进行遍历框取,将处于同一所述集群窗口内的所有在用服务器识别为隶属于同一所述服务器集群;
资源调节器创建单元,用于为各个所述服务器集群配置所述资源调节器。
可选地,所述网络资源的分配设备还包括:
预期分配比例计算单元,用于将所述初始分配比例以及各个采集时刻获取得到的所述实际占用比例作为训练样本,将所述训练样本导入多层前馈神经网络,计算出预期分配比例;
扩容操作响应单元,用于若接收到系统扩容指令,则确定扩容服务器的硬件资源量,并根据所述预期分配比例对所述扩容服务器的硬件资源量执行硬件资源划分操作。
可选地,所述网络资源的分配设备还包括:
资源占用率采集单元,用于向各个所述资源调节器广播资源上报指令,以使各个所述资源调节器采集对应的所述服务器集群的资源占用率;
扩容提示信息发送单元,用于若任一所述资源占用率均大于预设的扩容阈值,则向管理员的终端发送扩容提示信息。
因此,本申请实施例提供的网络资源的分配设备中,用于虚拟主机租赁的服务器以及用于实时云计算的服务器布放于同一服务器集群内,甚至同一服务器也能够响应两种不同的服务请求,通过资源调节器管理服务器集群内的硬件资源,并动态分配两个服务的资源区域,提高了资源的利用率。
图7是本申请另一实施例提供的一种终端设备的示意图。如图7所示,该实施例的终端设备7包括:处理器70、存储器71以及存储在所述存储器71中并可在所述处理器70 上运行的计算机可读指令72,例如网络资源的分配程序。所述处理器70执行所述计算机可读指令72时实现上述各个网络资源的分配方法实施例中的步骤,例如图1所示的S101至S105。或者,所述处理器70执行所述计算机可读指令72时实现上述各装置实施例中各单元的功能,例如图6所示模块61至65功能。
示例性的,所述计算机可读指令72可以被分割成一个或多个单元,所述一个或者多个单元被存储在所述存储器71中,并由所述处理器70执行,以完成本申请。所述一个或多个单元可以是能够完成特定功能的一系列计算机可读指令指令段,该指令段用于描述所述计算机可读指令72在所述终端设备7中的执行过程。例如,所述计算机可读指令72可以被分割成资源调节器配置单元、硬件资源池建立单元、硬件资源池划分单元、实际占用比例获取单元以及硬件资源调整单元,各单元具体功能如上所述。
所述终端设备7可以是桌上型计算机、笔记本、掌上电脑及云端服务器等计算设备。所述终端设备可包括,但不仅限于,处理器70、存储器71。本领域技术人员可以理解,图7仅仅是终端设备7的示例,并不构成对终端设备7的限定,可以包括比图示更多或更少的部件,或者组合某些部件,或者不同的部件,例如所述终端设备还可以包括输入输出设备、网络接入设备、总线等。所称处理器70可以是中央处理单元(Central Processing Unit,CPU),还可以是其他通用处理器、数字信号处理器(Digital Signal Processor,DSP)、专用集成电路(Application Specific Integrated Circuit,ASIC)、现成可编程门阵列(Field-Programmable Gate Array,FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。所述存储器71可以是所述终端设备7的内部存储单元,例如终端设备7的硬盘或内存。所述存储器71也可以是所述终端设备7的外部存储设备,例如所述终端设备7上配备的插接式硬盘,智能存储卡(Smart Media Card,SMC),安全数字(Secure Digital,SD)卡,闪存卡(Flash Card)等。进一步地,所述存储器71还可以既包括所述终端设备7的内部存储单元也包括外部存储设备。所述存储器71用于存储所述计算机可读指令以及所述终端设备所需的其他程序和数据。所述存储器71还可以用于暂时地存储已经输出或者将要输出的数据。
以上所述实施例仅用以说明本申请的技术方案,而非对其限制;尽管参照前述实施例对本申请进行了详细的说明,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申请各实施例技术方案的精神和范围,均应包含在本申请的保护范围之内。

Claims (20)

  1. 一种网络资源的分配方法,其特征在于,包括:
    获取目标服务系统的可用服务器列表,将所述可用服务器列表内的所有在用服务器划分为多个服务器集群,为各个所述服务器集群配置资源调节器;
    调用所述资源调节器采集所述服务器集群内各个所述在用服务器的硬件资源量,建立所述服务器集群的硬件资源池;
    基于预设的初始分配比例,将所述硬件资源池划分为用于响应虚拟主机租赁服务的第一资源区域以及用于响应实时云计算服务的第二资源区域;
    通过所述资源调节器采集所述第一资源区域的第一占用率以及第二资源区域的第二占用率,根据所述第一占用率以及所述第二占用率计算实际占用比例;
    若所述实际占用比例与所述初始分配比例之差的绝对值大于预设的调整阈值,则将所述实际占用比例设置为初始分配比例,基于所述实际占用比例调整所述第一资源区域以及所述第二资源区域。
  2. 根据权利要求1所述的分配方法,其特征在于,所述通过所述资源调节器采集所述第一资源区域的第一占用率以及第二资源区域的第二占用率,根据所述第一占用率以及所述第二占用率计算实际占用比例,包括:
    若满足预设的采集条件,则以预设的采集频率获取所述服务器集群内各个所述在用服务器的多个实际资源占用参数;所述实际资源占用参数包括:用于响应虚拟主机租赁服务的第一占用参量以及用于响应实时云计算服务的第二占用参量;
    根据各个所述在用服务器的多个所述第一占用参量,计算所述第一资源区域的第一占用均值以及第一占用标准差;
    根据各个所述在用服务器的多个所述第二占用参量,计算所述第二资源区域的第二占用均值以及第二占用标准差;
    将所述第一占用均值、所述第一占用标注差、所述第二占用均值以及所述第二占用标准差导入实际占用率计算模型,确定所述实际占用比例;所述实际占用率计算模型具体为:
    Figure PCTCN2019091499-appb-100001
    其中,ActualRate为所述实际占用比例;
    Figure PCTCN2019091499-appb-100002
    为第一占用均值;ξ VM为第一占用标准差;
    Figure PCTCN2019091499-appb-100003
    为第二占用均值;ξ Count为第二占用标准差;VMmax为第一资源区域的硬件资源量; Countmax为第二资源区域的硬件资源量。
  3. 根据权利要求1所述的分配方法,其特征在于,所述将所述可用服务器列表内的所有在用服务器划分为多个服务器集群,为各个所述服务器集群配置资源调节器,包括:
    获取各个所述在用服务器的安装位置,并根据所述安装位置在预设的地图界面上标记各个所述在用服务器;
    通过预设的集群窗口在所述地图界面上进行遍历框取,将处于同一所述集群窗口内的所有在用服务器识别为隶属于同一所述服务器集群;
    为各个所述服务器集群配置所述资源调节器。
  4. 根据权利要求1-3任一项所述的分配方法,其特征在于,在所述根据所述第一占用率以及所述第二占用率计算实际占用比例之后,还包括:
    将所述初始分配比例以及各个采集时刻获取得到的所述实际占用比例作为训练样本,将所述训练样本导入多层前馈神经网络,计算出预期分配比例;
    若接收到系统扩容指令,则确定扩容服务器的硬件资源量,并根据所述预期分配比例对所述扩容服务器的硬件资源量执行硬件资源划分操作。
  5. 根据权利要求1-3任一项所述的分配方法,其特征在于,还包括:
    向各个所述资源调节器广播资源上报指令,以使各个所述资源调节器采集对应的所述服务器集群的资源占用率;
    若任一所述资源占用率均大于预设的扩容阈值,则向管理员的终端发送扩容提示信息。
  6. 一种网络资源的分配设备,其特征在于,包括:
    资源调节器配置单元,用于获取目标服务系统的可用服务器列表,将所述可用服务器列表内的所有在用服务器划分为多个服务器集群,为各个所述服务器集群配置资源调节器;
    硬件资源池建立单元,用于调用所述资源调节器采集所述服务器集群内各个所述在用服务器的硬件资源量,建立所述服务器集群的硬件资源池;
    硬件资源池划分单元,用于基于预设的初始分配比例,将所述硬件资源池划分为用于响应虚拟主机租赁服务的第一资源区域以及用于响应实时云计算服务的第二资源区域;
    实际占用比例获取单元,用于通过所述资源调节器采集所述第一资源区域的第一占用率以及第二资源区域的第二占用率,根据所述第一占用率以及所述第二占用率计算实际占用比例;
    硬件资源调整单元,用于若所述实际占用比例与所述初始分配比例之差的绝对值大于预设的调整阈值,则将所述实际占用比例设置为初始分配比例,基于所述实际占用比例调整所述第一资源区域以及所述第二资源区域。
  7. 根据权利要求6所述的分配设备,其特征在于,所述实际占用比例获取单元包括:
    采集触发单元,用于若满足预设的采集条件,则以预设的采集频率获取所述服务器集群内各个所述在用服务器的多个实际资源占用参数;所述实际资源占用参数包括:用于响应虚拟主机租赁服务的第一占用参量以及用于响应实时云计算服务的第二占用参量;
    第一占用参量计算单元,用于根据各个所述在用服务器的多个所述第一占用参量,计算所述第一资源区域的第一占用均值以及第一占用标准差;
    第二占用参量计算单元,用于根据各个所述在用服务器的多个所述第二占用参量,计算所述第二资源区域的第二占用均值以及第二占用标准差;
    实际占用比例计算单元,用于将所述第一占用均值、所述第一占用标注差、所述第二占用均值以及所述第二占用标准差导入实际占用率计算模型,确定所述实际占用比例;所述实际占用率计算模型具体为:
    Figure PCTCN2019091499-appb-100004
    其中,ActualRate为所述实际占用比例;
    Figure PCTCN2019091499-appb-100005
    为第一占用均值;ξ VM为第一占用标准差;
    Figure PCTCN2019091499-appb-100006
    为第二占用均值;ξ Count为第二占用标准差;VMmax为第一资源区域的硬件资源量;Countmax为第二资源区域的硬件资源量。
  8. 根据权利要求6所述的分配设备,其特征在于,所述资源调节器配置单元包括:
    安装位置获取单元,用于获取各个所述在用服务器的安装位置,并根据所述安装位置在预设的地图界面上标记各个所述在用服务器;
    服务器集群识别单元,用于通过预设的集群窗口在所述地图界面上进行遍历框取,将处于同一所述集群窗口内的所有在用服务器识别为隶属于同一所述服务器集群;
    资源调节器创建单元,用于为各个所述服务器集群配置所述资源调节器。
  9. 根据权利要求6-8任一项所述的分配设备,其特征在于,所述资源调节器配置单元包括:
    预期分配比例计算单元,用于将所述初始分配比例以及各个采集时刻获取得到的所述实际占用比例作为训练样本,将所述训练样本导入多层前馈神经网络,计算出预期分配比例;
    扩容操作响应单元,用于若接收到系统扩容指令,则确定扩容服务器的硬件资源量,并根据所述预期分配比例对所述扩容服务器的硬件资源量执行硬件资源划分操作。
  10. 根据权利要求6-8任一项所述的分配设备,其特征在于,所述网络资源的分配设备 还包括:
    资源占用率采集单元,用于向各个所述资源调节器广播资源上报指令,以使各个所述资源调节器采集对应的所述服务器集群的资源占用率;
    扩容提示信息发送单元,用于若任一所述资源占用率均大于预设的扩容阈值,则向管理员的终端发送扩容提示信息。
  11. 一种终端设备,其特征在于,包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机可读指令,所述处理器执行所述计算机可读指令时实现如下步骤:
    获取目标服务系统的可用服务器列表,将所述可用服务器列表内的所有在用服务器划分为多个服务器集群,为各个所述服务器集群配置资源调节器;
    调用所述资源调节器采集所述服务器集群内各个所述在用服务器的硬件资源量,建立所述服务器集群的硬件资源池;
    基于预设的初始分配比例,将所述硬件资源池划分为用于响应虚拟主机租赁服务的第一资源区域以及用于响应实时云计算服务的第二资源区域;
    通过所述资源调节器采集所述第一资源区域的第一占用率以及第二资源区域的第二占用率,根据所述第一占用率以及所述第二占用率计算实际占用比例;
    若所述实际占用比例与所述初始分配比例之差的绝对值大于预设的调整阈值,则将所述实际占用比例设置为初始分配比例,基于所述实际占用比例调整所述第一资源区域以及所述第二资源区域。
  12. 根据权利要求11所述的终端设备,其特征在于,所述通过所述资源调节器采集所述第一资源区域的第一占用率以及第二资源区域的第二占用率,根据所述第一占用率以及所述第二占用率计算实际占用比例,包括:
    若满足预设的采集条件,则以预设的采集频率获取所述服务器集群内各个所述在用服务器的多个实际资源占用参数;所述实际资源占用参数包括:用于响应虚拟主机租赁服务的第一占用参量以及用于响应实时云计算服务的第二占用参量;
    根据各个所述在用服务器的多个所述第一占用参量,计算所述第一资源区域的第一占用均值以及第一占用标准差;
    根据各个所述在用服务器的多个所述第二占用参量,计算所述第二资源区域的第二占用均值以及第二占用标准差;
    将所述第一占用均值、所述第一占用标注差、所述第二占用均值以及所述第二占用标准差导入实际占用率计算模型,确定所述实际占用比例;所述实际占用率计算模型具体为:
    Figure PCTCN2019091499-appb-100007
    其中,ActualRate为所述实际占用比例;
    Figure PCTCN2019091499-appb-100008
    为第一占用均值;ξ VM为第一占用标准差;
    Figure PCTCN2019091499-appb-100009
    为第二占用均值;ξ Count为第二占用标准差;VMmax为第一资源区域的硬件资源量;Countmax为第二资源区域的硬件资源量。
  13. 根据权利要求11所述的终端设备,其特征在于,所述将所述可用服务器列表内的所有在用服务器划分为多个服务器集群,为各个所述服务器集群配置资源调节器,包括:
    获取各个所述在用服务器的安装位置,并根据所述安装位置在预设的地图界面上标记各个所述在用服务器;
    通过预设的集群窗口在所述地图界面上进行遍历框取,将处于同一所述集群窗口内的所有在用服务器识别为隶属于同一所述服务器集群;
    为各个所述服务器集群配置所述资源调节器。
  14. 根据权利要求11-13任一项所述的终端设备,其特征在于,在所述根据所述第一占用率以及所述第二占用率计算实际占用比例之后,所述处理器执行所述计算机可读指令时还实现如下步骤:
    将所述初始分配比例以及各个采集时刻获取得到的所述实际占用比例作为训练样本,将所述训练样本导入多层前馈神经网络,计算出预期分配比例;
    若接收到系统扩容指令,则确定扩容服务器的硬件资源量,并根据所述预期分配比例对所述扩容服务器的硬件资源量执行硬件资源划分操作。
  15. 根据权利要求11-13任一项所述的终端设备,其特征在于,所述处理器执行所述计算机可读指令时还实现如下步骤:
    向各个所述资源调节器广播资源上报指令,以使各个所述资源调节器采集对应的所述服务器集群的资源占用率;
    若任一所述资源占用率均大于预设的扩容阈值,则向管理员的终端发送扩容提示信息。
  16. 一种计算机非易失性可读存储介质,所述计算机非易失性可读存储介质存储有计算机可读指令,其特征在于,所述计算机可读指令被处理器执行时实现如下步骤:
    获取目标服务系统的可用服务器列表,将所述可用服务器列表内的所有在用服务器划分为多个服务器集群,为各个所述服务器集群配置资源调节器;
    调用所述资源调节器采集所述服务器集群内各个所述在用服务器的硬件资源量,建立所述服务器集群的硬件资源池;
    基于预设的初始分配比例,将所述硬件资源池划分为用于响应虚拟主机租赁服务的第一资源区域以及用于响应实时云计算服务的第二资源区域;
    通过所述资源调节器采集所述第一资源区域的第一占用率以及第二资源区域的第二占用率,根据所述第一占用率以及所述第二占用率计算实际占用比例;
    若所述实际占用比例与所述初始分配比例之差的绝对值大于预设的调整阈值,则将所述实际占用比例设置为初始分配比例,基于所述实际占用比例调整所述第一资源区域以及所述第二资源区域。
  17. 根据权利要求16所述的计算机非易失性可读存储介质,其特征在于,所述通过所述资源调节器采集所述第一资源区域的第一占用率以及第二资源区域的第二占用率,根据所述第一占用率以及所述第二占用率计算实际占用比例,包括:
    若满足预设的采集条件,则以预设的采集频率获取所述服务器集群内各个所述在用服务器的多个实际资源占用参数;所述实际资源占用参数包括:用于响应虚拟主机租赁服务的第一占用参量以及用于响应实时云计算服务的第二占用参量;
    根据各个所述在用服务器的多个所述第一占用参量,计算所述第一资源区域的第一占用均值以及第一占用标准差;
    根据各个所述在用服务器的多个所述第二占用参量,计算所述第二资源区域的第二占用均值以及第二占用标准差;
    将所述第一占用均值、所述第一占用标注差、所述第二占用均值以及所述第二占用标准差导入实际占用率计算模型,确定所述实际占用比例;所述实际占用率计算模型具体为:
    Figure PCTCN2019091499-appb-100010
    其中,ActualRate为所述实际占用比例;
    Figure PCTCN2019091499-appb-100011
    为第一占用均值;ξ VM为第一占用标准差;
    Figure PCTCN2019091499-appb-100012
    为第二占用均值;ξ Count为第二占用标准差;VMmax为第一资源区域的硬件资源量;Countmax为第二资源区域的硬件资源量。
  18. 根据权利要求16所述的计算机非易失性可读存储介质,其特征在于,所述将所述可用服务器列表内的所有在用服务器划分为多个服务器集群,为各个所述服务器集群配置资源调节器,包括:
    获取各个所述在用服务器的安装位置,并根据所述安装位置在预设的地图界面上标记各个所述在用服务器;
    通过预设的集群窗口在所述地图界面上进行遍历框取,将处于同一所述集群窗口内的 所有在用服务器识别为隶属于同一所述服务器集群;
    为各个所述服务器集群配置所述资源调节器。
  19. 根据权利要求16-18任一项所述的计算机非易失性可读存储介质,其特征在于,在所述根据所述第一占用率以及所述第二占用率计算实际占用比例之后,所述处理器执行所述计算机可读指令时还实现如下步骤:
    将所述初始分配比例以及各个采集时刻获取得到的所述实际占用比例作为训练样本,将所述训练样本导入多层前馈神经网络,计算出预期分配比例;
    若接收到系统扩容指令,则确定扩容服务器的硬件资源量,并根据所述预期分配比例对所述扩容服务器的硬件资源量执行硬件资源划分操作。
  20. 如权利要求16-18任一项所述的计算机非易失性可读存储介质,其特征在于,所述处理器执行所述计算机可读指令时还实现如下步骤:
    向各个所述资源调节器广播资源上报指令,以使各个所述资源调节器采集对应的所述服务器集群的资源占用率;
    若任一所述资源占用率均大于预设的扩容阈值,则向管理员的终端发送扩容提示信息。
PCT/CN2019/091499 2019-01-08 2019-06-17 一种网络资源的分配方法及设备 WO2020143164A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910017051.8A CN109857518B (zh) 2019-01-08 2019-01-08 一种网络资源的分配方法及设备
CN201910017051.8 2019-01-08

Publications (1)

Publication Number Publication Date
WO2020143164A1 true WO2020143164A1 (zh) 2020-07-16

Family

ID=66894238

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/091499 WO2020143164A1 (zh) 2019-01-08 2019-06-17 一种网络资源的分配方法及设备

Country Status (2)

Country Link
CN (1) CN109857518B (zh)
WO (1) WO2020143164A1 (zh)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112473141A (zh) * 2020-12-14 2021-03-12 网易(杭州)网络有限公司 一种游戏集群管理方法及装置
CN112965795A (zh) * 2021-02-23 2021-06-15 卓望数码技术(深圳)有限公司 集群调度方法、电子设备及存储介质
CN116069513A (zh) * 2023-04-04 2023-05-05 上海钐昆网络科技有限公司 成本确定方法、装置、电子设备及存储介质
CN116662020A (zh) * 2023-08-01 2023-08-29 鹏城实验室 应用服务动态管理方法、系统、电子设备及存储介质
CN117009091A (zh) * 2023-10-07 2023-11-07 浪潮(山东)计算机科技有限公司 一种资源调整方法、装置、设备及可读存储介质

Families Citing this family (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109857518B (zh) * 2019-01-08 2022-10-14 平安科技(深圳)有限公司 一种网络资源的分配方法及设备
CN110795233B (zh) * 2019-09-18 2022-04-05 北京你财富计算机科技有限公司 分布式资源配给方法、装置和电子设备
CN110750312A (zh) * 2019-10-17 2020-02-04 中科寒武纪科技股份有限公司 硬件资源配置方法、装置、云侧设备和存储介质
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
CN110995856B (zh) * 2019-12-16 2022-09-13 上海米哈游天命科技有限公司 一种服务器扩展的方法、装置、设备及存储介质
CN113315719A (zh) * 2020-02-27 2021-08-27 阿里巴巴集团控股有限公司 流量调度方法、设备、系统及存储介质
CN111538597B (zh) * 2020-04-27 2024-02-27 贝壳技术有限公司 资源配置方法、装置、计算机可读存储介质及电子设备
CN111708629B (zh) * 2020-04-30 2023-09-19 咪咕文化科技有限公司 一种资源分配方法、装置、电子设备和存储介质
CN111562985B (zh) * 2020-05-09 2024-03-22 上海商汤智能科技有限公司 资源管理方法及装置、电子设备和存储介质
CN113938392B (zh) * 2020-07-09 2023-11-14 亚信科技(南京)有限公司 资源分配方法、装置、电子设备及计算机可读存储介质
CN112367708B (zh) * 2020-10-30 2023-05-26 新华三技术有限公司 一种网络资源分配方法及装置
CN112631730A (zh) * 2020-12-30 2021-04-09 中国建设银行股份有限公司 一种模型的处理方法及装置、设备、计算机可读存储介质
CN113342885B (zh) * 2021-06-15 2022-08-12 深圳前海微众银行股份有限公司 数据导入方法、装置、设备及计算机程序产品
CN115550315B (zh) * 2022-08-10 2023-08-29 北京中关村软件园发展有限责任公司 一种基于下一代互联网的数字化云服务管理方法和系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110138019A1 (en) * 2009-12-09 2011-06-09 Electronics And Telecommunications Research Institute System and method for providing multi-layered content using plurality of servers
CN103812789A (zh) * 2013-09-18 2014-05-21 广东电网公司佛山供电局 云服务资源自动分配方法和系统
CN105912397A (zh) * 2016-03-31 2016-08-31 乐视控股(北京)有限公司 一种资源管理方法和装置
CN107205030A (zh) * 2017-05-31 2017-09-26 成都博视美达文化传播有限公司 服务器资源调度方法及系统
CN109857518A (zh) * 2019-01-08 2019-06-07 平安科技(深圳)有限公司 一种网络资源的分配方法及设备

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104714851B (zh) * 2015-03-30 2018-11-02 中国联合网络通信集团有限公司 一种实现资源分配的方法及装置
CN106656533B (zh) * 2015-10-29 2019-11-19 大唐移动通信设备有限公司 一种集群系统的负荷处理监控方法及装置
CN106959889A (zh) * 2016-01-11 2017-07-18 阿里巴巴集团控股有限公司 一种服务器资源调整的方法和装置
MX2018010803A (es) * 2016-03-10 2019-03-28 Velocity Tech Solutions Inc Sistemas y metodos para la administracion de recursos de computacion en la nube para sistemas de informacion.
CN108667859A (zh) * 2017-03-27 2018-10-16 中兴通讯股份有限公司 一种实现资源调度的方法及装置
CN107547622B (zh) * 2017-06-28 2021-10-12 新华三技术有限公司 一种资源调整方法及装置
CN107404523A (zh) * 2017-07-21 2017-11-28 中国石油大学(华东) 云平台自适应资源调度系统和方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110138019A1 (en) * 2009-12-09 2011-06-09 Electronics And Telecommunications Research Institute System and method for providing multi-layered content using plurality of servers
CN103812789A (zh) * 2013-09-18 2014-05-21 广东电网公司佛山供电局 云服务资源自动分配方法和系统
CN105912397A (zh) * 2016-03-31 2016-08-31 乐视控股(北京)有限公司 一种资源管理方法和装置
CN107205030A (zh) * 2017-05-31 2017-09-26 成都博视美达文化传播有限公司 服务器资源调度方法及系统
CN109857518A (zh) * 2019-01-08 2019-06-07 平安科技(深圳)有限公司 一种网络资源的分配方法及设备

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112473141A (zh) * 2020-12-14 2021-03-12 网易(杭州)网络有限公司 一种游戏集群管理方法及装置
CN112965795A (zh) * 2021-02-23 2021-06-15 卓望数码技术(深圳)有限公司 集群调度方法、电子设备及存储介质
CN116069513A (zh) * 2023-04-04 2023-05-05 上海钐昆网络科技有限公司 成本确定方法、装置、电子设备及存储介质
CN116069513B (zh) * 2023-04-04 2023-06-23 上海钐昆网络科技有限公司 成本确定方法、装置、电子设备及存储介质
CN116662020A (zh) * 2023-08-01 2023-08-29 鹏城实验室 应用服务动态管理方法、系统、电子设备及存储介质
CN116662020B (zh) * 2023-08-01 2024-03-01 鹏城实验室 应用服务动态管理方法、系统、电子设备及存储介质
CN117009091A (zh) * 2023-10-07 2023-11-07 浪潮(山东)计算机科技有限公司 一种资源调整方法、装置、设备及可读存储介质
CN117009091B (zh) * 2023-10-07 2023-12-19 浪潮(山东)计算机科技有限公司 一种资源调整方法、装置、设备及可读存储介质

Also Published As

Publication number Publication date
CN109857518A (zh) 2019-06-07
CN109857518B (zh) 2022-10-14

Similar Documents

Publication Publication Date Title
WO2020143164A1 (zh) 一种网络资源的分配方法及设备
CN108776934B (zh) 分布式数据计算方法、装置、计算机设备及可读存储介质
CN106776005B (zh) 一种面向容器化应用的资源管理系统及方法
WO2019001092A1 (zh) 负载均衡引擎,客户端,分布式计算系统以及负载均衡方法
CN106489251B (zh) 应用拓扑关系发现的方法、装置和系统
CN109032801B (zh) 一种请求调度方法、系统及电子设备和存储介质
US11206193B2 (en) Method and system for provisioning resources in cloud computing
US8024737B2 (en) Method and a system that enables the calculation of resource requirements for a composite application
CN105743962A (zh) 端对端数据中心性能控制
KR101882383B1 (ko) 데스크톱 가상화 환경에서의 컨테이너 자원할당 장치 및 방법
WO2015085969A1 (zh) 推荐算法优化方法、装置及系统
WO2020119060A1 (zh) 容器资源调度方法和系统、服务器及计算机可读存储介质
CN106383764A (zh) 一种数据采集方法和设备
CN110599148B (zh) 集群数据处理方法、装置、计算机集群及可读存储介质
WO2017075967A1 (zh) 在线媒体服务的带宽分配方法及系统
CN107070709B (zh) 一种基于底层numa感知的nfv实现方法
CN112000454A (zh) 一种多媒体数据的处理方法及设备
CN113608869A (zh) 任务调度方法、装置、电子设备和计算机存储介质
US9348391B1 (en) Managing resource power states in shared environments
CN111418187A (zh) 云网络中的可伸缩统计和分析机制
CN114153609A (zh) 资源控制方法及装置、电子设备、计算机可读存储介质
CN113946491A (zh) 微服务数据处理方法、装置、计算机设备及存储介质
CN112448833B (zh) 一种多管理域的通信方法和装置
WO2021051537A1 (zh) 深度学习的资源调度方法、装置及终端设备
CN113079062B (zh) 一种资源调整方法、装置、计算机设备和存储介质

Legal Events

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

Ref document number: 19909468

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19909468

Country of ref document: EP

Kind code of ref document: A1