WO2017128980A1 - 云平台中管理资源的方法和装置 - Google Patents

云平台中管理资源的方法和装置 Download PDF

Info

Publication number
WO2017128980A1
WO2017128980A1 PCT/CN2017/071274 CN2017071274W WO2017128980A1 WO 2017128980 A1 WO2017128980 A1 WO 2017128980A1 CN 2017071274 W CN2017071274 W CN 2017071274W WO 2017128980 A1 WO2017128980 A1 WO 2017128980A1
Authority
WO
WIPO (PCT)
Prior art keywords
cloud platform
application
health
isw
resources
Prior art date
Application number
PCT/CN2017/071274
Other languages
English (en)
French (fr)
Inventor
张琦
牛杰
吴佳青
Original Assignee
华为技术有限公司
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 华为技术有限公司 filed Critical 华为技术有限公司
Priority to EP17743595.5A priority Critical patent/EP3402163B1/en
Publication of WO2017128980A1 publication Critical patent/WO2017128980A1/zh
Priority to US16/047,789 priority patent/US10805385B2/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
    • H04L67/1031—Controlling of the operation of servers by a load balancer, e.g. adding or removing servers that serve requests
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44—Arrangements for executing specific programs
    • G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
    • G06F9/45533—Hypervisors; Virtual machine monitors
    • G06F9/45558—Hypervisor-specific management and integration aspects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5061—Partitioning or combining of resources
    • G06F9/5077—Logical partitioning of resources; Management or configuration of virtualized resources
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0893—Assignment of logical groups to network elements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0894—Policy-based network configuration management
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
    • H04L43/0805—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability
    • H04L43/0817—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability by checking functioning
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L65/00—Network arrangements, protocols or services for supporting real-time applications in data packet communication
    • H04L65/40—Support for services or applications
    • 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
    • H04L67/1034—Reaction to server failures by a load balancer
    • 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
    • H04L67/34—Network arrangements or protocols for supporting network services or applications involving the movement of software or configuration parameters 
    • 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
    • H04L67/51—Discovery or management thereof, e.g. service location protocol [SLP] or web services
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44—Arrangements for executing specific programs
    • G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
    • G06F9/45533—Hypervisors; Virtual machine monitors
    • G06F9/45558—Hypervisor-specific management and integration aspects
    • G06F2009/45591—Monitoring or debugging support
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44—Arrangements for executing specific programs
    • G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
    • G06F9/45533—Hypervisors; Virtual machine monitors
    • G06F9/45558—Hypervisor-specific management and integration aspects
    • G06F2009/45595—Network integration; Enabling network access in virtual machine instances
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0895—Configuration of virtualised networks or elements, e.g. virtualised network function or OpenFlow elements

Definitions

  • the present application relates to the field of information technology, and more particularly, to a method and apparatus for managing resources in a cloud platform.
  • the cloud platform In the era of cloud computing, a large number of applications are hosted on the cloud platform. In addition to ensuring high availability, the cloud platform also provides high availability guarantees for applications hosted on it.
  • App International full name: Application, referred to as App
  • the promotion and promotion activities of the app are sent by the social platform, and 500,000 users are promoted at a time, and the concurrent visits of the user are sent out in the event information.
  • the embodiments of the present invention provide a method and a device for managing resources in a cloud platform, which can implement a fast and effective dynamic capacity increase in a situation in which an existing service is stable and available to cope with an increase in application access time in a short period of time.
  • the embodiment of the present application provides a method for managing resources in a cloud platform, where the cloud platform is configured to provide a running resource for an application deployed on the cloud platform, including: according to the first state information of the cloud platform. And determining a resource adjustment policy, where the resource adjustment policy is used to adjust a quantity of resources allocated to the application; determining a size of the smart sliding window ISW according to the second state information of the cloud platform, where the ISW is used to indicate a unit The maximum integrated access amount PV allowed by the application in the time; wherein the first status information and the second status information indicate an operating status when the cloud platform provides a service for the application; Policy, adjusting the allocation to the application The number of resources, and the ISW is adjusted according to the determined size of the ISW.
  • the execution entity of the embodiment of the present application is a device for managing resources in the cloud platform, which may be an independent device, or may be integrated in an LB system in the cloud platform or an elastic extension system.
  • the device that manages the resource can adjust the size of the resource allocated to the application and the size of the smart sliding window according to the state information of the cloud platform, and dynamically adjust the size of the ISW to ensure the application when the amount of access of the application increases rapidly. Provides a stable and highly available service while achieving a rhythmic, on-demand, incremental capacity by dynamically adjusting the amount of resources allocated to the application.
  • the determining a resource adjustment policy according to the first state information of the cloud platform includes: determining, according to the first state information, the cloud platform The health of the cloud platform indicates the performance status of the cloud platform when the service is provided for the application.
  • the device for managing resources in the cloud platform may determine the health of the cloud platform according to the running state information of the cloud platform, and then determine a resource adjustment policy for the health level according to the health of the cloud platform, thereby The adjustment policy adjusts the amount of resources allocated to the application.
  • the determining, according to the health of the cloud platform, the resource adjustment policy including: determining a plurality of health level levels Determining a health level corresponding to the health of the cloud platform; determining a resource adjustment policy corresponding to the health level according to a correspondence between the plurality of resource adjustment policies pre-configured by the cloud platform and the plurality of health levels.
  • the device for managing resources in the cloud platform may pre-configure different resource adjustment policies for different health levels. After obtaining the health degree of the cloud platform, the device may adopt corresponding health level corresponding to the health degree.
  • the resource adjustment policy adjusts the amount of resources allocated to the application. Therefore, different resource adjustment strategies can be adopted according to different health levels, so that a reasonable resource amount can be configured according to the health of the cloud platform, and the resource utilization rate of the cloud platform is improved.
  • the method further includes: determining, according to the fluctuation amount of the PV of the application and/or the quantity of remaining resources of the cloud platform, The current running state of the cloud platform is an emergency state; determining an emergency plan in the plan database of the cloud platform that matches the current running state of the cloud platform; and adjusting the allocation to the application according to the solution of the emergency plan The number of resources and the ISW.
  • the device for managing resources in the cloud platform may quickly match the emergency plan in the plan database when determining that the cloud platform is in an emergency state according to the fluctuation amount of the application and/or the number of remaining resources of the cloud platform, according to the An emergency plan solution that quickly allocates the number of resources to the application and the size of the ISW. Therefore, obtaining a solution by matching the emergency plan improves the decision-making speed for emergencies, and therefore, in the face of sudden increase in access When the amount is measured, it can also respond in time.
  • the determining, by the determining, the emergency plan in the plan database of the cloud platform that matches a current running state of the cloud platform includes: Determining, according to at least one of a PV of the application, a service type of the application, a quantity of remaining resources of the cloud platform, and a response time of the application, a current running state of the cloud platform and the emergency plan matching .
  • the determining the resource adjustment policy according to the health of the cloud platform includes: if the health of the cloud platform The first health degree threshold is greater than the second health degree threshold, and the health level of the cloud platform belongs to the first health level, and the resource adjustment policy corresponding to the first health level is based on the health of the cloud platform.
  • the resource adjustment policy corresponding to the level is based on the cloud platform
  • the resource adjustment suggestion is sent to the elastic scalability system, so that the elastic scalability system adjusts the quantity of resources allocated to the application according to the resource adjustment suggestion; if the cloud platform is healthy
  • the third health level threshold is determined to be the third health level, and the resource adjustment policy corresponding to the third health level is determined from the plan database of the cloud platform and the cloud platform.
  • the device for managing resources in the cloud platform may establish a correspondence between different health levels and health level levels, so as to determine a health level according to the health level after obtaining the health degree information, thereby determining The resource adjustment strategy corresponding to the health level.
  • the first state information of the cloud platform includes at least one of: a fluctuation amount of the PV of the application, the The number of remaining resources of the cloud platform, the average response time of the application, and the service health factor ⁇ of the application, the ⁇ is a performance indicator of the cloud platform fed back by the elastic scalability system; the second state of the cloud platform
  • the information includes at least one of the following: a maximum PV allowed for the application per unit time, a number of remaining resources of the cloud platform, and a time from startup to loading of the server serving the application.
  • an apparatus for managing resources in a cloud platform comprising modules for performing the method of the first aspect or any one of the implementations of the first aspect.
  • a third aspect provides a device for managing resources in a cloud platform, including a transceiver device, a software device, and a hardware device component;
  • the method in any one of the first aspect or the first aspect is implemented by a software device and/or a hardware device.
  • an apparatus for managing resources in a cloud platform including an input device, an output device, a processor, a memory, and a bus system.
  • the input device, the output device, the processor, and the memory are connected by a bus system, and the processor executes the instruction stored by the memory by calling an operation instruction stored in the memory, and the execution of the operation instruction stored in the memory causes the processor to execute the first
  • the method and device for managing resources in the cloud platform of the embodiment of the present application can adjust the quantity of resources allocated to the application and the size of the smart sliding window according to the state information of the cloud platform, and therefore, the user access amount is Incremental time, it is possible to provide a stable and highly available service by adjusting the number of resources allocated to the application and the size of the ISW.
  • FIG. 1 is a schematic diagram of a cloud computing system to which the embodiments of the present application are applied.
  • FIG. 2 is a schematic diagram of a Paas platform to which the embodiments of the present application are applied.
  • FIG. 3 is a schematic diagram of an apparatus for managing resources in a cloud platform according to an embodiment of the present application.
  • FIG. 4 is a schematic flowchart of a method for managing resources in a cloud platform according to an embodiment of the present application.
  • FIG. 5 is a schematic flowchart of a method for managing resources in a cloud platform according to another embodiment of the present application.
  • FIG. 6 is a logic diagram of an apparatus for managing resources in a cloud platform according to an embodiment of the present application.
  • FIG. 7 is a schematic block diagram of an apparatus for managing resources in a cloud platform according to an embodiment of the present application.
  • FIG. 8 is a schematic block diagram of an apparatus for managing resources in a cloud platform according to another embodiment of the present application.
  • FIG. 9 is a schematic block diagram of an apparatus for managing resources in a cloud platform according to still another embodiment of the present application.
  • FIG. 10 is a schematic block diagram of an apparatus for managing resources in a cloud platform according to still another embodiment of the present application.
  • the cloud computing system includes: a cloud infrastructure, and an operating system running on the cloud infrastructure;
  • the cloud infrastructure may include hardware resources provided by a plurality of physical machines (such as servers), such as a central processing unit (Central Processing Unit: CPU), memory, hard disk, network bandwidth, etc., and may also include security. Resources for power supply or refrigeration.
  • the operating system in a cloud computing system is commonly referred to as a cloud operating system and is used to implement functions such as abstraction, management, and scheduling of hardware resources.
  • the application Application, referred to as the application for short
  • the application is developed by the developer and deployed to the cloud platform.
  • the cloud platform serves as a bridge between the application and the underlying operating system, and can provide the deployment environment required for the application.
  • the execution environment, and the system resources such as computing and storage.
  • the cloud management system in the cloud platform can dynamically adjust the amount of resources allocated to the application through the load balancing system and/or the elastic scaling system. For example, when the application traffic increases, more resources can be allocated to the application.
  • the cloud platform in the cloud computing system can be provided to the user or the developer as a service. This mode is usually called platform-as-a-service (English name: Platform-as-a-Service, abbreviated as: PaaS), therefore, the cloud computing system
  • PaaS Platform-as-a-Service
  • the cloud platform in China is also commonly referred to as the PaaS cloud platform or the PaaS platform.
  • 2 is a schematic diagram of a PaaS platform according to an embodiment of the present application, which is divided according to a cloud computing logical structure.
  • the PaaS platform is located in an intermediate layer in a cloud computing system, and the upper layer is software as a service (English full name: Software-as-a -Service, abbreviation: SaaS), SaaS is responsible for maintaining and managing the hardware and software facilities in the cloud platform, and charging users for free or on-demand.
  • the lower layer is infrastructure-as-a-service (English name: Infrastructure-as-a-Service, IaaS for short), which is used to provide infrastructure services such as virtual computing, storage, and database. Users can obtain the required computing or storage from the supplier. Wait for resources to load related applications and pay only for the part of the resources that they are renting.
  • the executor of the embodiment of the present application is a device for managing resources in a cloud platform.
  • the device for managing resources may be multiple, and is used to manage applications of different types and different service types in the cloud platform.
  • each device that manages resources may be an independent device, or may be integrated into an LB system in the cloud platform, or may be integrated into the elastic telescopic system, which is not limited in this embodiment of the present application.
  • the device for managing resources may be a buffer (Bumper) in FIG. 2, which may be a separate module in the PaaS platform, or may be integrated in an LB system in the PaaS platform, or the Bumper may be integrated.
  • the method of managing resources in the elastic expansion system (ie, HA/retraction in FIG. 2) in the PaaS platform or the cloud platform in the embodiment of the present application can also be completed by combining the Bumper and the elastic extension system. This embodiment of the present application does not limit this.
  • the apparatus for managing resources of the execution entity of the embodiment of the present application may be referred to as a Bumper in FIG. 2, and may be other names.
  • the embodiment of the present application does not limit the name of the execution subject, and the Bumper may be an independent device.
  • the embodiment of the present application is not limited to the elastic extension system, or the LB system, or may be integrated into the elastic expansion system or the LB system.
  • the embodiment of the present application describes the execution body as a Bumper. .
  • the Bumper may include a state analysis module 310, a plan management module 320, and a resource management module 330.
  • the state analysis module 310 is configured to analyze and predict the running state of the cloud platform, and determine the health of the cloud platform.
  • the size of the smart sliding window (English full name: Intelligent Sliding Window, referred to as: ISW) (the size of the ISW is the maximum integrated access allowed by the application per unit time), so that the resource management module 330 Adjusting the ISW according to the determined size of the smart sliding window to mitigate the impact of the sudden increase of the access request, or making a resource adjustment scheme according to the health of the cloud platform, for example, increasing the amount of resources allocated to the application by starting the virtual machine, etc.
  • the amount of resources allocated to the application refers to the number of resources (such as CPU, Memory) allocated by the cloud platform to the application providing service; the plan management module 320 is used to manage the historical plan, because the types of applications in the cloud platform are numerous, the service Diversified types, the plan management module 320 can extract various effective historical plans to form a plan library for different services, and provide a solution for subsequent cases. For example, if an emergency occurs on the cloud platform, a plan similar to the current event can be quickly matched from the plan base.
  • a plan similar to the current event can be matched according to factors such as the number of access requests of the App or the response time of the App, and then The matching solution of the plan adjusts or adds VMs and the like to the ISW; the resource management module 330 can be used to adjust the amount of resources allocated to the application according to the solution of the matching plan, or according to the health of the cloud platform.
  • the state analysis module 310 may determine, according to the PV of the App, the response time (English name: Response Time, RT: shorthand), and one or more data of historical health data and historical smart sliding window data. The health of the current cloud platform and the size of the ISW.
  • the state analysis module 310 can also push the determined health data and the ISW data to the plan management module 320, and the plan management module 320 can match the health plan data and the ISW data with the historical plan in the cloud platform plan database to determine The solution for health data and ISW data.
  • the plan management module 320 can also push the solution to the resource management module 330, so that the resource management module 330 modulates the ISW or the amount of resources allocated to the application according to the solution.
  • the state analysis module 310 can also push the determined health data and ISW data to the resource management module 330, so that the resource management module 330 can formulate a corresponding solution according to the health degree data and the ISW data.
  • the resource management module 330 may further push the formulated solution to the plan management module 320 as a reference plan for the later system plan formulation.
  • the resource management module 330 may further push the formulated solution to the elastic extension system, so that the elastic extension system adjusts the ISW or the amount of resources allocated to the application according to the solution.
  • the state analysis module 310 can also push the determined health data and the ISW data to the elastic extension system, so that the elastic extension system formulates a corresponding resource adjustment policy according to the health degree data and the ISW data, and then according to the The resource adjustment policy adjusts the ISW and the amount of resources allocated to the application.
  • Bumper adjusts the amount of resources allocated to the application mainly includes two aspects: increasing the amount of resources allocated to the application (referred to as “energy capacity”), for example, increasing the number of virtual machine servers serving the App, etc., or reducing The amount of resources allocated to the application (referred to as “reduction”), for example, reducing the number of VMs that provide services to the App, and so on.
  • FIG. 4 is a schematic flowchart of a method 400 for managing resources in a cloud platform according to an embodiment of the present application. As shown in FIG. 4, the method 400 includes:
  • S410 Determine, according to the first state information of the cloud platform, a resource adjustment policy, where the resource adjustment policy is used to adjust a quantity of resources allocated to the application.
  • S420 Determine, according to the second state information of the cloud platform, a size of the smart sliding window ISW, where the ISW is used to indicate a maximum integrated access amount PV allowed by the application in a unit time; wherein the first state information and the second The status information indicates the operating status of the cloud platform when the service is provided for the application;
  • the execution body of the method 400 may be the buffer in FIG. 3, or may be an elastic extension system.
  • the embodiment of the present application is not limited by the embodiment of the present application.
  • the buffer determines a resource adjustment policy for adjusting the quantity of resources allocated to the application according to the first state information that provides the cloud platform.
  • the buffer may further determine the size of the ISW according to the second state information of the cloud platform, and adjust the size of the ISW to adjust the maximum allowed access amount of the application per unit time.
  • the first state information and the second state information both indicate an operating state when the cloud platform provides a service for the application, and the first state information and the second state information may be the same information or different information. There is no limit to this.
  • the first state information of the cloud platform may include at least one of: a fluctuation amount of the PV of the application, a quantity of remaining resources of the cloud platform, an average response time of the application, and a service health of the application.
  • a coefficient ⁇ , the ⁇ is a performance index of the cloud platform fed back by the elastic stretching system;
  • the second state information of the cloud platform includes at least one of the following: a maximum PV allowed by the application per unit time, and a remaining of the cloud platform The number of resources and the time the server is servicing the application from startup to loading.
  • the Bumper can predict that the user's visit volume will continue to increase in the future based on the fluctuation of the PV of the application in the current time period. If the operation continues, the performance of the system may deteriorate. At this time, the Bumper According to the fluctuation amount, the amount of resources allocated to the application and the size of the ISW are determined. Optionally, the Bumper can determine to increase the capacity of one VM when the fluctuation amount of the PV is 40%, and adjust the size of the smart sliding window to be original. 80%, or when the fluctuation of the PV is 80%, it is determined that the capacity of 4 VMs is increased, and the size of the smart sliding window is adjusted to 40%.
  • the method for managing resources in the cloud platform of the embodiment of the present application can adjust the number of resources allocated to the application and the size of the smart sliding window according to the state information of the cloud platform. Therefore, when the user access amount increases rapidly, the method can ensure At the same time as capacity expansion, stable and highly available services are provided by adjusting the size of the ISW.
  • the determining a resource adjustment policy according to the first state information of the cloud platform includes:
  • the resource adjustment strategy is determined.
  • the Bumper may first determine the health of the cloud platform according to the first state information of the cloud platform, and then determine a resource adjustment policy for the health of the cloud platform according to the health of the cloud platform, thereby A resource adjustment strategy that adjusts the amount of resources allocated to an application.
  • the first state information of the cloud platform may include at least one of: a fluctuation amount of a PV of the application, a quantity of remaining resources of the cloud platform, an average response time of the application, and a service of the application.
  • the health factor ⁇ which is a performance index of the cloud platform fed back by the elastic telescopic system.
  • the health of the cloud platform may be determined according to the first state information, and the health may be determined according to one or more of the foregoing state information.
  • the number of remaining resources of the cloud platform may be set, and the cloud platform may be set.
  • the health of the cloud platform may be represented by a score.
  • the health of the cloud platform may be represented by 0 to 100 points, and 100 points may be set to indicate that the cloud platform is running at an optimal state, and the score is higher. It indicates that the running state of the cloud platform is better, or the value of 0 can be set to indicate that the cloud platform is running in an optimal state, and the higher the score, the worse the operating state of the cloud platform is, and the like, which is not limited in this embodiment.
  • the health of the cloud platform may also be expressed by a health level.
  • the health of the cloud platform may be set into four health levels: health, sub-health, low risk, and high risk, and health level. Yunping The station is in good working condition and can provide high-performance services.
  • the sub-health level indicates that the cloud platform can provide high-performance services for a period of time, but if it encounters a sudden increase in user traffic, it may cause the performance of the cloud platform system to decline. That is to say, the performance of the cloud platform system is deteriorating at this time; when the cloud platform system is at a high risk level, it indicates that the cloud platform system is in a bad state and needs urgent measures, for example, increasing the number of resources allocated to the application. .
  • the Bumper may adjust the quantity of resources allocated to the application if the health of the cloud platform meets a preset condition, for example, if the health of the cloud platform is represented by a score (0 to 100), and The higher the score, the better the performance of the system.
  • the health of the cloud platform meets the preset condition.
  • the health of the cloud platform may be lower than a certain threshold.
  • the threshold may be 70 points, or if the cloud platform is The health degree is expressed by the above four health level, and the health of the cloud platform satisfies the preset condition, and the health of the cloud platform may be lower than a certain health level.
  • the health of the cloud platform satisfies the preset condition, indicating that the running state of the cloud platform system is deteriorating.
  • the Bumper can adjust the size of the ISW according to the second state information of the cloud platform to alleviate the impact of the surge in access requests. For example, if the health of the current cloud platform is slightly lower than the first threshold, that is, the operating state of the system is not very bad, but the situation of the sudden increase in user access cannot be supported, then the buffer may choose to adjust the ISW. Size to mitigate the impact of a sudden increase in user traffic, for example, the size of the ISW can be adjusted from 15k/s to 10k/s, so that the number of users allowed by the application per unit time is reduced, thereby reducing the system The load.
  • the buffer may optionally increase by 1 based on the amount of user visits in the future.
  • One or more VMs to ensure that the cloud platform system continues to provide stable and high-performance services. If the current cloud platform health indicates that the current cloud platform system is running poorly, if the amount of resources allocated to the application is not rapidly increased, the application access may be at risk of crashing. Alternatively, the Bumper may pass the matching plan library.
  • the emergency plan matching the current running status of the cloud platform, and then adjusting the amount of resources allocated to the application and the size of the ISW according to the solution of the emergency plan, for example, if the solution in the system plan is to add 4 VMs, Then the Bumper can add 4 VMs rhythmically according to the solution, or the Bumper can also send the solution to the elastic scaling system, and then the elastic scaling system can urgently add 4 VMs according to the solution; or the Bumper
  • the resource adjustment policy may be formulated according to the current running state of the cloud platform, so that the number of resources allocated to the application and the size of the ISW are adjusted according to the resource adjustment policy, or the Bumper may also store the formulated resource adjustment policy into the plan database, so that Learning and reference in later cases.
  • the method for managing resources in the cloud platform of the embodiment of the present application can adjust the number of resources allocated to the application and the size of the smart sliding window according to the running state information of the cloud platform system. Therefore, when the user access amount increases suddenly, It is guaranteed to provide stable and highly available services by adjusting the size of the ISW while increasing capacity.
  • the resource adjustment policy determined by the Bumper may be different according to the actual application scenario, the service type, or the user requirements of the application, and the embodiment of the present application does not perform the resource adjustment policy in the specific scenario. Out of restrictions.
  • the second state information of the cloud platform includes at least one of the following:
  • the maximum PV allowed for the application per unit time the number of remaining resources of the cloud platform, and the time from the start to the loading of the server serving the application.
  • the Bumper may determine the size of the ISW according to the second state information of the cloud platform system, and the Bumper may determine the size of the ISW according to one or more items in the foregoing information.
  • the size of the ISW can be determined according to the number of remaining resources of the cloud platform and the time from the startup to the loading service of the server serving the application. Ground, the size of the ISW can be determined according to the following formula:
  • ISW Min(Size,(C total -C Used )/T run )
  • Size is the maximum PV allowed for the application per unit time
  • C total indicates the total number of resources of the cloud platform
  • C Used indicates the number of used resources
  • C total -C Used indicates the number of remaining resources
  • T Run is the time from the start of the server to the loading of the service. (C total -C Used )/T run indicates the amount of user access that the number of remaining resources can support. Min indicates that the value is smaller.
  • the ISW can be determined according to the maximum amount of PV allowed by the application per unit time, that is, the number of users allowed to be supported and the number of remaining resources (C total - C Used ) / T run , if the size is greater than ( C total -C Used )/T run , ISW takes (C total -C Used )/T run, otherwise, ISW takes the value of Size.
  • determining the resource adjustment policy according to the health of the cloud platform including:
  • the resource adjustment policy corresponding to the health level is determined according to the correspondence between the multiple resource adjustment policies pre-configured by the cloud platform and the multiple health level.
  • the Bumper may determine a resource adjustment policy for the health according to the health of the cloud platform, and then adjust the quantity of resources allocated to the application according to the resource adjustment policy. For example, when the health of the cloud platform is A, it is determined that the number of added resources is M, and when the health of the cloud platform is B, it is determined that the number of added resources is N, and if the system runs at the health degree A, the performance is Better than the performance of the system running at health B, you can set N>M, which means that you need to increase the amount of resources to reduce the access pressure of the app under the state of health B.
  • the health of the cloud platform may correspond to multiple health level levels, and each health level may be configured with different resource adjustment policies, and after the device that manages the resources obtains the health of the cloud platform, The health of the cloud platform determines a health level corresponding to the health level, and then determines a resource adjustment policy corresponding to the health level.
  • the health of the cloud platform determines a health level corresponding to the health level, and then determines a resource adjustment policy corresponding to the health level.
  • the four health level levels may also respectively correspond to four score segments of the health of the cloud platform (combined from 0 to 100).
  • the correspondence between the health level, the health score, and the resource adjustment policy can be as shown in Table 1.
  • M 1 , M 2 , and M 3 are integers greater than 0, and M 1 ⁇ M 2 ⁇ M 3 .
  • the value of the M 1 , M 2 , and M 3 values may be determined according to a specific application scenario, which is not limited by the embodiment of the present application.
  • the data in Table 1 is merely an example and is not a limitation.
  • the health level corresponding to the health degree of the current cloud platform may be determined to be a sub-health level according to the correspondence between the health level and the health score in Table 1.
  • the Bumper can determine the resource adjustment policy corresponding to the health level according to the health level.
  • the resource adjustment policy corresponding to the sub-health level can be determined as the capacity expansion M 1 .
  • VM the resource adjustment policy corresponding to the sub-health level can be determined as the capacity expansion M 1 .
  • the resource adjustment Bumper adjustment policy allocated to an application for example, if the policy for the resource adjustment compatibilizer one VM M 1, which can be adjusted Bumper M 1 start one VM Serve applications.
  • the Bumper may re-determine the health of the cloud platform, so as to determine a resource adjustment policy for the latest health degree data according to the real-time change of the health of the cloud platform.
  • the Bumper can adjust the size of the ISW synchronously to meet the access needs of more users. Therefore, Bumper can gradually increase the capacity according to the change of the health of the cloud platform, instead of waiting for the system to face the collapse of the emergency capacity, resulting in the speed of capacity increase can not keep up with the increase in concurrent pressure, and just increase capacity The server immediately faced tremendous access pressure and immediately crashed.
  • the method for managing resources in the cloud platform of the embodiment of the present application can perform rhythmic capacity expansion while ensuring that the system provides stable services, and can also adjust the size of the ISW during the process of capacity expansion.
  • the size of the small ISW eases the pressure from a sudden increase in user traffic.
  • Health rating Health score Resource adjustment strategy health 71 ⁇ 100 0 Subhealth 66 ⁇ 70 M 1 low risk 51 ⁇ 65 M 2 high risk 0 to 50 M 3
  • the correspondence between the health score and the health level in Table 1 is for the sake of example only, and is not limited to the embodiment of the present application.
  • the embodiment of the present application may also adopt 81-100, 71-80, 61- 70, 0 to 60 respectively correspond to health, sub-health, low-risk and high-risk four health levels, etc., in the embodiment of the present application, the health of the cloud platform can also be evaluated by using 0 to 200 points or 0 to 10 points. It should be understood that the embodiment of the present application may further classify the health level of the cloud platform into five levels or three levels.
  • the resource adjustment policy in Table 1 is also for the sake of example. In the specific application, the amount of resources to be added is different from the actual application scenario and the user requirement.
  • the method for managing resources in the cloud platform of the embodiment of the present application can adjust the number of resources allocated to the application and the size of the smart sliding window according to the running state information of the cloud platform system, thereby providing the user with stable and highly available. service.
  • the method 400 further includes:
  • the number of resources allocated to the application and the ISW are adjusted.
  • the Bumper determines that the system is in an emergency state according to the fluctuation amount of the PV of the current application and/or the remaining resources of the cloud platform system, for example, the Bumper analyzes the current PV fluctuation amount according to the current PV fluctuation amount.
  • the Bumper can search for an emergency plan matching the current running state of the cloud platform system from the preview database. For example, the Bumper can determine the current cloud platform system according to factors such as the current user access amount and the service type of the application. The operational status matches the emergency plan.
  • the Bumper can then manage the number of resources allocated to the application and the size of the ISW based on the solution to the emergency plan in the plan repository. For example, if the solution of the emergency plan in the plan library is to increase the capacity of 4 VMs and adjust the ISW to 2 times, then the Bumper can adjust the resources of the cloud platform system according to the solution.
  • determining the current running status of the cloud platform in the plan database of the cloud platform Matching emergency plans including:
  • the current running state of the cloud platform and the emergency plan match are determined according to at least one of a PV of the application, a service type of the application, a quantity of remaining resources of the cloud platform, and a response time of the application.
  • the Bumper may determine the current running status of the system and the plan base according to one or more items of the PV of the application, the service type of the application, the number of remaining resources of the cloud platform system, and the response time of the application.
  • the status of the emergency plan matches.
  • the Bumper can determine the solution of the emergency plan when the service type of the application is consistent, and the access amount of the application PV and the emergency plan application match, determine the running state of the cloud platform system and the emergency plan match, thereby determining the solution of the emergency plan.
  • the policy is adjusted for the resources in the current running state of the system, and then the resources of the system and the size of the ISW can be managed according to the solution.
  • determining the health resource adjustment policy of the cloud platform according to the health of the cloud platform may further include:
  • the health of the cloud platform is determined to be the first health level, and the resource adjustment policy corresponding to the first health level is based on the cloud platform. Health, adjust the size of the smart sliding window ISW;
  • the health of the cloud platform is less than the second health threshold, the health of the cloud platform belongs to the second health level, and the resource adjustment policy corresponding to the second health level is based on the cloud platform.
  • the health degree, the resource adjustment proposal is formulated, and the resource adjustment proposal is sent to the elastic scalability system, so that the elastic scalability system adjusts the quantity of resources allocated to the application according to the resource adjustment proposal;
  • the health of the cloud platform is less than the third health threshold, determine that the health of the cloud platform belongs to a third health level, and the resource adjustment policy corresponding to the third health level is determined from the plan database of the system.
  • the platform's health-matching plan adjusts the amount of resources allocated to the application based on the solution used to process the plan;
  • the second health threshold is less than the first health threshold
  • the third health threshold is less than the second health threshold
  • the health of the cloud platform is less than the first health threshold is greater than the second health threshold, determining that the health of the cloud platform belongs to the first health level, optionally, if the health of the cloud platform is divided into Health, sub-health, low-risk and high-risk are classified into four health levels. At this time, the health of the cloud platform can be considered as a sub-health level. If the health of the cloud platform is divided by 100 points, the first health The degree threshold may be 70 points, and the second health level threshold may be 65 points. In this state, although the system can provide high-performance services for a period of time, if the PV continues to grow, it may cause system performance shortage.
  • the Bumper can reduce the impact of the surge in the access request by adjusting the size of the ISW.
  • the current ISW can be reduced, that is, the maximum number of user access allowed in the system per unit time is reduced.
  • the performance of the system can be improved by adjusting the size of the ISW; or the buffer can also send the health information of the cloud platform to the elastic extension system, and then the elastic extension system can According to the health information of the cloud platform, a corresponding resource adjustment scheme is formulated to adjust the quantity of resources allocated to the application.
  • the elastic scalability system can determine to increase the capacity of one VM according to the health information to cope with the current system. situation.
  • the health of the cloud platform is determined to be the second health level.
  • the health of the cloud platform is healthy, The four health levels of health, low risk and high risk are differentiated. At this time, the health level of the cloud platform can be considered to be at a low risk level.
  • the second health threshold is It can be 65 points, and the third health threshold can be 50 points.
  • the Bumper can send the cloud platform to the elastic extension system.
  • the health scalability information, and then the elastic scalability system can formulate a corresponding resource adjustment scheme according to the health information of the cloud platform, and adjust the quantity of resources allocated to the application.
  • the elastic scalability system is based on the health of the current cloud platform.
  • Information to determine the need to increase the capacity of a VM; or the Bumper can also compare the health information of the cloud platform with the health information or service type of the case in the plan library to determine the status of the current system.
  • the plan uses the solution of the plan as a resource adjustment strategy to adjust the amount of resources allocated to the application based on the resource adjustment policy.
  • the solution can be increased by increasing the number of resources allocated to the application and adjusting the size of the ISW.
  • the amount of capacity can be determined according to the health of the cloud platform, instead of simply increasing the amount of resources allocated to the application, causing some system resources to be idle, and the user needs to add these systems.
  • the resources of the cloud platform in the embodiment of the present application also improve the utilization of system resources.
  • the health of the cloud platform is less than the third health threshold, determine that the health of the cloud platform belongs to the third health level.
  • the health of the cloud platform is classified into health, sub-health, low risk, and high risk. The four health levels are differentiated. At this time, the health of the cloud platform can be considered to be at a high risk level. If the health of the cloud platform is divided by 100 points, the third health threshold can be 50 points.
  • the Bumper can compare the health information of the cloud platform with the health information of the case in the plan database, determine a plan that matches the running status of the current system, and then determine the plan.
  • the solution acts as a resource adjustment policy to adjust the amount of resources allocated to the application based on the resource adjustment policy.
  • the Bumper may send the resource adjustment policy to the elastic extension system through the emergency channel, and then the elastic extension system may perform emergency capacity expansion according to the resource adjustment policy.
  • the solution of adjusting the size of the ISW and increasing the number of VMs is not isolated and can be performed in combination.
  • the number of VMs can be increased first, and then the size of the ISW can be adjusted.
  • the number of VMs can be increased while the size of the ISW can be adjusted, or
  • the embodiment of the present application does not limit the size of the ISW, and then increases the number of VMs.
  • the above examples only represent three possible implementation manners, and do not limit the conditions for performing the above three implementation manners, for example, when the health degree is less than the first health degree threshold is greater than the second health threshold. Adjust the number of resources allocated to the application or the size of the ISW by matching the emergency plan in the plan repository.
  • the embodiment of the present application only takes the health degree and the better performance as an example, and is not limited thereto, and the health degree is set to be larger, and the system performance is worse, that is, the health degree may be inversely proportional to the system performance.
  • the health thresholds in the foregoing embodiments are only examples, and are not limited thereto, and the values of the thresholds may also be different according to actual application scenarios or requirements.
  • the first state information of the cloud platform may include at least one of the following:
  • the fluctuation of the PV of the application, the amount of remaining resources of the cloud platform, the average response time of the application, and the service health factor ⁇ of the application, the ⁇ is a performance index of the cloud platform fed back by the elastic scalability system.
  • the health of the cloud platform is determined by the first state information of the system, that is, the health of the cloud platform can also be determined according to at least one of the following:
  • the fluctuation of the PV of the application, the amount of remaining resources of the cloud platform, the average response time of the application, and the service health factor ⁇ of the application, the ⁇ is a performance indicator of the system fed back by the elastic expansion system.
  • the health of the platform can be represented by cloud o H, o H can be determined according to the following formula:
  • SUM represents a summation, that is, o, p, q, r
  • the sum of 1 the Wcur represents the fluctuation of the PV of the application in the current time period
  • the C total is the total amount of resources of the cloud platform
  • the C Used is the number of used resources
  • the C Increasing is the next time The number of resources that need to be added in the segment.
  • the RTavg is the average response time of the application
  • the RTstd is the maximum response time of the application.
  • ⁇ includes the central processing unit (English name: central processing unit: CPU), memory (memory), disk (disk), etc. of the server deployed by the App system and the App, which affect other factors such as system expansion. Obtained in a
  • the health of the cloud platform can be determined according to the following formula:
  • the embodiment of the present application determines the health degree by adopting the principle of 631, that is, the weight of the PV of the App in the current time period occupies a weight of 6, the number of remaining resources of the cloud platform occupies a weight of 3, and the average response time of the App.
  • the weight is 1.
  • the embodiment of the present application is used to determine the health of the present invention.
  • the 631 principle is used as an example for the purpose of illustration.
  • the embodiment of the present application may also adopt the 811 principle and the like. This example does not limit this.
  • the fluctuation amount W cur of the PV of the App in the current time period may be determined according to the following formula (1):
  • W cur (PV cur -PV prev )/max(PV cur ,PV prev ) (1)
  • PVcur is the PV applied in the current time period
  • PVprev is the PV applied in the previous time period
  • the average fluctuation amount W avg of the PV of the App in a unit time period may be determined according to the following formula (2):
  • W 1 , W 2 , . . . , W n are PV fluctuation amounts of n time periods, and Avg represents average value.
  • the PV in the next time period can be predicted according to the following formula (3):
  • PV next Sum(50% ⁇
  • the embodiment of the present application predicts that PV next adopts the principle of 515, that is, collects PV data of six sampling points, sets the weight of the first five sampling points to 1, and sets the weight of the current sampling point, that is, the current sampling point by 5, It should be noted that, in the embodiment of the present application, the 515 principle is used to predict the PV next .
  • the description is only for the sake of example, and the embodiment of the present application may also adopt the 811 principle, that is, the collection 3
  • the weight of the first two points is 1 and the weight of the nearest sampling point is 8 or the like. This embodiment does not limit this.
  • the amount of increase of PV for some time in the App PV increasing (4) is determined according to the following formula:
  • PV increasing (W avg ⁇ PV next ) ⁇ T run (4)
  • the W avg is an average fluctuation amount of the PV of the App in a unit time period
  • the T run is a time from the startup to the joining of the server.
  • the average response time RT avg of the App in the unit time period may be determined according to the following formula (5):
  • RT avg Sum (80% ⁇ RT cur , 10% ⁇ RT prev1 , 10% ⁇ RT prev2 ) (5)
  • the RT cur represents the response time of the App in the current time period
  • the RT prev1 and the RT prev2 represent the response time of the App in the two time periods before the current time period.
  • the embodiment of the present application determines that the RT avg adopts the 811 principle, that is, collects the RT data of three sampling points, sets the weight of the first two sampling points to 1, and sets the weight of the current sampling point, that is, the current sampling point by 8, It should be noted that the embodiment of the present application determines that the RT avg is based on the principle of the 811. The description is only for the sake of example, and the embodiment of the present application may also adopt the 515 principle, that is, the collection 6 The weight of the first 5 sampling points is 1 and the weight of the nearest sampling point is 5, which is not limited in the embodiment of the present application.
  • FIG. 5 illustrates a schematic diagram of a method 500 of managing resources in a cloud platform, which may be performed by the various modules in FIG. 6 in accordance with an embodiment of the present application.
  • the LB-Bumper can be used as a modular unit of the LB system. It can also be connected to the elastic extension system to send health data to the elastic extension system.
  • the elastic telescopic system makes the decision to increase or shrink.
  • the LB-Bumper can also maintain a connection with the elastic extension system through the Keep-alive link. For example, if the health of the cloud platform is at a sub-health level, the LB-Bumper can communicate through the common channel and the elastic extension system. In case of emergency, the LB-Bumper can also communicate through the emergency channel and the elastic telescopic system.
  • the PV at this time is low, and Bumper analyzes the health of the cloud platform to 95 according to the current PV situation of the application, and determines that the system is in good running state.
  • the Bumper communicates with the elastic telescopic system through a common channel at this time;
  • the Bumper determines the health of the current cloud platform and the size of the ISW according to the current PV situation. According to the analysis result, the current system is in good health.
  • the current ISW size is 1w, that is, the user allowed to access at this time.
  • the number of visits is 10,000;
  • the Bumper may analyze the state according to the fluctuation of the applied PV for a period of time;
  • the Bumper determines that the current cloud platform has a health score of 40 according to the fluctuation of the applied PV. At this time, the current system is in a high-risk state.
  • the Bumper searches the plan library for an emergency plan that matches the current running state of the system.
  • the solution of the emergency plan is used as a solution for the current system's operating conditions. For example, the solution is to urgently capacity up to 4 VMs.
  • the Bumper can notify the elastic telescopic system through the emergency channel;
  • the Bumper also The size of the ISW can be adjusted according to the actual operating conditions of the system (for example, from 1w/s ⁇ 5k / s ⁇ 3k / s ⁇ 4.5k / s ⁇ 1w / s).
  • the health of the cloud platform becomes 75, and the system performance is improved.
  • the PV is stable at 2w/s, and the health of the cloud platform reaches 80, that is, the health is greater than 75, so it is not required.
  • Bumper analyzes the current system's operating conditions, and the system may have the risk of overload or performance degradation for the next period of time;
  • the Bumper sends the current health information to the elastic extension system. Since the running state of the system is not very bad at this time, the Bumper can send the health information to the elastic extension system through an ordinary channel;
  • the elastic telescopic system determines to increase the capacity of one VM according to the health information
  • the cloud platform has a health of 75, and stops increasing capacity.
  • the above example mainly includes the implementation of the method for managing resources in the cloud platform in two scenarios (one is a scenario in which the PV is suddenly increased, and the other is a scenario in which the PV is slightly increased).
  • the Bumper can According to the emergency plan, the emergency expansion capacity of the elastic expansion system is notified, and the size of the ISW can be continuously adjusted during the emergency capacity increase to alleviate the system impact caused by the sudden increase of PV; when the PV increase is not large
  • the Bumper may determine the capacity increase according to the health information of the cloud platform, or may push the health information of the cloud platform to the elastic scalability system, so that the elastic expansion system determines the capacity expansion policy according to the health information.
  • the size of the ISW can be adjusted according to the real-time operating condition of the system.
  • the method for managing resources in the cloud platform of the embodiment of the present application can adjust the number of resources allocated to the application and the size of the smart sliding window according to the running state information of the cloud platform system. Therefore, when the user access amount increases suddenly, It is guaranteed to provide stable and highly available services by adjusting the size of the ISW while increasing capacity.
  • FIG. 7 is a schematic block diagram of an apparatus 700 for managing resources in a cloud platform according to an embodiment of the present application. As shown in FIG. 7, the apparatus 700 includes:
  • a determining module 710 configured to determine, according to the first state information of the cloud platform, a resource adjustment policy, where the resource adjustment policy is used to adjust a quantity of resources allocated to the application;
  • the determining module 710 is further configured to determine, according to the second state information of the cloud platform, a size of the smart sliding window ISW, where the ISW is used to indicate a maximum integrated access amount PV allowed by the application in a unit time; wherein the first state The information and the second status information indicate an operation status of the cloud platform when the service is provided for the application;
  • the management module 720 is configured to adjust the quantity of resources allocated to the application according to the resource adjustment policy, and adjust the ISW according to the determined size of the ISW.
  • the determining module 710 is equivalent to the function of the state analyzing module 310 in FIG. 3.
  • the function of the determining module 710 can be implemented by a software program.
  • the software program can be implemented in one process or can be implemented by a hardware chip.
  • the software module is implemented by a software module or a combination of hardware and software modules.
  • the determining module 710 may be integrated in the LB system or the elastic stretching system, or may be a separate module or the like.
  • one determining module 710 may be configured for each application, that is, the determining module 710 may be in an application unit, or may share a determining module 710, etc., for several applications. This embodiment of the present application does not limit this.
  • the management module 720 is equivalent to the functions of the plan management module 320 and the resource management module 330 in FIG. 2.
  • the function of the management module 720 can be implemented by a software program.
  • the software program can be placed in a
  • the implementation in the process can also be implemented by a software module on a hardware chip, or by a combination of hardware and software modules.
  • the management module 720 can be integrated into the LB system or the elastic extension system, and push the resource adjustment policy for the LB system or the elastic extension system, or can be a separate module or the like.
  • one management module 720 can be configured for each application, that is, the management module 720 can be an application unit, or a plurality of applications can share a management module 720. The embodiment does not limit this.
  • the device for managing resources in the cloud platform of the embodiment of the present application can adjust the number of resources for providing the application service and the size of the smart sliding window according to the running state information of the system, and therefore, when the user access amount increases rapidly, the device can ensure At the same time as capacity expansion, stable and highly available services are provided by adjusting the size of the ISW.
  • the apparatus 700 for managing resources in a cloud platform according to an embodiment of the present application may correspond to a Bumper in the method 400 of managing resources in a cloud platform according to an embodiment of the present application, and the above and other operations of the respective modules in the apparatus 700 and/or The functions are respectively implemented in order to implement the corresponding processes of the foregoing various methods, and are not described herein for brevity.
  • the embodiment of the present application further provides a schematic block diagram of an apparatus 800 for managing resources in a cloud platform, where the apparatus 800 includes:
  • Transceiver devices software devices, and hardware device parts
  • the transceiver device is a hardware circuit for completing packet transmission and reception
  • Hardware devices can also be called “hardware processing modules", or simpler, or simply “hardware”. Hardware devices mainly include dedicated hardware circuits based on FPGAs, ASICs (and other supporting devices, such as memory). The hardware circuits of certain functions are often processed much faster than general-purpose processors, but once the functions are customized, they are difficult to change. Therefore, they are not flexible to implement and are usually used to handle some fixed functions. It should be noted that the hardware device may also include an MCU (microprocessor, such as a single chip microcomputer) or a processor such as a CPU in practical applications, but the main function of these processors is not to complete the processing of big data, but mainly used for processing. Some control is performed. In this application scenario, the system that is paired with these devices is a hardware device.
  • MCU microprocessor, such as a single chip microcomputer
  • Software devices mainly include general-purpose processors (such as CPU) and some supporting devices (such as memory, hard disk and other storage devices), which can be programmed to let the processor have the corresponding processing functions.
  • general-purpose processors such as CPU
  • some supporting devices such as memory, hard disk and other storage devices
  • the processed data can be sent through the transceiver device through the hardware device, or the processed data can be sent to the transceiver device through an interface connected to the transceiver device.
  • the software device or the hardware device is used to determine the size of the resource adjustment policy and the ISW mentioned in the foregoing embodiment, and adjust the quantity and ISW of resources allocated to the application.
  • the device for managing resources in the cloud platform of the embodiment of the present application can adjust the number of resources allocated to the application and the size of the smart sliding window according to the running state information of the cloud platform system. Therefore, when the user access amount increases rapidly, It is guaranteed to provide stable and highly available services by adjusting the size of the ISW while increasing capacity.
  • the apparatus 800 for managing resources in a cloud platform according to an embodiment of the present application may correspond to a Bumper in the method 400 of managing resources in a cloud platform according to an embodiment of the present application, and the above and other operations of the respective modules in the apparatus 800 and/or The functions are respectively implemented in order to implement the corresponding processes of the foregoing various methods, and are not described herein for brevity.
  • the device for managing resources in the cloud platform provided by the embodiment of the present application may be a cloud host in the cloud computing system, and the cloud host may be a virtual machine running on the physical machine.
  • the physical machine 900 includes a hardware layer 910, a VMM (Virtual Machine Monitor) 920 running on the hardware layer 910, and a host Host 901 running on the VMM 920 and several virtual machines.
  • Machine VM, Virtual Machine
  • the hardware layer includes but is not limited to: I/O device, CPU, and memory.
  • the device for managing resources in the cloud platform provided by the embodiment of the present application may be a virtual machine in the physical machine 900, such as a VM 940.
  • the VM 940 runs one or more cloud applications, where each cloud application is used. To implement the corresponding business functions, such as database applications, map applications, etc., these cloud applications can be developed by developers and then deployed to cloud computing systems.
  • the VM 940 is also configured to execute a program, and the VM 940 runs the executable program, and calls the hardware resource of the hardware layer 910 through the host Host 930 during the running of the program to implement the management resource in the cloud platform in the cloud platform.
  • the function of the determining module and the management module of the device may be included in the executable program in the form of a software module or a function, for example, the executable program may include: a determining module and a management module
  • the VM 940 runs the executable program by calling resources such as CPU and Memory in the hardware layer 910 to implement the functions of the determining module and the management module. For brevity, details are not described herein again.
  • the embodiment of the present application further provides a schematic block diagram of a device 1000 for managing resources, where the device 1000 includes a processor 1010, a memory 1020, a bus system 1030, an input device 1040, and an output device 1050. .
  • Memory 1020 can include read only memory and random access memory and provides instructions and data to processor 1010. A portion of memory 1020 may also include non-volatile random access memory (NVRAM).
  • NVRAM non-volatile random access memory
  • the memory 1020 stores the following elements, executable modules or data structures, or a subset thereof, or an extended set thereof:
  • Operation instructions include various operation instructions for implementing various operations.
  • Operating system Includes a variety of system programs for implementing various basic services and handling hardware-based tasks.
  • the processor 1010 performs the following operations by calling an operation instruction stored in the memory 1020 (which can be stored in the operating system):
  • the processor 1010 controls the operation of the device 1000, which may also be referred to as a CPU (Central Processing Unit).
  • Memory 1020 can include read only memory and random access memory and provides instructions and data to processor 1010.
  • a portion of memory 1020 may also include non-volatile random access memory (NVRAM).
  • NVRAM non-volatile random access memory
  • the various components of the device 1000 are coupled together by a bus system 1030, which may include, in addition to the data bus, a power bus, a control bus, a status signal bus, and the like.
  • bus system 1030 may include, in addition to the data bus, a power bus, a control bus, a status signal bus, and the like.
  • various buses are labeled as the bus system 1030 in the figure. For ease of representation, only one thick line is shown in FIG. 10, but it does not mean that the bus system 1030 has only one bus or one type of bus.
  • the method disclosed in the foregoing embodiment of the present application may be applied to the processor 1010 or implemented by the processor 1010.
  • the processor 1010 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the foregoing method may be completed by an integrated logic circuit of hardware in the processor 1010 or an instruction in a form of software.
  • the processor 1010 described above may be a general purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, or discrete hardware. Component.
  • DSP digital signal processor
  • ASIC application specific integrated circuit
  • FPGA off-the-shelf programmable gate array
  • the methods, steps, and logical block diagrams disclosed in the embodiments of the present application can be implemented or executed.
  • the general purpose processor may be a microprocessor or the processor or any conventional processor or the like. Combined with the implementation of the application The steps of the disclosed method may be directly embodied by the hardware decoding processor being executed or by a combination of hardware and software modules in the decoding processor.
  • the software module can be located in a conventional storage medium such as random access memory, flash memory, read only memory, programmable read only memory or electrically erasable programmable memory, registers, and the like.
  • the storage medium is located in the memory 1020, and the processor 1010 reads the information in the memory 1020 and performs the steps of the above method in combination with its hardware.
  • the device for managing resources in the cloud platform of the embodiment of the present application can adjust the number of resources for providing the application service and the size of the smart sliding window according to the running state information of the system, and therefore, when the user access amount increases rapidly, the device can ensure At the same time as capacity expansion, stable and highly available services are provided by adjusting the size of the ISW.
  • the apparatus 1000 for managing resources in a cloud platform according to an embodiment of the present application may correspond to a Bumper in the method 400 of managing resources in a cloud platform according to an embodiment of the present application, and the above and other operations of the respective modules in the apparatus 1000 and/or The functions are respectively implemented in order to implement the corresponding processes of the foregoing various methods, and are not described herein for brevity.
  • the size of the sequence numbers of the foregoing processes does not mean the order of execution sequence, and the order of execution of each process should be determined by its function and internal logic, and should not be applied to the embodiment of the present application.
  • the implementation process constitutes any limitation.
  • the disclosed systems, devices, and methods may be implemented in other manners.
  • the device embodiments described above are merely illustrative.
  • the division of the unit is only a logical function division.
  • there may be another division manner for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in an electrical, mechanical or other form.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the functions may be stored in a computer readable storage medium if implemented in the form of a software functional unit and sold or used as a standalone product.
  • the technical solution of the present application which is essential or contributes to the prior art, or a part of the technical solution, may be embodied in the form of a software product, which is stored in a storage medium, including Several instructions to make a computer device (can It is a personal computer, server, or network device, etc.) that performs all or part of the steps of the methods described in various embodiments of the present application.
  • the foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and the like. .

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Environmental & Geological Engineering (AREA)
  • Multimedia (AREA)
  • Debugging And Monitoring (AREA)
  • Stored Programmes (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

本申请公开了一种云平台中管理资源的方法和装置，该方法包括：根据该云平台的第一状态信息，确定资源调整策略，该资源调整策略用于调整分配给该应用的资源的数量；根据该云平台的第二状态信息，确定智能滑窗ISW的大小，该ISW用于指示单位时间内该应用被允许的最大综合访问量PV；其中，该第一状态信息和该第二状态信息指示该云平台为该应用提供服务时的运行状况；根据该资源调整策略，调整分配给该应用的资源的数量，并根据确定的该ISW的大小调整该ISW。因此，本申请实施例的云平台中管理资源的方法和装置，在用户访问量骤增时，能够通过调整分配给应用的资源的数量和ISW的大小提供稳定而高可用的服务。

Description

云平台中管理资源的方法和装置
本申请要求于2016年01月29日提交中国专利局、申请号为201610067172.X、发明名称为“云平台中管理资源的方法和装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及信息技术领域，并且更具体地，涉及一种云平台中管理资源的方法和装置。
背景技术
在云计算时代，大量的应用都被托管在云平台上。云平台除了要保证自身的高可用，还要对托管在其上的应用提供高可用的保证。当应用程序(英文全称：Application，简称：App)的用户访问量骤增时，例如，该App的促销和推广活动由社交平台发出，一次推广50万用户，用户的并发访问量在活动信息发出后的一分钟内突然从2千/秒(2k/s)突然剧增到5万/秒(5w/s)；或者，该App突然在社交平台上得到某位知名人士推荐，可能在毫无预兆的情况下突然用户访问量大增，在一天内并发访问量可能从2千/秒增加到10万/秒。
面对访问量剧增的情况，为了保持应用稳定可用，现有技术有两种方案，一种是使用负载均衡(英文全称：Load Blance，简称：LB)系统，例如，设置一个App服务性能上限，在综合浏览量(英文全称：Page View，简称：PV)达到这个上限值时，只处理上限值范围内的请求，对于超限的请求进行等待或类似拒绝服务。采用这种负载均衡系统来应对访问量骤增的情况，会导致应用无法满足所有用户的访问需求，即对部分用户不可用。现有技术的另一种解决方案是采用弹性伸缩系统监控App的运行状态等信息，当应用的访问量增加时，进行动态增容，例如，根据应用的访问量的增加量，增加提供App服务的虚拟机(英文全称：Virtual Machine，简称：VM)或服务器的数量。但是当访问量骤增时，增容速度往往赶不上并发访问量增加的速度，从而会导致在增容完成之前已经有虚拟机或服务器崩溃，而新增容的虚拟机或服务器立刻就承受巨大的访问压力进而宕机。因此，在用户访问量骤增时，两种方案都难以确保为用户提供稳定而高可用的服务。
发明内容
本申请实施例提供一种云平台中管理资源的方法和装置，能够在保证现有服务稳定可用的情况下，实现快速、有效的动态增容以应对应用访问量短时间内增大的状况。
第一方面，本申请实施例提供了一种云平台中管理资源的方法，该云平台用于为部署在其上的应用提供运行所需资源，包括：根据所述云平台的第一状态信息，确定资源调整策略，所述资源调整策略用于调整分配给所述应用的资源的数量；根据所述云平台的第二状态信息，确定智能滑窗ISW的大小，所述ISW用于指示单位时间内所述应用被允许的最大综合访问量PV；其中，所述第一状态信息和所述第二状态信息指示所述云平台为所述应用提供服务时的运行状况；根据所述资源调整策略，调整分配给所述应用的 资源的数量，并根据确定的所述ISW的大小调整所述ISW。
本申请实施例的执行主体为云平台中管理资源的装置，它可以为独立的装置，也可以集成在云平台中的LB系统中或弹性伸缩系统中等，在第一方面所述的实现方式中，该管理资源的装置通过根据云平台的状态信息，调整分配给应用的资源的数量以及智能滑窗的大小，通过动态调整ISW的大小，在应用的访问量骤增时，能够保证为该应用提供稳定而高可用的服务，同时通过动态调整分配给应用的资源数量实现有节奏的按需逐步增容。
结合第一方面，在第一方面的第一种实现方式中，所述根据所述云平台的第一状态信息，确定资源调整策略，包括：根据所述第一状态信息，确定所述云平台的健康度，所述云平台的健康度指示所述云平台为所述应用提供服务时的性能状况。在该实现方式中，云平台中管理资源的装置可以根据云平台的运行状态信息，确定云平台的健康度，然后根据云平台的健康度确定针对该健康度的资源调整策略，从而根据该资源调整策略调整分配给应用的资源的数量。
结合第一方面及其上述实现方式，在第一方面的第二种实现方式中，所述根据所述云平台的健康度，确定所述资源调整策略，包括：确定多个健康度等级中所述云平台的健康度对应的健康度等级；根据所述云平台预配置的多个资源调整策略与所述多个健康度等级的对应关系，确定所述健康度等级对应的资源调整策略。在该实现方式中，云平台中管理资源的装置可以针对不同的健康度等级预配置不同的资源调整策略，在获取云平台的健康度后，可以根据该健康度对应的健康度等级，采用相应的资源调整策略调整分配给应用的资源的数量。因此，能够根据不同的健康度，采用差异化的资源调整策略，从而能根据云平台的健康度配置合理的资源量，同时提高了云平台的资源的利用率。
结合第一方面及其上述实现方式，在第一方面的第三种实现方式中，所述方法还包括：根据应用的PV的波动量和/或所述云平台的剩余资源的数量，确定所述云平台当前的运行状态属于紧急状态；确定所述云平台的预案库中与所述云平台当前的运行状态匹配的紧急预案；根据所述紧急预案的解决方案，调整分配给所述应用的资源的数量和所述ISW。在该实现方式中，云平台中管理资源的装置在根据应用的波动量和/或云平台的剩余资源的数量，确定云平台处于紧急状态时，可以快速匹配预案库中的紧急预案，根据该紧急预案的解决方案，快速分配给所述应用的资源的数量和ISW的大小，因此，通过匹配紧急预案的方法获取解决方案提高了对于紧急事件的决策速度，因此，在面对骤增的访问量时，也能及时作出反应。
结合第一方面及其上述实现方式，在第一方面的第四种实现方式中，所述确定所述云平台的预案库中与所述云平台当前的运行状态匹配的紧急预案，包括：根据所述应用的PV、所述应用的服务类型、所述云平台的剩余资源的数量和所述应用的响应时间中的至少一项来确定所述云平台当前的运行状态和所述紧急预案匹配。
结合第一方面及其上述实现方式，在第一方面的第五种实现方式中，所述根据所述云平台的健康度，确定所述资源调整策略，包括：若所述云平台的健康度小于第一健康度阈值大于第二健康度阈值，确定所述云平台的健康度属于第一健康度等级，所述第一健康度等级对应的资源调整策略为根据所述云平台的健康度，调整智能滑窗ISW的大小；若所述云平台的健康度小于第二健康度阈值大于第三健康度阈值，确定所述云平台的健康度属于第二健康度等级，所述第二健康度等级对应的资源调整策略为根据所述云平台 的健康度，制定资源调整建议，向弹性伸缩系统发送所述资源调整建议，以便于所述弹性伸缩系统根据所述资源调整建议调整分配给应用的资源的数量；若所述云平台的健康度小于第三健康度阈值，确定所述云平台的健康度属于第三健康度等级，所述第三健康度等级对应的资源调整策略为从所述云平台的预案库确定跟所述云平台的健康度匹配的预案，根据处理所述预案采用的解决方案调整分配给应用的资源的数量；其中，所述第二健康度阈值小于所述第一健康度阈值，所述第三健康度阈值小于所述第二健康度阈值。在该实现方式中，该云平台中管理资源的装置可以建立不同的健康度和健康度等级的对应关系，以便于在获取健康度信息后，根据该健康度确定对应的健康度等级，从而确定该健康度等级对应的资源调整策略。
结合第一方面及其上述实现方式，在第一方面的第六种实现方式中，所述云平台的第一状态信息包括以下中的至少一项：所述应用的PV的波动量、所述云平台的剩余资源的数量、所述应用的平均应答时间和所述应用的服务健康系数β，所述β为弹性伸缩系统反馈的所述云平台的性能指标；所述云平台的第二状态信息包括以下中的至少一项：单位时间内所述应用被允许的最大PV、所述云平台的剩余的资源的数量和为应用提供服务的服务器从启动到加载入服的时间。
第二方面，提供了一种云平台中管理资源的装置，该装置包括用于执行第一方面或第一方面的任一种实现方式中的方法的各模块。
第三方面，提供了一种云平台中管理资源的装置,包括收发器件、软件器件以及硬件器件部分；
在本申请实施例中，通过软件器件和/或硬件器件实现第一方面或第一方面的任一种实现方式中的方法。
第四方面，提供了一种云平台中管理资源的装置,包括输入设备、输出设备、处理器、存储器和总线系统。其中，输入设备、输出设备、处理器、存储器通过总线系统相连，处理器通过调用存储器存储的操作指令执行该存储器存储的指令，并且对该存储器中存储的操作指令的执行使得该处理器执行第一方面或第一方面的任一种实现方式中的方法。
基于上述技术方案，本申请实施例的云平台中管理资源的方法和装置，能够根据云平台的状态信息，调整分配给应用的资源的数量以及智能滑窗的大小，因此，在用户访问量骤增时，能够通过调整分配给应用的资源的数量和ISW的大小提供稳定而高可用的服务。
附图说明
图1是根据本申请实施例所适用的云计算系统的示意图。
图2是根据本申请实施例所适用的Paas平台的示意图。
图3是根据本申请实施例的云平台中管理资源的装置的原理图。
图4是根据本申请实施例的云平台中管理资源的方法的示意性流程图。
图5是根据本申请另一实施例的云平台中管理资源的方法的示意性流程图。
图6是根据本申请实施例的云平台中管理资源的装置的一种逻辑图。
图7是根据本申请实施例的云平台中管理资源的装置的示意性框图。
图8是根据本申请另一实施例的云平台中管理资源的装置的示意性框图。
图9是根据本申请再一实施例的云平台中管理资源的装置的示意性框图。
图10是根据本申请再一实施例的云平台中管理资源的装置的示意性框图。
具体实施方式
本申请实施例的方法可以典型地应用于如图1所示的云计算系统中，如图1所示，云计算系统包括：云基础设施，以及运行于云基础设施之上的操作系统；其中，云基础设施可以包括由众多物理机(如服务器)所提供的硬件资源，例如中央处理器(英文全称：Central Processing Unit，简称：CPU)，内存，硬盘，网络带宽等资源，还可以包括安全、供电或制冷等方面的资源。云计算系统中的操作系统通常称为云操作系统，用于实现硬件资源的抽象、管理和调度等功能。应用程序(Application，本申请实施例简称为应用)由开发者开发完成之后，被部署到云平台上，云平台作为应用与底层操作系统之间的桥梁，能为应用提供运行所需的部署环境、执行环境，以及运算、存储等系统资源，进一步地，云平台中的云管理系统可以通过负载均衡系统和/或弹性伸缩系统动态调整为应用分配的资源量。例如，在应用访问量增大时，可以给该应用分配较多的系统资源。
云计算系统中的云平台可以作为一种服务提供给用户或开发者，这种模式通常称为平台即服务(英文全称：Platform-as-a-Service，简称：PaaS)，因此，云计算系统中的云平台通常也称为PaaS云平台或PaaS平台。图2为本申请实施例提供的一个PaaS平台的示意图，按照云计算逻辑结构来划分，该PaaS平台在云计算系统中位于中间层，其上层是软件即服务(英文全称：Software-as-a-Service，简称：SaaS)，SaaS负责维护和管理云平台中的软硬件设施，同时以免费或者按需使用的方式向用户收费。其下层是基础设施即服务(英文全称：Infrastructure-as-a-Service，简称：IaaS)，用于提供虚拟计算、存储、数据库等基础设施服务，用户可以从供应商那里获得需要的计算或者存储等资源来装载相关应用，并只需为其所租用的那部分资源付费。
本申请实施例的执行主体为云平台中管理资源的装置，在云平台中该管理资源的装置可以为多个，用于管理云平台中不同种类，不同服务类型的应用。可选地，每一个管理资源的装置都可以为独立的装置，也可以集成于云平台中的LB系统中，或者还可以集成于弹性伸缩系统中，本申请实施例对此不作限制。例如，该管理资源的装置可以为图2中的缓冲器(Bumper)，它可以为该PaaS平台中的独立的模块，也可以集成在该PaaS平台中的LB系统中，或者该Bumper还可以集成于该PaaS平台中的弹性伸缩系统(即图2中的HA/伸缩)中，或者本申请实施例的云平台中管理资源的方法也可以通过该Bumper与弹性伸缩系统的结合来完成。本申请实施例对此不作限制。
应理解，本申请实施例的执行主体管理资源的装置可以简称为图2中的Bumper，当然也可以为其他的名称，本申请实施例对执行主体的名称不作限制，该Bumper可以为独立的装置，或者也可以为弹性伸缩系统，或LB系统，或者也可以集成在该弹性伸缩系统或LB系统中等，本申请实施例对此不作限制，为了简洁，本申请实施例以执行主体为Bumper来描述。
如图3所示，该Bumper可以包括状态分析模块310，预案管理模块320，资源管理模块330，其中，该状态分析模块310用于对云平台的运行状态进行分析和预测，确定云平台的健康度和智能滑窗(英文全称：Intelligent Sliding Window，简称：ISW)的大小(ISW的大小为单位时间内应用被允许的最大综合访问量)，以便于资源管理模块330 根据确定的智能滑窗的大小调整ISW以疏缓访问请求骤增带来的冲击，或根据该云平台的健康度做出资源调整方案，例如，通过启动虚拟机增加给应用分配的资源量等；给应用分配的资源量指的是云平台分配给应用提供服务的资源(比如CPU、Memory)的数量；该预案管理模块320用于管理历史预案，因为云平台中的应用的种类繁多，服务类型多样化，该预案管理模块320可以抽取各类有效的历史预案，以形成针对不同服务的预案库，方便为后续的案例提供解决方案。例如，若云平台发生紧急事件，可以快速从预案库匹配跟当前事件类似的预案，例如，可以根据App的访问请求数或App的响应时间等因素来匹配跟当前事件类似的预案，然后可以根据匹配出来的预案的解决方案对ISW进行调整或增加VM等；该资源管理模块330可以用于根据匹配的预案的解决方案调整ISW或分配给应用的资源量，或根据该云平台的健康度确定相应的资源调整建议，然后根据该资源调整建议调整分配给应用的资源量，或者也可以根据该状态分析模块310确定的ISW的大小调整ISW。例如，该状态分析模块310可以根据App的PV、响应时间(英文全称：Response Time，简称：RT)以及历史的健康度数据和历史的智能滑窗的数据中的一种或几种数据，确定当前云平台的健康度和ISW的大小。该状态分析模块310还可以将确定的健康度数据和ISW数据推送给预案管理模块320，该预案管理模块320可以根据健康度数据和ISW数据跟云平台预案库中的历史预案进行匹配，确定针对该健康度数据和ISW数据的解决方案。可选地，该预案管理模块320还可以将该解决方案推送到资源管理模块330，以便于该资源管理模块330根据该解决方案调制ISW或分配给应用的资源量。该状态分析模块310也可以将确定的健康度数据和ISW数据推送到资源管理模块330，以便于该资源管理模块330可以根据健康度数据和ISW数据制定相应的解决方案。可选地，该资源管理模块330还可以将制定的该解决方案推送到预案管理模块320，作为后期的系统预案制定时的参考预案。可选地，该资源管理模块330还可以将该制定的解决方案推送给弹性伸缩系统，以便于该弹性伸缩系统根据该解决方案调整ISW或分配给应用的资源量。可选地，该状态分析模块310也可以将确定的健康度数据和ISW数据推送到弹性伸缩系统，以便于该弹性伸缩系统根据该健康度数据和ISW数据制定相应的资源调整策略，然后根据该资源调整策略调整ISW和分配给应用的资源量。
应理解，Bumper调整分配给应用的资源量主要包括两个方面：增加分配给应用的资源量(简称为“增容”)，例如，增加给App提供服务的虚拟机服务器的数量等，或减少分配给应用的资源量(简称为“缩容”)，例如，减少给App提供服务的VM的数量等。
图4示出了根据本申请实施例的云平台中管理资源的方法400的示意性流程图，如图4所示，该方法400包括：
S410，根据该云平台的第一状态信息，确定资源调整策略，该资源调整策略用于调整分配给该应用的资源的数量；
S420，根据该云平台的第二状态信息，确定智能滑窗ISW的大小，该ISW用于指示单位时间内该应用被允许的最大综合访问量PV；其中，该第一状态信息和该第二状态信息指示该云平台为该应用提供服务时的运行状况；
S430，根据该资源调整策略，调整分配给该应用的资源的数量，并根据确定的该ISW的大小调整该ISW。
具体而言，该方法400的执行主体为可以为图3中的缓冲器，也可以为弹性伸缩系 统，或者也可以由该缓冲器结合弹性伸缩系统来完成，本申请实施例对此不作限制。首先，缓冲器根据提供云平台的第一状态信息，确定资源调整策略，该资源调整策略用于调整分配给应用的资源的数量。该缓冲器还可以根据该云平台的第二状态信息，确定ISW的大小，通过调整该ISW的大小，从而调整单位时间内应用被允许的最大访问量。该第一状态信息和该第二状态信息都指示该云平台为应用提供服务时的运行状态，该第一状态信息和该第二状态信息可以为相同的信息或不同的信息，本申请实施例对此不作限制。可选地，该云平台的第一状态信息可以包括以下中的至少一项：该应用的PV的波动量、该云平台的剩余资源的数量、该应用的平均应答时间和该应用的服务健康系数β，该β为弹性伸缩系统反馈的该云平台的性能指标；该云平台的第二状态信息包括以下中的至少一项：单位时间内该应用被允许的最大PV、该云平台的剩余的资源的数量和为应用提供服务的服务器从启动到加载入服的时间。
例如，该Bumper可以根据当前时间段内该应用的PV的波动量，预测未来一段时间内，用户访问量有持续增长的趋势，如果继续运行下去，系统的性能可能变差，此时，该Bumper根据该波动量，确定分配给应用的资源量以及ISW的大小，可选地，该Bumper可以在该PV的波动量为40％时，确定增容1台VM，调整智能滑窗大小为原来的80％，或者在该PV的波动量为80％时，确定增容4台VM，调整智能滑窗大小为原来的40％等。
因此，本申请实施例的云平台中管理资源的方法，能够根据云平台的状态信息，调整分配给应用的资源的数量以及智能滑窗的大小，因此，在用户访问量骤增时，能够保证在增容的同时，通过调整ISW的大小提供稳定而高可用的服务。
可选地，作为一个实施例，该根据该云平台的第一状态信息，确定资源调整策略，包括：
根据该第一状态信息，确定该云平台的健康度，该云平台的健康度指示该云平台给该应用提供服务时的性能状况；
根据该云平台的健康度，确定该资源调整策略。
具体地，该Bumper可以首先根据该云平台的第一状态信息，确定云平台的健康度，然后可以根据该云平台的健康度，确定针对该云平台的健康度的资源调整策略，从而根据该资源调整策略，调整分配给应用的资源的数量。可选地，因为该云平台的第一状态信息可以包括以下中的至少一项：该应用的PV的波动量、该云平台的剩余资源的数量、该应用的平均应答时间和该应用的服务健康系数β，该β为弹性伸缩系统反馈的该云平台的性能指标。该云平台的健康度可以根据该第一状态信息确定，那么该健康度也可以根据上述状态信息中的一项或几项确定，例如，可以设置云平台的剩余资源的数量越大，云平台的健康度越高，或PV的正向波动量越大，云平台的健康度越低，或者也可以设置App的响应时间越长，云平台的健康度越低等。
可选地，该云平台的健康度可以用分值表示，例如，云平台的健康度可以用0～100分来表示，可以设置100分表示云平台运行在最佳状态，分值越高，表示云平台的运行状态越好，或者也可以设置0分表示云平台运行在最佳状态，分值越高，表示云平台的运行状态越差等，本申请实施例对此不作限制。
可选地，该云平台的健康度也可以用健康度等级来表示，例如，可以设置云平台的健康度可以分为四个健康度等级：健康、亚健康、低风险和高风险，健康等级表示云平 台运行状态良好，可以提供高性能的服务，亚健康等级表示云平台虽然一段时间内可以提供高性能的服务，但是如果遇到骤增的用户访问量，有可能引起云平台系统性能的下降，也就是说此时云平台系统的性能有变坏的趋势；云平台系统处于高风险等级时，表示云平台系统运行状态很差，需要采取紧急措施，例如，增大分配给应用的资源的数量。
可选地，该Bumper可以在云平台的健康度满足预设条件的情况下，调整分配给应用的资源的数量，例如，若该云平台的健康度用分值表示(0～100)，并且，分值越高表示系统性能越好时，该云平台的健康度满足预设条件可以为云平台的健康度低于某个阈值，例如，该阈值可以为70分，或者若该云平台的健康度用上述四个健康度等级表示，那么云平台的健康度满足预设条件可以为云平台的健康度低于某个健康度等级。该云平台的健康度满足预设条件表示该云平台系统的运行状况有变差的趋势，如果遇到骤增的用户访问量，可能会引起系统性能的下降。可选地，该Bumper可以根据云平台的第二状态信息调整ISW的大小以疏缓访问请求骤增的冲击。例如，若当前云平台的健康度略低于第一阈值，也就是系统的运行状态还不是很差，只是不能支持骤增的用户访问量的情况的话，那么，该缓冲器可以选择调整ISW的大小来疏缓用户访问量的骤增带来的冲击，例如，可以将ISW的大小由15k/s调整到10k/s，以使得单位时间内应用被允许的用户的访问量减少，从而减轻系统的负荷。
上述解决方案往往适用于用户访问量不会持续增多的情况，如果用户的访问量有持续增多的趋势，可选地，该缓冲器可以根据未来一段时间内的用户访问量的波动量选择增加1台或多台VM，以保证云平台系统持续提供稳定高性能的服务。若当前云平台的健康度指示当前云平台系统的运行状态很差，如果不迅速增加分配给应用的资源量，应用的访问有可能面临崩溃的风险，可选地，该Bumper可以通过匹配预案库中的跟云平台的当前运行状况匹配的紧急预案，然后根据该紧急预案的解决方案调整分配给应用的资源的数量和ISW的大小，例如，若系统预案中的解决方案为增加4台VM，那么该Bumper可以根据该解决方案有节奏的增加4台VM，或者该Bumper也可以将该解决方案发送给弹性伸缩系统，然后弹性伸缩系统可以根据该解决方案，紧急增加4台VM；或者该Bumper本身可以根据当前云平台的运行状态制定资源调整策略，从而根据该资源调整策略调整分配给应用的资源的数量和ISW的大小，或者该Bumper还可以将制定的资源调整策略存储到预案库，以便于后期案例的学习和参考。
因此，本申请实施例的云平台中管理资源的方法，能够根据云平台系统的运行状态信息，调整分配给应用的资源的数量以及智能滑窗的大小，因此，在用户访问量骤增时，能够保证在增容的同时，通过调整ISW的大小提供稳定而高可用的服务。
应理解，以上举例仅为示例不作限定，Bumper确定的资源调整策略，还可以根据应用的实际应用场景、服务类型或用户需求的不同而不同，本申请实施例不对具体场景下的资源调整策略做出限制。
可选地，作为一个实施例，该云平台的第二状态信息包括以下中的至少一项：
单位时间内该应用被允许的最大PV、该云平台的剩余的资源的数量和为应用提供服务的服务器从启动到加载入服的时间。
具体地，该Bumper可以该云平台系统的第二状态信息，确定ISW的大小，该Bumper可以根据上述信息中的一项或几项确定ISW的大小。例如，该ISW的大小可以根据云平台的剩余的资源的数量和为应用提供服务的服务器从启动到加载入服的时间确定，可选 地，该ISW的大小可以根据以下公式确定：
ISW＝Min(Size,(Ctotal-CUsed)/Trun)
其中，Size为单位时间内应用被允许的最大的PV，Ctotal表示云平台总的资源的数量，CUsed表示已使用的资源的数量，Ctotal-CUsed表示剩余的资源的数量，该Trun为服务器从启动到加载入服务的时间，(Ctotal-CUsed)/Trun表示剩余的资源的数量能支持的用户访问量，Min表示取较小值。
也就是说，该ISW可以根据单位时间内应用被允许的最大的PV即Size和剩余的资源的数量能支持的用户访问量(Ctotal-CUsed)/Trun来确定，若该Size大于(Ctotal-CUsed)/Trun，ISW取(Ctotal-CUsed)/Trun，否则，ISW取Size的值。
可选地，作为一个实施例，该根据该云平台的健康度，确定该资源调整策略，包括：
确定多个健康度等级中该云平台的健康度对应的健康度等级；
根据该云平台预配置的多个资源调整策略与该多个健康度等级的对应关系，确定该健康度等级对应的资源调整策略。
具体地，该Bumper可以根据该云平台的健康度，确定针对该健康度的资源调整策略，然后根据该资源调整策略调整分配给应用的资源的数量。例如，可以在云平台的健康度为A时，确定增加的资源的数量为M，云平台的健康度为B时，确定增加的资源的数量为N，若系统运行在健康度A时的性能优于系统运行在健康度B时的性能，可以设置N>M，也就是说要缓解健康度为B的状态下的App的访问压力需要增加更多的资源量。可选地，该云平台的健康度可以对应多个健康度等级，每个健康度等级可以分别配置不同的资源调整策略，那么在该管理资源的装置获取该云平台的健康度以后，可以根据该云平台的健康度确定该健康度对应的健康度等级，然后确定该健康度等级对应的资源调整策略。例如，若该云平台的健康度分为四个健康度等级：健康、亚健康、低风险和高风险，可以为该四个健康度等级分别配置4种不同的资源调整策略，可选地，该四个健康度等级还可以分别对应云平台的健康度的四个分值段(以0～100分为例)。例如，该健康度等级、健康度分值和资源调整策略的对应关系可以如表1所示。其中，M1、M2、M3为大于0的整数，且M1<M2<M3。M1、M2、M3值的大小可以根据具体的应用场景确定，本申请实施例对此不作限制，表1中的数据仅为示例而非限定。例如，若确定的云平台的健康度为70，可以对照表1中该健康度等级和健康度分值的对应关系，确定当前的云平台的健康度对应的健康度等级为亚健康等级。然后Bumper可以根据该健康度等级确定该健康度等级对应的资源调整策略，通过查看表1中健康度等级与资源调整策略的对应关系，可以确定亚健康等级对应的资源调整策略为增容M1台VM。在确定相应的资源调整策略后，该Bumper可以根据该资源调整策略调整分配给应用的资源的数量，例如，若该资源调整策略为增容M1台VM，该Bumper可以调整启动M1台VM为应用提供服务。可选地，在每台VM加入服务后，该Bumper可以重新确定该云平台的健康度，以便于根据云平台的健康度的实时变化，确定针对最新的健康度数据的资源调整策略。可选地，在该每台VM加入服务的过程中，能够为应用提供服务的资源的数量在增大，系统的性能在变好，也就是说，可以满足更多的用户访问量，可选地，该Bumper可以同步调整ISW的大小，以满足更多用户的访问需求。因此，Bumper可以根据云平台的健康度的变化，逐步有节奏的进行增容，而不是等到系统面临崩溃的时候再紧急增容，从而导致增容速度赶不上并发压力增加的速度，而刚增容的服务器立即面临巨大的访问压 力而立即崩溃。因此，本申请实施例的云平台中管理资源的方法，能够在保证系统提供稳定的服务的情况下，进行有节奏的增容，同时在增容的过程中，还可以通过调整ISW的大小，提供稳定而高可用的服务。例如，在逐步增容过程中，健康度迅速提高，可以增大ISW的大小，以保证更多的用户享受到服务，或者在逐步增容过程中，用户的访问量持续骤增，可以通过减小ISW的大小以舒缓用户访问量骤增带来的压力。
表1
健康度等级 健康度分值 资源调整策略
健康 71～100 0
亚健康 66～70 M1
低风险 51～65 M2
高风险 0～50 M3
应理解，表1中的健康度分值和健康度等级的对应关系仅仅是为了示例，并不对本申请实施例构成任何限定，本申请实施例还可以采用81～100、71～80、61～70、0～60分别对应健康、亚健康、低风险和高风险四个健康度等级等，本申请实施例还可以采用0～200分或0～10分等来评估云平台的健康度，还应理解，本申请实施例还可以将云平台的健康度等级分为5级或3级等。表1中的资源调整策略也仅仅是为了示例，具体应用中，需要增加的资源量跟实际的应用场景和用户需求的不同而不同，本申请实施例对此不作限制。
因此，本申请实施例的云平台中管理资源的方法，能够根据云平台系统的运行状态信息，调整分配给应用的资源的数量和智能滑窗的大小，从而能够为用户提供稳定而高可用的服务。
可选地，作为一个实施例，该方法400还包括：
根据应用的PV的波动量和/或该云平台的剩余资源的数量，确定该云平台当前的运行状态属于紧急状态；
确定该云平台的预案库中与该云平台当前的运行状态匹配的紧急预案；
根据该紧急预案的解决方案，调整分配给该应用的资源的数量和该ISW。
具体地，该Bumper根据当前的应用的PV的波动量和/或云平台系统的剩余资源的数量，确定系统处于紧急状态，例如，该Bumper根据当前的PV的波动量分析当前PV的波动量下需要新增的资源的数量远大于云平台系统的剩余资源的数量时，确定云平台系统处于紧急状态，需要紧急增容。可选地，该Bumper可以从预案库中查找跟云平台系统当前的运行状态匹配的紧急预案，例如，该Bumper可以根据当前的用户访问量以及该应用的服务类型等因素确定当前云平台系统的运行状态与该紧急预案匹配。然后该Bumper可以根据预案库中该紧急预案的解决方案，管理分配给应用的资源的数量和ISW的大小。例如，若该预案库中该紧急预案的解决方案为增容4台VM，调整ISW为原来的2倍，那么该Bumper可以根据该解决方案对云平台系统的资源作出相应的调整。
可选地，作为一个实施例，该确定该云平台的预案库中与该云平台当前的运行状态 匹配的紧急预案，包括：
根据该应用的PV、该应用的服务类型、该云平台的剩余资源的数量和该应用的响应时间中的至少一项来确定该云平台当前的运行状态和该紧急预案匹配。
具体地，该Bumper可以根据该应用的PV、该应用的服务类型、云平台系统的剩余资源的数量和该应用的响应时间中的一项或几项确定该系统当前的运行状态和预案库中的紧急预案的状态匹配。例如，该Bumper可以在确定应用的服务类型一致，且应用的PV和紧急预案的应用的访问量匹配时，确定云平台系统的运行状态和该紧急预案匹配，从而可以确定该紧急预案的解决方案为当前系统的运行状态下的资源调整策略，然后可以根据该解决方案管理系统的资源和ISW的大小。
可选地，作为另一个实施例，该根据该云平台的健康度，确定云平台的健康度资源调整策略，还可以包括：
若该云平台的健康度小于第一健康度阈值大于第二健康度阈值，确定该云平台的健康度属于第一健康度等级，该第一健康度等级对应的资源调整策略为根据该云平台的健康度，调整智能滑窗ISW的大小；
若该云平台的健康度小于第二健康度阈值大于第三健康度阈值，确定该云平台的健康度属于第二健康度等级，该第二健康度等级对应的资源调整策略为根据该云平台的健康度，制定资源调整建议，向弹性伸缩系统发送该资源调整建议，以便于该弹性伸缩系统根据该资源调整建议调整分配给应用的资源的数量；
若该云平台的健康度小于第三健康度阈值，确定该云平台的健康度属于第三健康度等级，该第三健康度等级对应的资源调整策略为从该系统的预案库确定跟该云平台的健康度匹配的预案，根据处理该预案采用的解决方案调整分配给应用的资源的数量；
其中，该第二健康度阈值小于该第一健康度阈值，该第三健康度阈值小于该第二健康度阈值。
具体地，若该云平台的健康度小于第一健康度阈值大于第二健康度阈值，确定该云平台的健康度属于第一健康度等级，可选地，若该云平台的健康度分为健康、亚健康、低风险和高风险四个健康度等级区分，此时该云平台的健康度可以认为处于亚健康等级，若该云平台的健康度以100分值来划分，该第一健康度阈值可以为70分，该第二健康度阈值可以为65分。在这种状态下，系统虽然一段时间内能够提供高性能服务，但是如果PV持续增长，可能会引起系统性能的不足。可选地，该Bumper可以通过调整ISW的大小来疏缓访问请求骤增的冲击，例如，可以将当前的ISW调小一些，也就是降低该系统单位时间内允许的最大的用户访问数，换句话说，若系统的性能不是很差，此时可以通过调整ISW的大小来提高系统的性能；或者该缓冲器也可以向弹性伸缩系统发送该云平台的健康度信息，然后该弹性伸缩系统可以根据该云平台的健康度信息，制定相应的资源调整方案，从而调整分配给应用的资源的数量，例如，该弹性伸缩系统可以根据该健康度信息决定增容1台VM，以应对当前的系统状况。
若该云平台的健康度小于第二健康度阈值大于第三健康度阈值，确定该云平台的健康度属于第二健康度等级，可选地，若该云平台的健康度分为健康、亚健康、低风险和高风险四个健康度等级区分，此时该云平台的健康度等级可以认为处于低风险等级，若该云平台的健康度以100分值来划分，该第二健康度阈值可以为65分，该第三健康度阈值可以为50分。在这种状况下，可选地，该Bumper可以向弹性伸缩系统发送该云平台 的健康度信息，然后该弹性伸缩系统可以根据该云平台的健康度信息，制定相应的资源调整方案，调整分配给应用的资源的数量，例如，该弹性伸缩系统根据当前的云平台的健康度信息，确定需要增容1台VM；或者该Bumper也可以将该云平台的健康度信息跟预案库中的案例的健康度信息或服务类型等信息对比，确定出跟当前的系统的状态相匹配的预案，然后将该预案的解决方案作为资源调整策略，从而根据该资源调整策略调整分配给应用的资源的数量。换句话说，若系统的运行状态有变差的风险，通过调整ISW的大小，已经不能维持系统的高性能，此时，可以通过增加分配给应用的资源的数量和调整ISW大小的解决方案来保证系统的高性能，增容的数量可以根据云平台的健康度来确定，而不是一味增大分配给应用的资源的数量，造成部分系统的资源的闲置，而用户还需为这些增加的系统的资源支付大量费用，因此，本申请实施例的云平台中管理资源的方法，也提高了系统资源的利用率。
若云平台的健康度小于第三健康度阈值，确定该云平台的健康度属于第三健康度等级，可选地，若该云平台的健康度分为健康、亚健康、低风险和高风险四个健康度等级区分，此时该云平台的健康度可以认为处于高风险等级，若该云平台的健康度以100分值来划分，该第三健康度阈值可以为50分。在这种状况下，可选地，该Bumper可以将该云平台的健康度信息跟预案库中的案例的健康度信息对比，确定跟当前的系统的运行状态相匹配的预案，然后将该预案的解决方案作为资源调整策略，从而根据该资源调整策略调整分配给应用的资源的数量。可选地，该Bumper可以将该资源调整策略通过紧急通道发送给弹性伸缩系统，然后弹性伸缩系统可以根据该资源调整策略进行紧急增容。
应理解，调整ISW大小、增加VM数量等解决方案不是孤立的，可以结合起来执行，例如，可以先增加VM数量，再调整ISW大小；或者可以边增加VM数量，边调整ISW大小，或者也可以先调整ISW大小，再增加VM数量等，本申请实施例对此不作限制。
还应理解，以上示例仅表示三种可能的实现方式，并不对执行上述三种实现方式的条件作出限制，例如，在该健康度小于第一健康度阈值大于第二健康度阈值时，也可以通过匹配预案库中的紧急预案来调整分配给应用的资源的数量或ISW的大小。另外，本申请实施例仅以健康度越大，性能越优作为示例，而非限定，也可以设置健康度越大，系统性能越差，也就是健康度也可以和系统性能成反比等。还需要说明的是，以上实施例中的健康度阈值仅为示例，而非限定，各阈值的取值还可以根据实际应用场景或需求的不同而不同。
可选地，作为一个实施例，该云平台的第一状态信息可以包括以下中的至少一项：
该应用的PV的波动量、该云平台的剩余资源的数量、该应用的平均应答时间和该应用的服务健康系数β，该β为弹性伸缩系统反馈的该云平台的性能指标。
因为该云平台的健康度由该系统的第一状态信息确定，也就是说，该云平台的健康度也可以根据以下中的至少一项确定：
该应用的PV的波动量、该云平台的剩余资源的数量、该应用的平均应答时间和该应用的服务健康系数β，该β为弹性伸缩系统反馈的该系统的性能指标。
例如，该云平台的健康度可以用oH表示，该oH可以根据以下公式确定：
oH＝SUM(o×(1-Wcur)，
p×Ctotal(1-(CUsed+CIncreasing)/Ctotal)，
q×((RTavg-RTstd)/RTavg)，
r×β)
其中，该o、该p、该q，该r皆大于或等于0且小于1，SUM(o,p,q,r)＝1，SUM表示求和，也就是说o、p、q、r的和为1，该Wcur表示当前时间段内该应用的PV的波动量，该Ctotal为云平台总的资源的数量，该CUsed为已使用的资源的数量，该CIncreasing为下一时间段内需要增加的资源的数量，该RTavg为应用的平均应答时间，该RTstd为该应用的最大应答时间。β包括App系统、App所部署的服务器的中央处理器(英文全称：central processing unit简称：CPU)、存储器(memory)、磁盘(disk)等影响系统扩容的其他因素，该指标主要从弹性伸缩的监控系统或其他类似系统中获取。
具体地，若o＝60％，p＝30％，q＝10％，r＝0，那么该云平台的健康度可以根据以下公式确定：
oH＝SUM(60％×(1-Wcur),
30％×Ctotal(1-(CUsed+CIncreasing)/Ctotal),
10％×((RTavg-RTstd)/RTavg))
其中，若RTavg-RTstd<0，该项取0，oH∈[0，100]，oH小于0时，oH取值为0。
本申请实施例确定健康度采用的是631原则，也就是当前时间段内该App的PV的波动量占的权重为6，云平台剩余的资源的数量占的权重为3，App的平均应答时间占的权重为1。需要说明的是，本申请实施例以确定健康度采用631原则为例进行介绍仅仅是为了示例，而不应对本申请实施例构成任何限定，本申请实施例还可以采用811原则等，本申请实施例对此不作限制。
可选地，作为一个实施例，当前时间段内该App的PV的波动量Wcur可以根据以下公式(1)确定：
Wcur＝(PVcur-PVprev)/max(PVcur,PVprev)             (1)
其中，PVcur为当前时间段内应用的PV，PVprev为前一时间段内应用的PV。
可选地，作为一个实施例，单位时间段内该App的PV的平均波动量Wavg可以根据以下公式(2)确定：
Wavg＝Avg(W1，W2，…，Wn)            (2)
其中，该W1，W2，…，Wn为n个时间段的PV波动量,Avg表示求平均值。
可选地，作为一个实施例，下一个时间段内的PV可以根据以下公式(3)预测：
PVnext＝Sum(50％×|PVcur|,10％×|PVprev1|,10％×|PVprev2|，10％×|PVprev3|，10％×|PVprev4|，10％×|PVprev5|)           (3)
其中，PVnext为该下一个时间段内的PV，|PVprev1|、|PVprev2|、|PVprev3|、|PVprev4|和|PVprev5|为当前时间段前的5个时间段内应用的PV的绝对值；
本申请实施例预测PVnext采用的是515原则，也就是采集6个采样点的PV数据，给前5个采样点的权重设置1，给最近的采样点也就是当前采样点的权重设置5，需要说明的是，本申请实施例以预测PVnext采用515原则为例进行介绍仅仅是为了示例，而不应对本申请实施例构成任何限定，本申请实施例还可以采用811原则，也就是采集3个点，前两个点的权重为1，最近的采样点权重为8等，本申请实施例对此不作限制。
可选地，作为一个实施例，未来一段时间内该App的PV的增长量PVincreasing根据以下公式(4)确定：
PVincreasing＝(Wavg×PVnext)×Trun           (4)
其中，该Wavg为单位时间段内该App的PV的平均波动量，该Trun为服务器从启动到加入服务的时间。
可选地，作为一个实施例，该单位时间段内App的平均应答时间RTavg可以根据以下公式(5)确定：
RTavg＝Sum(80％×RTcur，10％×RTprev1，10％×RTprev2)           (5)
其中，该RTcur表示当前时间段内的该App的应答时间，该RTprev1和该RTprev2表示当前时间段前的2个时间段内该App的应答时间。
本申请实施例确定RTavg采用的是811原则，也就是采集3个采样点的RT数据，给前2个采样点的权重设置1，给最近的采样点也就是当前采样点的权重设置8，需要说明的是，本申请实施例以确定RTavg采用811原则为例进行介绍仅仅是为了示例，而不应对本申请实施例构成任何限定，本申请实施例还可以采用515原则，也就是采集6个点，前5个采样点的权重为1，最近的采样点权重为5等，本申请实施例对此不作限制。
图5示出了根据本申请一个具体实施例的云平台中管理资源的方法500的示意图，该方法500可以由图6中的各模块配合来执行。
如图6所示为简约的LB-Bumper的实现方案，LB-Bumper可以作为LB系统的一个模块单元进行工作，也可以与弹性伸缩系统对接，向该弹性伸缩系统发送健康度数据，以便于该弹性伸缩系统做出增容或缩容决策。可选地，该LB-Bumper还可以通过Keep-alive链路与弹性伸缩系统保持连接，例如，若云平台的健康度处于亚健康等级，该LB-Bumper可以通过普通通道和弹性伸缩系统进行通信，若遇到紧急情况，该LB-Bumper还可以通过紧急通道和弹性伸缩系统进行通信。
下面结合图5所示的具体实施例详细介绍根据本申请实施例的云平台中管理资源的方法，在该实施例中，设置云平台的健康度达到75时，停止增容。
S501，此时的PV较低，Bumper根据当前的应用的PV情况分析云平台的健康度为95，确定系统运行状态良好。可选地，该Bumper此时通过普通通道与弹性伸缩系统进行通信；
S502，开始有用户访问接入，但是此时的PV较低(2k/s)，也就是1s内的用户访问数为2k；
S503，Bumper根据当前的PV情况，确定当前云平台的健康度和ISW的大小，根据分析结果，确定当前系统的健康状态良好，当前的ISW大小为1w，也就是此时一次允许接入的用户访问数为1万；
S504，经过一段时间的运行，PV从2k/s骤增至10w/s，Bumper根据应用的PV的波动量分析这种状态可能会持续一段时间；
S505，Bumper根据应用的PV的波动量确定当前云平台的健康度分值为40，此时当前系统处于高风险状态，可选地，Bumper从预案库查找跟系统当前的运行状态匹配的紧急预案，将该紧急预案的解决方案作为当前系统的运行状况下的解决方案。例如，该解决方案为紧急增容4台VM。可选地，该Bumper可以将该解决方案通过紧急通道通知弹性伸缩系统；
S506，该弹性伸缩系统根据该解决方案，紧急增容4台VM；
在S507、S509、S510、S512中，4台VM不断加入服务，随着4台VM的不断入服，系统的性能越来越好，可选地，在4台VM不断加入服务的过程中，该Bumper还 可以根据系统的实际运行状况调整ISW的大小(例如，从1w/s→5k/s→3k/s→4.5k/s→1w/s)。
S508，处理一段时间后，云平台的健康度变为75，系统性能变好，到了S411，PV稳定在2w/s，云平台的健康度达到了80，也就是健康度大于75，因此不需要继续增容了；
S513，在之后的某个时间点，PV突然由2w/s骤增到5w/s；
S514，Bumper根据当前系统的运行状况分析，接下来的一段时间内，系统可能有超负荷或性能下降的风险；
S515，Bumper将当前的健康度信息发送给弹性伸缩系统，由于此时系统的运行状态还不是很差，因此，该Bumper可以通过普通通道向该弹性伸缩系统发送健康度信息；
S516，弹性伸缩系统根据该健康度信息确定增容1台VM；
S517、在增容1台VM后经过一段时间，PV稳定在3w/s；
S518、Bumper根据该系统的运行状况，调整ISW为1w/s；
S519、云平台的健康度为75，停止增容。
以上示例主要包括了两种场景下(一种是PV骤增的场景，另一种是PV小幅增加的场景)的云平台中管理资源的方法的实现方式，当PV骤增时，该Bumper可以根据通过匹配紧急预案，通知该弹性伸缩系统紧急增容，还可以在紧急增容的过程中不断调整ISW的大小，以疏缓PV骤然增大带来的系统冲击；当PV增幅不是很大时，该Bumper可以根据该云平台的健康度信息，确定增容数量，或者可以将该云平台的健康度信息推送到弹性伸缩系统，以便于该弹性伸缩系统根据该健康度信息确定增容策略，在增容的同时可以根据该系统的实时运行状况调整ISW的大小。
因此，本申请实施例的云平台中管理资源的方法，能够根据云平台系统的运行状态信息，调整分配给应用的资源的数量以及智能滑窗的大小，因此，在用户访问量骤增时，能够保证在增容的同时，通过调整ISW的大小提供稳定而高可用的服务。
图7示出了根据本申请实施例的云平台中管理资源的装置700的示意性框图，如图7所示，该装置700包括：
确定模块710，用于根据该云平台的第一状态信息，确定资源调整策略，该资源调整策略用于调整分配给该应用的资源的数量；
该确定模块710还用于根据该云平台的第二状态信息，确定智能滑窗ISW的大小，该ISW用于指示单位时间内该应用被允许的最大综合访问量PV；其中，该第一状态信息和该第二状态信息指示该云平台为该应用提供服务时的运行状况；
管理模块720，用于根据该资源调整策略，调整分配给该应用的资源的数量，并根据确定的该ISW的大小该调整该ISW。
具体地，该确定模块710与图3中的状态分析模块310的功能等同，该确定模块710的功能可以由软件程序实现，例如，该软件程序可以放在一个进程中实现，也可以由硬件芯片上的软件模块来实现，或者由硬件和软件模块的组合来实现等。可选地，该确定模块710可以集成在LB系统或弹性伸缩系统中，或者可以为独立的模块等。可选地，在Pass云平台中，可以为每个应用配置一个该确定模块710，也就是说该确定模块710可以是以应用为单位的，或者也可以几个应用共用一个确定模块710等，本申请实施例对此不作限制。该管理模块720与图2中的预案管理模块320和资源管理模块330的功能等同，该管理模块720的功能可以由软件程序实现，例如，该软件程序可以放在一个 进程中实现，也可以由硬件芯片上的软件模块来实现，或者由硬件和软件模块的组合来实现等。可选地，该管理模块720可以集成在LB系统或弹性伸缩系统中，为LB系统或弹性伸缩系统推送资源调整策略，或者可以为独立的模块等。可选地，在Pass云平台中，可以为每个应用配置一个管理模块720，也就是说管理模块720可以是以应用为单位的，或者也可以几个应用共用一个管理模块720等，本申请实施例对此不作限制。
因此，本申请实施例的云平台中管理资源的装置，能够根据系统的运行状态信息，调整提供应用服务的资源的数量以及智能滑窗的大小，因此，在用户访问量骤增时，能够保证在增容的同时，通过调整ISW的大小提供稳定而高可用的服务。
根据本申请实施例的云平台中管理资源的装置700可对应于根据本申请实施例的云平台中管理资源的方法400中的Bumper，并且装置700中的各个模块的上述和其它操作和/或功能分别为了实现前述各个方法的相应流程，为了简洁，在此不再赘述。
如图8所示，本申请实施例还提供了一种云平台中管理资源的装置800的示意性框图，该装置800包括：
收发器件、软件器件以及硬件器件部分；
收发器件为用于完成包收发的硬件电路；
硬件器件也可称“硬件处理模块”，或者更简单的，也可简称为“硬件”，硬件器件主要包括基于FPGA、ASIC之类专用硬件电路(也会配合其他配套器件，如存储器)来实现某些特定功能的硬件电路，其处理速度相比通用处理器往往要快很多，但功能一经定制，便很难更改，因此，实现起来并不灵活，通常用来处理一些固定的功能。需要说明的是，硬件器件在实际应用中，也可以包括MCU(微处理器，如单片机)、或者CPU等处理器，但这些处理器的主要功能并不是完成大数据的处理，而主要用于进行一些控制，在这种应用场景下，由这些器件搭配的系统为硬件器件。
软件器件(或者也简单“软件”)主要包括通用的处理器(例如CPU)及其一些配套的器件(如内存、硬盘等存储设备)，可以通过编程来让处理器具备相应的处理功能，用软件来实现时，可以根据业务需求灵活配置，但往往速度相比硬件器件来说要慢。软件处理完后，可以通过硬件器件将处理完的数据通过收发器件进行发送，也可以通过一个与收发器件相连的接口向收发器件发送处理完的数据。
在本申请实施例中，软件器件或者硬件器件用于进行上述实施例中提到的确定资源调整策略和ISW的大小，以及调整分配给应用的资源的数量和ISW。
通过本实施例软硬结合的方法，既保证了处理的速度，又具有灵活性。
因此，本申请实施例的云平台中管理资源的装置，能够根据云平台系统的运行状态信息，调整分配给应用的资源的数量以及智能滑窗的大小，因此，在用户访问量骤增时，能够保证在增容的同时，通过调整ISW的大小提供稳定而高可用的服务。
根据本申请实施例的云平台中管理资源的装置800可对应于根据本申请实施例的云平台中管理资源的方法400中的Bumper，并且装置800中的各个模块的上述和其它操作和/或功能分别为了实现前述各个方法的相应流程，为了简洁，在此不再赘述。
需要说明的是，本申请实施例提供的云平台中管理资源的装置，具体可以为云计算系统中的一台云主机，该云主机可以为运行在物理机上的虚拟机。如图9所示，物理机900包括硬件层910，运行在硬件层910之上的VMM(Virtual Machine Monitor，虚拟机监视器)920，以及运行在VMM 920之上的宿主机Host 901和若干虚拟机(VM，Virtual  Machine)，其中，硬件层包括但不限于：I/O设备、CPU和memory。本申请实施例提供的云平台中管理资源的装置具体可以为物理机900中的一台虚拟机，比如VM 940，VM 940上运行有一个或多个云应用，其中，每一个云应用都用于实现相应的业务功能，比如数据库应用、地图应用等等，这些云应用可以由开发者开发然后部署到云计算系统中。此外VM940还运行有可以执行程序，VM 940通过运行该可执行程序，并在程序运行的过程中通过宿主机Host 930来调用硬件层910的硬件资源，以实现云平台中的云平台中管理资源的装置的确定模块和管理模块的功能，具体而言，确定模块和管理模块可以以软件模块或函数的形式被包含在上述可执行程序中，比如该可执行程序可以包括：确定模块和管理模块，VM940通过调用硬件层910中的CPU、Memory等资源，以运行该可执行程序，从而实现确定模块和管理模块的功能，为了简洁，这里不再赘述。
如图10所示，本申请实施例还提供了一种管理资源的装置1000的示意性框图，该管理资源的装置1000包括处理器1010、存储器1020、总线系统1030、输入设备1040和输出设备1050。
存储器1020可以包括只读存储器和随机存取存储器，并向处理器1010提供指令和数据。存储器1020的一部分还可以包括非易失性随机存取存储器(NVRAM)。
存储器1020存储了如下的元素，可执行模块或者数据结构，或者它们的子集，或者它们的扩展集:
操作指令：包括各种操作指令，用于实现各种操作。
操作系统：包括各种系统程序，用于实现各种基础业务以及处理基于硬件的任务。
在本申请实施例中，处理器1010通过调用存储器1020存储的操作指令(该操作指令可存储在操作系统中)，执行如下操作：
根据该云平台的第一状态信息，确定资源调整策略，该资源调整策略用于调整分配给该应用的资源的数量；根据该云平台的第二状态信息，确定智能滑窗ISW的大小，该ISW用于指示单位时间内该应用被允许的最大综合访问量PV；其中，该第一状态信息和该第二状态信息指示该云平台为该应用提供服务时的运行状况；根据该资源调整策略，调整分配给该应用的资源的数量，并根据确定的该ISW的大小该调整该ISW。
处理器1010控制装置1000的操作，处理器1010还可以称为CPU(Central Processing Unit，中央处理单元)。存储器1020可以包括只读存储器和随机存取存储器，并向处理器1010提供指令和数据。存储器1020的一部分还可以包括非易失性随机存取存储器(NVRAM)。具体的应用中，装置1000的各个组件通过总线系统1030耦合在一起，其中总线系统1030除包括数据总线之外，还可以包括电源总线、控制总线和状态信号总线等。但是为了清楚说明起见，在图中将各种总线都标为总线系统1030。为便于表示，图10中仅用一条粗线表示，但并不表示总线系统1030仅有一根总线或一种类型的总线。
上述本申请实施例揭示的方法可以应用于处理器1010中，或者由处理器1010实现。处理器1010可能是一种集成电路芯片，具有信号的处理能力。在实现过程中，上述方法的各步骤可以通过处理器1010中的硬件的集成逻辑电路或者软件形式的指令完成。上述的处理器1010可以是通用处理器、数字信号处理器(DSP)、专用集成电路(ASIC)、现成可编程门阵列(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。可以实现或者执行本申请实施例中的公开的各方法、步骤及逻辑框图。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。结合本申请实施 例所公开的方法的步骤可以直接体现为硬件译码处理器执行完成，或者用译码处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器，闪存、只读存储器，可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器1020，处理器1010读取存储器1020中的信息，结合其硬件完成上述方法的步骤。
因此，本申请实施例的云平台中管理资源的装置，能够根据系统的运行状态信息，调整提供应用服务的资源的数量以及智能滑窗的大小，因此，在用户访问量骤增时，能够保证在增容的同时，通过调整ISW的大小提供稳定而高可用的服务。
根据本申请实施例的云平台中管理资源的装置1000可对应于根据本申请实施例的云平台中管理资源的方法400中的Bumper，并且装置1000中的各个模块的上述和其它操作和/或功能分别为了实现前述各个方法的相应流程，为了简洁，在此不再赘述。
应理解，本文中术语“和/或”，仅仅是一种描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B这三种情况。另外，本文中字符“/”，一般表示前后关联对象是一种“或”的关系。
应理解，在本申请的各种实施例中，上述各过程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。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统、装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统、装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，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，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，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，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案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)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应以所述权利要求的保护范围为准。

Claims (12)

  1. 一种云平台中管理资源的方法，其特征在于，所述云平台用于为部署在其上的应用提供运行所需的资源，包括：
    根据所述云平台的第一状态信息，确定资源调整策略，所述资源调整策略用于调整分配给所述应用的资源的数量；
    根据所述云平台的第二状态信息，确定智能滑窗ISW的大小，所述ISW用于指示单位时间内所述应用被允许的最大综合访问量PV；其中，所述第一状态信息和所述第二状态信息指示所述云平台为所述应用提供服务时的运行状况；
    根据所述资源调整策略，调整分配给所述应用的资源的数量，并根据确定的所述ISW的大小调整所述ISW。
  2. 根据权利要求1所述的方法，其特征在于，所述根据所述云平台的第一状态信息，确定资源调整策略，包括：
    根据所述第一状态信息，确定所述云平台的健康度，所述云平台的健康度指示所述云平台为所述应用提供服务时的性能状况；
    根据所述云平台的健康度，确定所述资源调整策略。
  3. 根据权利要求2所述的方法，其特征在于，所述根据所述云平台的健康度，确定所述资源调整策略，包括：
    确定多个健康度等级中所述云平台的健康度对应的健康度等级；
    根据所述云平台预配置的多个资源调整策略与所述多个健康度等级的对应关系，确定所述健康度等级对应的资源调整策略。
  4. 根据权利要求1至3中任一项所述的方法，其特征在于，所述方法还包括：
    根据应用的PV的波动量和/或所述云平台剩余资源的数量，确定所述云平台当前的运行状态属于紧急状态；
    确定所述云平台的预案库中与所述云平台当前的运行状态匹配的紧急预案；
    根据所述紧急预案的解决方案，调整分配给所述应用的资源的数量和所述ISW。
  5. 根据权利要求4所述的方法，其特征在于，所述确定所述云平台的预案库中与所述云平台当前的运行状态匹配的紧急预案，包括：
    根据所述应用的PV、所述应用的服务类型、所述云平台的剩余资源的数量和所述应用的响应时间中的至少一项来确定所述云平台当前的运行状态和所述紧急预案匹配。
  6. 根据权利要求1至5中任一项所述的方法，其特征在于，所述云平台的第一状态信息包括以下中的至少一项：
    所述应用的PV的波动量、所述云平台的剩余资源的数量、所述应用的平均应答时间和所述应用的服务健康系数β，所述β为弹性伸缩系统反馈的所述云平台的性能指标；
    所述云平台的第二状态信息包括以下中的至少一项：
    单位时间内所述应用被允许的最大PV、所述云平台的剩余的资源的数量和为所述应用提供服务的服务器从启动到加载入服的时间。
  7. 一种云平台中管理资源的装置，其特征在于，所述云平台用于为部署在其上的应用提供运行所需的资源，所述装置包括：
    确定模块，用于根据所述云平台的第一状态信息，确定资源调整策略，所述资源调整策略用于调整分配给所述应用的资源的数量；
    所述确定模块还用于根据所述云平台的第二状态信息，确定智能滑窗ISW的大小，所述ISW用于指示单位时间内所述应用被允许的最大综合访问量PV；其中，所述第一状态信息和所述第二状态信息指示所述云平台为所述应用提供服务时的运行状况；
    管理模块，用于根据所述资源调整策略，调整分配给所述应用的资源的数量，并根据确定的所述ISW的大小调整所述ISW。
  8. 根据权利要求7所述的装置，其特征在于，所述确定模块具体用于：
    根据所述第一状态信息，确定所述云平台的健康度，所述云平台的健康度指示所述云平台为所述应用提供服务时的性能状况；
    根据所述云平台的健康度，确定所述资源调整策略。
  9. 根据权利要求8所述的装置，其特征在于，所述确定模块还用于：
    确定多个健康度等级中所述云平台的健康度对应的健康度等级；
    根据所述云平台预配置的多个资源调整策略与所述多个健康度等级的对应关系，确定所述健康度等级对应的资源调整策略。
  10. 根据权利要求7至9中任一项所述的装置，其特征在于，所述确定模块还用于：
    根据应用的PV的波动量和/或所述云平台的剩余资源的数量，确定所述云平台当前的运行状态属于紧急状态；
    确定所述云平台的预案库中与所述云平台当前的运行状态匹配的紧急预案；
    根据所述紧急预案的解决方案，调整分配给所述应用的资源的数量和所述ISW。
  11. 根据权利要求10所述的装置，其特征在于，所述确定模块还用于：
    根据所述应用的PV、所述应用的服务类型、所述云平台的剩余资源的数量和所述应用的响应时间中的至少一项来确定所述云平台当前的运行状态和所述紧急预案匹配。
  12. 根据权利要求7至11中任一项所述的装置，其特征在于，所述云平台的第一状态信息包括以下中的至少一项：
    所述应用的PV的波动量、所述云平台的剩余资源的数量、所述应用的平均应答时间和所述应用的服务健康系数β，所述β为弹性伸缩系统反馈的所述云平台的性能指标；
    所述云平台的第二状态信息包括以下中的至少一项：
    单位时间内所述应用被允许的最大PV、所述云平台的剩余的资源的数量和为应用提供服务的服务器从启动到加载入服的时间。
PCT/CN2017/071274 2016-01-29 2017-01-16 云平台中管理资源的方法和装置 WO2017128980A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP17743595.5A EP3402163B1 (en) 2016-01-29 2017-01-16 Method and device for managing resources in cloud platform
US16/047,789 US10805385B2 (en) 2016-01-29 2018-07-27 Method and apparatus for managing resource on cloud platform

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610067172.XA CN107026877B (zh) 2016-01-29 2016-01-29 云平台中管理资源的方法和装置
CN201610067172.X 2016-01-29

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US16/047,789 Continuation US10805385B2 (en) 2016-01-29 2018-07-27 Method and apparatus for managing resource on cloud platform

Publications (1)

Publication Number Publication Date
WO2017128980A1 true WO2017128980A1 (zh) 2017-08-03

Family

ID=59397384

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/071274 WO2017128980A1 (zh) 2016-01-29 2017-01-16 云平台中管理资源的方法和装置

Country Status (4)

Country Link
US (1) US10805385B2 (zh)
EP (1) EP3402163B1 (zh)
CN (1) CN107026877B (zh)
WO (1) WO2017128980A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112328367A (zh) * 2020-11-11 2021-02-05 中国电子科技集团公司第十五研究所 一种云平台计算资源调度方法及系统
CN113760528A (zh) * 2020-12-24 2021-12-07 北京沃东天骏信息技术有限公司 基于多云平台的资源处理方法和装置

Families Citing this family (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109936592A (zh) * 2017-12-15 2019-06-25 中国移动通信集团浙江有限公司 服务处理的方法、装置、电子设备和存储介质
CN109213451B (zh) * 2018-09-11 2021-11-26 郑州云海信息技术有限公司 云数据系统中存储资源的管理方法和装置
US10768996B2 (en) * 2018-10-23 2020-09-08 Vmware, Inc. Anticipating future resource consumption based on user sessions
US11379254B1 (en) * 2018-11-18 2022-07-05 Pure Storage, Inc. Dynamic configuration of a cloud-based storage system
CN111381959A (zh) * 2018-12-29 2020-07-07 中兴通讯股份有限公司 扩容方法及装置
KR20200133478A (ko) 2019-05-20 2020-11-30 삼성전자주식회사 클라우드 시스템에서 자원을 조정하기 위한 장치 및 방법
CN110221917B (zh) * 2019-05-23 2023-02-28 创新先进技术有限公司 用于分配流式数据的方法及装置
CN112346846A (zh) * 2019-08-06 2021-02-09 陆明 对云资源进行分析和编排的方法、装置以及存储介质
CN112395075A (zh) * 2019-08-15 2021-02-23 阿里巴巴集团控股有限公司 资源的处理方法、装置以及资源调度系统
CN113448728B (zh) * 2021-06-22 2022-03-15 腾讯科技（深圳）有限公司 一种云资源调度方法、装置、设备及存储介质
CN115460217B (zh) * 2022-11-10 2023-07-14 军事科学院系统工程研究院网络信息研究所 一种基于强化学习的云服务高可用决策方法
CN116760832B (zh) * 2023-06-01 2024-02-13 兴容(上海)信息技术股份有限公司 一种基于多源节点的云平台资源管理系统及方法
CN117112236B (zh) * 2023-10-23 2024-02-20 山东曙光照信息技术股份有限公司 基于数据涌流及波动性预测的辖区服务器配置方法及系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103425535A (zh) * 2013-06-05 2013-12-04 浙江大学 云环境下的敏捷弹性伸缩方法
US20140101226A1 (en) * 2012-10-08 2014-04-10 Motorola Mobility Llc Methods and apparatus for performing dynamic load balancing of processing resources
CN103780646A (zh) * 2012-10-22 2014-05-07 中国长城计算机深圳股份有限公司 一种云资源的调度方法及系统
CN104168133A (zh) * 2014-07-11 2014-11-26 广州联智信息科技有限公司 一种动态配置api访问量的方法和网关及系统
CN105208111A (zh) * 2015-08-31 2015-12-30 联想(北京)有限公司 一种信息处理的方法及物理机

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9015324B2 (en) * 2005-03-16 2015-04-21 Adaptive Computing Enterprises, Inc. System and method of brokering cloud computing resources
CN102427475B (zh) * 2011-12-08 2014-01-29 无锡城市云计算中心有限公司 一种云计算环境中负载均衡调度的系统
US20150341230A1 (en) * 2013-03-15 2015-11-26 Gravitant, Inc Advanced discovery of cloud resources
US9363190B2 (en) * 2013-07-31 2016-06-07 Manjrasoft Pty. Ltd. System, method and computer program product for energy-efficient and service level agreement (SLA)-based management of data centers for cloud computing
US11386442B2 (en) * 2014-03-31 2022-07-12 Liveperson, Inc. Online behavioral predictor

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140101226A1 (en) * 2012-10-08 2014-04-10 Motorola Mobility Llc Methods and apparatus for performing dynamic load balancing of processing resources
CN103780646A (zh) * 2012-10-22 2014-05-07 中国长城计算机深圳股份有限公司 一种云资源的调度方法及系统
CN103425535A (zh) * 2013-06-05 2013-12-04 浙江大学 云环境下的敏捷弹性伸缩方法
CN104168133A (zh) * 2014-07-11 2014-11-26 广州联智信息科技有限公司 一种动态配置api访问量的方法和网关及系统
CN105208111A (zh) * 2015-08-31 2015-12-30 联想(北京)有限公司 一种信息处理的方法及物理机

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3402163A4 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112328367A (zh) * 2020-11-11 2021-02-05 中国电子科技集团公司第十五研究所 一种云平台计算资源调度方法及系统
CN112328367B (zh) * 2020-11-11 2023-08-04 中国电子科技集团公司第十五研究所 一种云平台计算资源调度方法及系统
CN113760528A (zh) * 2020-12-24 2021-12-07 北京沃东天骏信息技术有限公司 基于多云平台的资源处理方法和装置

Also Published As

Publication number Publication date
EP3402163A1 (en) 2018-11-14
US20180337984A1 (en) 2018-11-22
EP3402163A4 (en) 2018-11-14
CN107026877B (zh) 2020-07-07
EP3402163B1 (en) 2020-07-29
CN107026877A (zh) 2017-08-08
US10805385B2 (en) 2020-10-13

Similar Documents

Publication Publication Date Title
WO2017128980A1 (zh) 云平台中管理资源的方法和装置
US10541931B2 (en) Data congestion control in hierarchical sensor networks
US8595366B2 (en) Method and system for dynamically creating and servicing master-slave pairs within and across switch fabrics of a portable computing device
US10169102B2 (en) Load calculation method, load calculation program, and load calculation apparatus
WO2018120990A1 (zh) 业务部署的方法与装置
US9652150B2 (en) Global memory sharing method and apparatus, and communications system
US20120204176A1 (en) Method and device for implementing load balance of data center resources
CN111176792A (zh) 一种资源调度方法、装置及相关设备
US20120221730A1 (en) Resource control system and resource control method
US9898329B1 (en) Flexible processor association for virtual machines
CN104375897A (zh) 基于最小相对负载不均衡度的云计算资源调度方法
CN107251486A (zh) 一种扩展联动的方法、装置及系统
CN103414657A (zh) 一种跨数据中心的资源调度方法、超级调度中心和系统
US11809289B2 (en) High-availability (HA) management networks for high performance computing platforms
CN112600761A (zh) 一种资源分配的方法、装置及存储介质
WO2017214986A1 (zh) 一种云应用伸缩方法及装置
WO2017070869A1 (zh) 一种内存配置方法、装置及系统
CN114003374B (zh) 基于云平台的节点调度方法、装置及电子设备和存储介质
WO2016095162A1 (zh) 一种确定虚拟机数量调整操作的装置和方法
WO2021180127A1 (zh) 网络设备的扩容方法和相关装置
US12106147B2 (en) Resource allocation in a cloud computing system based on predictions of workload probability parameters
JP2019053346A (ja) 情報処理装置、情報処理方法、情報処理プログラム、および情報処理システム
CN114035906A (zh) 虚拟机迁移方法、装置、电子设备及存储介质
US20230354101A1 (en) Resource allocation device, resource allocation method, and control circuit
US20220357991A1 (en) Information processing apparatus, computer-readable recording medium storing aggregation control program, and aggregation control method

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: 17743595

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2017743595

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 2017743595

Country of ref document: EP

Effective date: 20180806