WO2013107141A1 - 一种用于云计算的策略调度系统和方法 - Google Patents

一种用于云计算的策略调度系统和方法 Download PDF

Info

Publication number
WO2013107141A1
WO2013107141A1 PCT/CN2012/076114 CN2012076114W WO2013107141A1 WO 2013107141 A1 WO2013107141 A1 WO 2013107141A1 CN 2012076114 W CN2012076114 W CN 2012076114W WO 2013107141 A1 WO2013107141 A1 WO 2013107141A1
Authority
WO
WIPO (PCT)
Prior art keywords
policy
virtual machine
target virtual
module
configuration module
Prior art date
Application number
PCT/CN2012/076114
Other languages
English (en)
French (fr)
Inventor
余志强
Original Assignee
中兴通讯股份有限公司
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 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2013107141A1 publication Critical patent/WO2013107141A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46Multiprogramming arrangements
    • G06F9/50Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5061Partitioning or combining of resources
    • G06F9/5072Grid computing

Definitions

  • the present invention relates to the field of computing resource scheduling, and in particular to a policy scheduling system and method for cloud computing.
  • Cloud computing is an Internet-based computing method in which shared hardware and software resources and information can be provided to computers and other devices as needed.
  • the core idea of cloud computing is to uniformly manage and schedule a large number of virtual machines connected by networks to form a computing resource pool to provide services to users on demand. Faced with a large number of virtual machine cloud computing environments, how to ensure flexible scheduling of scheduling policies and rational scheduling of virtual machine resources has become an important issue for those skilled in the art.
  • the present invention provides a policy scheduling system and method for cloud computing, to solve the technical problem of how to ensure flexible scheduling of scheduling policies and rational scheduling of virtual machine resources.
  • the present invention provides a policy scheduling system for cloud computing, where the system includes a policy configuration module, a monitoring module, a policy scheduling module, and a policy execution module, where
  • the policy configuration module is configured to: configure a policy for the target virtual machine in the cloud computing resource pool;
  • the monitoring module is configured to: learn the target virtual machine from the policy configuration module, monitor performance parameters of the target virtual machine, and Sending the performance parameter to the policy scheduling module;
  • the policy scheduling module is configured to: determine whether a target virtual machine performance parameter from the monitoring module meets a trigger condition of a policy corresponding to the target virtual machine, and meet a target corresponding to the target virtual machine in the target virtual machine performance parameter. When the trigger condition is met, a request message is sent to the policy execution module; The policy execution module is configured to: after receiving the request message from the policy scheduling module, execute a policy configured by the policy configuration module that satisfies the trigger condition.
  • the policy configured by the policy configuration module includes one or more of the following: an extended virtual machine, a shrink virtual machine, a virtual machine alarm record, and a virtual machine alarm notification;
  • the policy components configured by the policy configuration module include: a target virtual machine, a policy trigger condition, and a policy action.
  • the policy configuration module is further configured to: detect whether a configured policy conflicts, and send a policy normal indication to the monitoring module if it is detected that the configured policy does not conflict;
  • the monitoring module is further configured to: after receiving the policy normal indication from the policy configuration module, learn the target virtual machine from the policy configuration module.
  • the policy configuration module is further configured to: the configured policy component further includes: the number of times that the policy trigger condition is continuously met;
  • the policy scheduling module is further configured to: if the target virtual machine performance parameter meets the trigger condition, the number of times that the trigger condition of the policy is continuously met is determined, and whether the number of consecutively satisfying the trigger condition of the policy reaches a preset threshold And if the number of times that the trigger condition of the policy is continuously met reaches a preset threshold, the request message is sent to the policy execution module that executes the policy.
  • the policy configuration module is configured to: configure multiple target virtual machines to correspond to the same policy, and the multiple target virtual machines form a target virtual machine list;
  • the monitoring module is configured to: monitor performance parameters of the virtual machine included in the target virtual machine list every preset period;
  • the policy execution module is further configured to: when the policy configured by the policy configuration module for the multiple target virtual machines is a shrinking virtual machine, after performing the shrinking virtual machine policy according to the policy action included in the policy configured by the configuration module And notifying the policy configuration module to delete the shrunk virtual machine from the target virtual machine list composed of the multiple target virtual machines; and configuring the policy for the multiple target virtual machines to be extended virtual in the policy configuration module After the extended virtual machine policy is executed according to the policy action included in the policy configured by the configuration module, the policy configuration module is further notified to add the extended virtual machine to the target virtual machine list composed of the multiple target virtual machines.
  • the policy execution module is further configured to: when the policy configured by the policy configuration module for the multiple virtual machines is a shrink virtual machine, perform a shrink virtual machine according to a policy action included in a policy configured by the configuration module. Before the policy, the number of the target virtual machines in the target virtual machine list is detected, and the shrinking virtual machine policy is continued only when the number is greater than the set minimum configured target virtual machine number.
  • the present invention further provides a policy scheduling method for cloud computing, the method comprising the following steps:
  • the policy includes one or more of the following: an extended virtual machine, a shrink virtual machine, a virtual machine alarm record, and a virtual machine alarm notification;
  • the policy components include: a target virtual machine, a policy trigger condition, and a policy action.
  • the method further includes:
  • the target virtual machine performance parameter satisfies the trigger condition
  • the number of times that the trigger condition of the policy is continuously satisfied is counted, and it is determined whether the number of times that the trigger condition of the policy is continuously met reaches a preset threshold, and the policy is continuously satisfied.
  • This policy is executed only when the number of trigger conditions reaches the preset threshold.
  • the plurality of target virtual machines are configured to correspond to the same policy, and the plurality of target virtual units are the target virtual machine list;
  • the policy configured for the multiple target virtual machines is a shrinking virtual machine
  • the shrunken virtual machine is also deleted from the target virtual machine list composed of the plurality of target virtual machines
  • the extended virtual machine is also added to the target virtual machine list composed of the plurality of target virtual machines.
  • the policy configured for the multiple virtual machines is a shrinking virtual machine
  • detecting the number of target virtual machines in the target virtual machine list only if the number is greater than The shrink virtual machine policy continues to be executed when the minimum number of configured target virtual machines is set.
  • the above technical solution can flexibly set the policy; and the triggering of the policy is not only related to the performance of the virtual machine, but the performance is the current performance of the virtual machine, ensuring that the scheduling policy truly reflects the resource situation; only detecting the virtual machine with the policy set, avoiding Unnecessary consumption of the system.
  • FIG. 1 is a block diagram of a policy scheduling system for cloud computing according to the embodiment
  • FIG. 2 is a flowchart of a policy scheduling method for cloud computing according to the embodiment. Preferred embodiment of the invention
  • FIG. 1 is a block diagram of a policy scheduling system for cloud computing according to the embodiment.
  • the system includes a policy configuration module, a monitoring module, a policy scheduling module, and a policy execution module, where
  • the policy configuration module is configured to configure a policy for a target virtual machine in a cloud computing resource pool;
  • the policy configured by the policy configuration module may include one or more of the following: Reduce virtual machine, virtual machine alarm record, and virtual machine alarm notification; this policy can be flexibly set by the user based on the actual application scenario;
  • the policy component configured by the policy configuration module may include: a target virtual machine, a policy trigger condition, and a policy action; of course, the composition of the policy is not limited to the parameters listed above, and may further include a policy identifier (ID); wherein, the target virtual machine may For a virtual machine or multiple virtual machines, when the target virtual machine is multiple virtual machines, the target virtual machine at this time is called the target virtual machine;
  • ID policy identifier
  • the policy action is corresponding to the policy configured by the policy configuration module.
  • the corresponding policy action is to start a new virtual machine, complete initialization of the virtual machine, and load related applications.
  • the policy configured by the policy configuration module is to shrink the virtual machine, the corresponding policy action is to close the related application, clean up the virtual machine data, release the virtual machine, and the like;
  • the corresponding policy action may further include detecting the number of target virtual machines in the target virtual machine list before shrinking the virtual machine, where the number is greater than the setting.
  • the shrinking virtual machine operation is continued; to avoid that the target virtual machine is all shrunk, and the user's application cannot be guaranteed to operate normally;
  • the corresponding policy action may further include adding the extended virtual machine to the target virtual machine list formed by the multiple target virtual machines, so that the monitoring module performs the extended virtual machine. Performance monitoring
  • the corresponding policy action may further include deleting the contracted virtual machine from the target virtual machine list composed of the plurality of target virtual machines, so as to save the monitoring module from shrinking.
  • the monitoring of virtual machines avoids unnecessary waste of system resources;
  • the foregoing policy configuration module is further configured to detect whether the configured policy conflicts, and send a policy normal indication to the monitoring module if it detects that the configured policy does not conflict.
  • the monitoring module is configured to learn a target virtual machine from the policy configuration module, monitor performance parameters of the target virtual machine, and send the performance parameter to the policy scheduling module;
  • the monitoring module is further configured to: after receiving the policy normal indication from the policy configuration module, the slave policy configuration The module knows the target virtual machine;
  • the monitoring module can monitor the performance parameters of the virtual machine included in the target virtual machine list every preset period to obtain the performance status of the current target virtual machine, thereby ensuring the effectiveness and accuracy of the policy scheduling.
  • the policy scheduling module is configured to determine whether a target virtual machine performance parameter from the monitoring module meets a trigger condition of a policy corresponding to the target virtual machine, and the target virtual machine performance parameter satisfies a policy corresponding to the target virtual machine.
  • the requesting message is sent to the policy execution module; the request message may include a policy action that is obtained by the policy scheduling module from the policy configuration module, or a policy identifier corresponding to the trigger condition;
  • the trigger condition may be a preset virtual machine CPU usage, a memory usage rate, a hard disk usage rate, and the like, and the trigger condition is obtained by the policy scheduling module from the policy configuration module;
  • the policy execution module is configured to: after receiving the request message from the policy scheduling module, execute a policy configured by the policy configuration module that satisfies the trigger condition;
  • the policy execution module may include multiple policy execution sub-modules, where each policy execution sub-module executes different policies, such as setting a virtual machine management sub-module to execute an extended virtual machine, shrinking a virtual machine's policy, and setting a record alarm sub-module to execute.
  • the virtual machine alarm record and the virtual machine alarm notification policy; the policy execution module may execute the corresponding policy according to the policy action parameter included in the policy configured by the configuration module after receiving the request message; if the request message includes a policy action
  • the policy execution module can directly execute the policy according to the policy action; if the request message includes the policy identifier, the policy execution module can obtain the corresponding policy action from the policy configuration module according to the policy identifier, and then execute the policy.
  • the policy execution module may directly modify the target virtual machine list configured in the policy configuration module, or the policy execution module may send the policy identification module to the policy configuration module.
  • the notification is modified, and the policy configuration module modifies the configured target virtual machine list according to the policy identifier.
  • the configuration of the policy configured by the policy configuration module may include a parameter that continuously satisfies the triggering condition of the policy, in addition to the foregoing parameters;
  • the policy scheduling module at this time, if the target virtual machine performance parameter satisfies the trigger condition, statistics the number of times that the trigger condition of the policy is continuously met, and continues to determine whether the number of consecutively satisfying the trigger condition of the policy reaches a preset width.
  • the value is sent to the policy execution module that executes the policy, and the number of times that the policy trigger condition is continuously satisfied is cleared to zero when the number of times that the trigger condition of the policy is continuously reached reaches the preset threshold. This eliminates the misjudgment caused by peaks in the target virtual machine.
  • the policy scheduling module determines that the number of times that the triggering condition of the policy is continuously met does not reach the preset threshold, the number of times that the statistics have been continuously met to meet the triggering condition of the policy is also cleared to ensure that the statistic trigger condition is met for each statistic.
  • the number of times is the number of times that the trigger condition of this policy is continuously satisfied.
  • FIG. 2 is a flowchart of a policy scheduling method for cloud computing according to the embodiment.
  • S201 configures a policy for the target virtual machine in the cloud computing resource pool
  • the plurality of target virtual machines can be configured to correspond to the same policy, and the multiple target virtual machines form a target virtual machine list;
  • the policy includes one or more of the following: an extended virtual machine, a shrink virtual machine, a virtual machine alarm record, and a virtual machine alarm notification;
  • the policy component includes: a target virtual machine, a policy trigger condition, and a policy action;
  • step S201 After performing step S201, it is also possible to detect whether the configured policy conflicts, and if it is detected that the configured policy does not conflict, step S202 is performed;
  • S202 monitors performance parameters of the target virtual machine.
  • the performance parameters of the target virtual machine included in the target virtual machine list may be monitored every preset period
  • step S203 determining whether the performance parameter of the target virtual machine meets the triggering condition of the policy corresponding to the target virtual machine, and if the performance parameter of the target virtual machine meets the triggering condition of the policy corresponding to the target virtual machine, step S204 is performed; Otherwise, step S205 is performed;
  • the number of times that the trigger condition of the policy is continuously met may be counted, and whether the number of times that the trigger condition of the policy is continuously met reaches a preset threshold is continuously satisfied.
  • the number of trigger conditions reaches the preset threshold, Step S204;
  • S204 executes a policy that satisfies the trigger condition
  • the policy configured for the multiple virtual machines is to shrink the virtual machine
  • the number of the target virtual machines in the target virtual machine list is detected, only if the number is greater than the set
  • the shrinking virtual machine policy is continued; and after the shrinking virtual machine policy is executed, the shrunken virtual machine is also deleted from the target virtual machine list composed of the plurality of target virtual machines;
  • the extended virtual machine is also added to the target virtual machine list composed of the plurality of target virtual machines.
  • the S205 process ends.
  • each module/unit in the foregoing embodiment may be implemented in the form of hardware, or may use software functions.
  • the form of the module is implemented. The invention is not limited to any specific form of combination of hardware and software.
  • the above technical solution can flexibly set the policy; and the triggering of the policy is not only related to the performance of the virtual machine, but the performance is the current performance of the virtual machine, ensuring that the scheduling policy truly reflects the resource situation; only detecting the virtual machine with the policy set, avoiding Unnecessary consumption of the system.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Mathematical Physics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Debugging And Monitoring (AREA)

Abstract

一种用于云计算的策略调度系统和方法,其中,该系统包括策略配置、监控、策略调度以及策略执行模块;策略配置模块设置为:配置针对云计算资源池中目标虚拟机的策略;监控模块设置为:从策略配置模块获知目标虚拟机,监控目标虚拟机的性能参数,并将该性能参数发送至策略调度模块;策略调度模块设置为:判断来自监控模块的目标虚拟机性能参数是否满足与该目标虚拟机对应的策略的触发条件,并在目标虚拟机性能参数满足与该目标虚拟机对应的策略的触发条件时,向策略执行模块发出请求消息;策略执行模块设置为:接收到来自策略调度模块的请求消息后,执行策略配置模块配置的满足触发条件的策略。本发明可灵活配置策略,合理调度虚拟机资源。

Description

一种用于云计算的策略调度系统和方法
技术领域
本发明涉及计算资源调度领域, 尤其涉及一种用于云计算的策略调度系 统和方法。
背景技术
近年来, 随着互联网的不断发展, 一种新型服务计算模型一一云计算 ( cloud computing ) ,快速发展起来。云计算,是一种基于互联网的计算方式, 通过这种方式,共享的软硬件资源和信息可以按需提供给计算机和其他设备。 云计算的核心思想, 是将大量用网络连接的虚拟机统一管理和调度, 构成一 个计算资源池向用户按需服务。 面对大量的虚拟机的云计算环境, 如何保证调度策略的灵活设置、 合理 调度虚拟机资源成为本领域技术人员面临的一个重要问题。
发明内容
本发明提供了一种用于云计算的策略调度系统和方法, 以解决如何保证 调度策略的灵活设置、 合理调度虚拟机资源的技术问题。
为解决上述技术问题, 本发明提供了一种用于云计算的策略调度系统, 所述系统包括策略配置模块、 监控模块、 策略调度模块以及策略执行模块, 其中,
所述策略配置模块设置为:配置针对云计算资源池中目标虚拟机的策略; 所述监控模块设置为: 从策略配置模块获知所述目标虚拟机, 监控所述 目标虚拟机的性能参数, 并将该性能参数发送至策略调度模块;
所述策略调度模块设置为: 判断来自所述监控模块的目标虚拟机性能参 数是否满足与该目标虚拟机对应的策略的触发条件, 并在目标虚拟机性能参 数满足与该目标虚拟机对应的策略的触发条件时, 向策略执行模块发出请求 消息; 所述策略执行模块设置为: 接收到来自策略调度模块的请求消息后, 执 行策略配置模块配置的满足所述触发条件的策略。
优选地, 所述策略配置模块配置的策略包括以下一种或多种: 扩展虚拟 机、 收缩虚拟机、 虚拟机告警记录和虚拟机告警通知;
所述策略配置模块配置的策略组成包括: 目标虚拟机、 策略触发条件和 策略动作。 优选地, 所述策略配置模块还设置为: 检测配置的策略是否发生冲突, 以及在检测出配置的策略没有冲突的条件下向所述监控模块发送策略正常指 示;
所述监控模块还设置为:在接收到来自策略配置模块的策略正常指示后, 从策略配置模块获知目标虚拟机。
优选地, 所述策略配置模块还设置为: 配置的策略组成还包括: 连续满 足本策略触发条件的次数;
所述策略调度模块还设置为: 在目标虚拟机性能参数满足所述触发条件 的情况下, 统计连续满足本策略触发条件的次数, 判断连续满足本策略触发 条件的次数是否达到预设的阔值, 在连续满足本策略触发条件的次数达到预 设的阔值的情况下, 才向执行所述本策略的策略执行模块发出请求消息。
优选地, 所述策略配置模块设置为: 配置多个目标虚拟机对应同一条策 略, 所述多个目标虚拟机组成目标虚拟机列表;
所述监控模块设置为: 每隔一段预设周期监控目标虚拟机列表中包含的 虚拟机的性能参数;
所述策略执行模块还设置为: 在所述策略配置模块为所述多个目标虚拟 机配置的策略为收缩虚拟机时, 在根据配置模块配置的策略中包含的策略动 作执行收缩虚拟机策略后, 还通知所述策略配置模块将收缩的虚拟机从所述 多个目标虚拟机组成的目标虚拟机列表中删除; 以及在所述策略配置模块为 所述多个目标虚拟机配置策略为扩展虚拟机时, 在根据配置模块配置的策略 中包含的策略动作执行扩展虚拟机策略后, 还通知所述策略配置模块将扩展 的虚拟机添加至所述多个目标虚拟机组成的目标虚拟机列表。 优选地, 所述策略执行模块还设置为: 在所述策略配置模块为所述多个 虚拟机配置的策略为收缩虚拟机时, 在根据配置模块配置的策略中包含的策 略动作执行收缩虚拟机策略前, 检测所述目标虚拟机列表中目标虚拟机的个 数, 仅在所述个数大于设定的最低配置目标虚拟机个数时, 才继续执行收缩 虚拟机策略。
为解决上述技术问题,本发明还提供了一种用于云计算的策略调度方法, 所述方法包括以下步骤:
配置针对云计算资源池中目标虚拟机的策略;
监控所述目标虚拟机的性能参数;
判断所述目标虚拟机的性能参数是否满足与该目标虚拟机对应的策略的 触发条件, 在所述目标虚拟机的性能参数满足与该目标虚拟机对应的策略的 触发条件时, 执行满足所述触发条件的策略。
优选地, 所述策略包括以下一种或多种: 扩展虚拟机、 收缩虚拟机、 虚 拟机告警记录和虚拟机告警通知;
所述策略组成包括: 目标虚拟机、 策略触发条件和策略动作。
优选地, 所述方法还包括:
配置针对目标虚拟机的策略后, 检测配置的策略是否发生冲突, 在检测 出配置的策略没有冲突的条件下, 监控目标虚拟机的性能参数。
优选地, 在目标虚拟机性能参数满足所述触发条件的情况下, 统计连续 满足本策略触发条件的次数, 判断连续满足本策略触发条件的次数是否达到 预设的阔值, 在连续满足本策略触发条件的次数达到预设的阔值的情况下, 才执行本策略。
优选地, 配置多个目标虚拟机对应同一条策略 , 所述多个目标虚拟机组 成目标虚拟机列表;
每隔一段预设周期监控目标虚拟机列表中包含的目标虚拟机的性能参 数; 当为所述多个目标虚拟机配置的策略为收缩虚拟机时 , 在执行收缩虚拟 机策略后, 还将收缩的虚拟机从所述多个目标虚拟机组成的目标虚拟机列表 中删除;
当为所述多个目标虚拟机配置策略为扩展虚拟机时, 在执行扩展虚拟机 策略后, 还将扩展的虚拟机添加至所述多个目标虚拟机组成的目标虚拟机列 表。
优选地, 当为所述多个虚拟机配置的策略为收缩虚拟机时, 在执行收缩 虚拟机策略前, 检测所述目标虚拟机列表中目标虚拟机的个数, 仅在所述个 数大于设定的最低配置目标虚拟机个数时, 才继续执行收缩虚拟机策略。
上述技术方案可灵活设置策略;且策略的触发不仅与虚拟机的性能有关, 而且该性能是虚拟机的当前性能, 保证调度策略真实的反映资源情况; 仅对 设置了策略的虚拟机检测, 避免了系统不必要的消耗。 附图概述
图 1为本实施例的用于云计算的策略调度系统模块图;
图 2为本实施例的用于云计算的策略调度方法流程图。 本发明的较佳实施方式
为使本发明的目的、 技术方案和优点更加清楚明白, 下文中将结合附图 对本发明的实施例进行详细说明。 需要说明的是, 在不冲突的情况下, 本申 请中的实施例及实施例中的特征可以相互任意组合。
图 1为本实施例的用于云计算的策略调度系统模块图。
该系统包括策略配置模块、监控模块、 策略调度模块以及策略执行模块, 其中,
所述策略配置模块, 用于配置针对云计算资源池中目标虚拟机的策略; 所述策略配置模块配置的策略可包括以下一种或多种: 扩展虚拟机、 收 缩虚拟机、 虚拟机告警记录和虚拟机告警通知; 该策略可由用户根基实际应 用场景灵活设置;
所述策略配置模块配置的策略组成可包括: 目标虚拟机、 策略触发条件 和策略动作; 当然, 策略的组成不限于上述列举的参数, 还可包括策略标识 ( ID ) ; 其中, 目标虚拟机可为一台虚拟机或多台虚拟机, 当目标虚拟机为 多台虚拟机时, 称此时的目标虚拟机为目标虚拟机组;
上述策略动作与策略配置模块配置的策略是对应的, 如当策略配置模块 配置的策略为扩展虚拟机时, 对应的策略动作为启动新的虚拟机, 完成该虚 拟机的初始化, 以及加载相关应用等; 如当策略配置模块配置的策略为收缩 虚拟机时, 对应的策略动作为关闭相关应用, 清理虚拟机数据, 释放虚拟机 等;
当策略配置模块配置的策略为收缩虚拟机时, 对应的策略动作还可包括 在收缩虚拟机前, 检测所述目标虚拟机列表中目标虚拟机的个数, 仅在所述 个数大于设定的最低配置目标虚拟机个数时, 才继续执行收缩虚拟机操作; 以避免目标虚拟机全部被收缩, 无法保证用户的应用能正常运行;
当策略配置模块配置的策略为扩展虚拟机时, 对应的策略动作还可包括 将扩展的虚拟机添加至所述多个目标虚拟机组成的目标虚拟机列表, 以便监 控模块对扩展的虚拟机进行性能监控;
当策略配置模块配置的策略为收缩虚拟机时, 对应的策略动作还可包括 将收缩的虚拟机从所述多个目标虚拟机组成的目标虚拟机列表中删除, 以省 去监控模块对收缩的虚拟机的监控, 避免了系统资源的不必要浪费;
为保证策略的可用性, 上述策略配置模块, 还用于检测配置的策略是否 发生冲突, 以及在检测出配置的策略没有冲突的条件下向所述监控模块发送 策略正常指示。
所述监控模块, 用于从策略配置模块获知目标虚拟机, 监控所述目标虚 拟机的性能参数, 并将该性能参数发送至策略调度模块;
在策略配置模块可用于检测配置的策略是否发生冲突的情况下, 所述监 控模块, 还用于在接收到来自策略配置模块的策略正常指示后, 从策略配置 模块获知目标虚拟机;
上述监控模块可每隔一段预设周期监控目标虚拟机列表中包含的虚拟机 的性能参数, 以获取当前目标虚拟机的性能状况, 进而保证策略调度的有效 性和准确性。 所述策略调度模块, 用于判断来自所述监控模块的目标虚拟机性能参数 是否满足与该目标虚拟机对应的策略的触发条件, 并在目标虚拟机性能参数 满足与该目标虚拟机对应的策略的触发条件时, 向策略执行模块发出请求消 息; 所述请求消息可包括策略调度模块从策略配置模块获取的与该触发条件 对应的策略动作, 或者与该触发条件对应的策略标识;
所述触发条件可以是预设的虚拟机 CPU使用率、 内存使用率、硬盘使用 率等, 该触发条件由策略调度模块从策略配置模块获取;
所述策略执行模块, 用于接收到来自策略调度模块的请求消息后, 执行 策略配置模块配置的满足所述触发条件的策略;
所述策略执行模块可包括多个策略执行子模块, 其中每个策略执行子模 块执行不同的策略, 如设置虚拟机管理子模块执行扩展虚拟机、 收缩虚拟机 的策略, 设置记录告警子模块执行虚拟机告警记录和虚拟机告警通知策略; 策略执行模块可在接收到所述请求消息后, 根据配置模块配置的策略中 包含的策略动作这一参数执行相应的策略; 若请求消息中包含策略动作, 则 策略执行模块可直接根据该策略动作执行策略;若请求消息中包含策略标识, 则策略执行模块可根据策略标识从策略配置模块获取相应的策略动作, 再执 行策略。
若策略执行模块执行的策略中涉及对目标虚拟机列表的操作, 可由策略 执行模块直接对策略配置模块中配置的目标虚拟机列表进行修改, 也可由策 略执行模块向策略配置模块发送携带策略标识的修改通知, 再由策略配置模 块根据该策略标识对配置的目标虚拟机列表进行修改。
在本发明的其他实施例中, 策略配置模块配置的策略的组成中除包含上 述参数外, 还可包含连续满足本策略触发条件的次数这一参数; 此时的策略调度模块, 在目标虚拟机性能参数满足所述触发条件的情况 下, 统计连续满足本策略触发条件的次数, 并继续判断该连续满足本策略触 发条件的次数是否达到预设的阔值, 在连续满足本策略触发条件的次数达到 预设的阔值的情况下, 才向执行所述本策略的策略执行模块发出请求消息, 此时可对连续满足策略触发条件的次数清零; 这样可消除目标虚拟机出现峰 值所造成的误判断。 若策略调度模块判断出连续满足本策略触发条件的次数 没有达到预设的阔值,也要对已统计的连续满足本策略触发条件的次数清零, 以保证每次统计的满足策略触发条件的次数是连续满足本策略触发条件的次 数。
图 2为本实施例的用于云计算的策略调度方法流程图。
S201配置针对云计算资源池中目标虚拟机的策略;
可配置多个目标虚拟机对应同一条策略, 所述多个目标虚拟机组成目标 虚拟机列表;
所述策略包括以下一种或多种: 扩展虚拟机、 收缩虚拟机、 虚拟机告警 记录和虚拟机告警通知;
所述策略组成包括: 目标虚拟机、 策略触发条件和策略动作;
在执行步骤 S201后,还可检测配置的策略是否发生冲突,在检测出配置 的策略没有冲突的条件下, 执行步骤 S202;
S202监控所述目标虚拟机的性能参数;
可每隔一段预设周期监控上述目标虚拟机列表中包含的目标虚拟机的性 能参数;
S203判断所述目标虚拟机的性能参数是否满足与该目标虚拟机对应的策 略的触发条件, 若所述目标虚拟机的性能参数满足与该目标虚拟机对应的策 略的触发条件, 执行步骤 S204;否则, 执行步骤 S205;
在目标虚拟机性能参数满足所述触发条件的情况下, 还可先统计连续满 足本策略触发条件的次数, 判断连续满足本策略触发条件的次数是否达到预 设的阔值, 在连续满足本策略触发条件的次数达到预设的阔值的情况下, 执 行步骤 S204;
S204执行满足所述触发条件的策略;
当为所述多个虚拟机配置的策略为收缩虚拟机时, 在执行收缩虚拟机策 略前, 检测所述目标虚拟机列表中目标虚拟机的个数, 仅在所述个数大于设 定的最低配置目标虚拟机个数时, 才继续执行收缩虚拟机策略; 并在执行收 缩虚拟机策略后, 还将收缩的虚拟机从所述多个目标虚拟机组成的目标虚拟 机列表中删除;
当为所述多个目标虚拟机配置策略为扩展虚拟机时, 在执行扩展虚拟机 策略后, 还将扩展的虚拟机添加至所述多个目标虚拟机组成的目标虚拟机列 表。
S205 流程结束。
本领域普通技术人员可以理解上述方法中的全部或部分步骤可通过程序 来指令相关硬件完成, 所述程序可以存储于计算机可读存储介质中, 如只读 存储器、 磁盘或光盘等。 可选地, 上述实施例的全部或部分步骤也可以使用 一个或多个集成电路来实现, 相应地, 上述实施例中的各模块 /单元可以釆用 硬件的形式实现, 也可以釆用软件功能模块的形式实现。 本发明不限制于任 何特定形式的硬件和软件的结合。
需要说明的是, 本发明还可有其他多种实施例, 在不背离本发明精神及 和变形, 但这些相应的改变和变形都应属于本发明所附的权利要求的保护范 围。
工业实用性
上述技术方案可灵活设置策略;且策略的触发不仅与虚拟机的性能有关, 而且该性能是虚拟机的当前性能, 保证调度策略真实的反映资源情况; 仅对 设置了策略的虚拟机检测, 避免了系统不必要的消耗。

Claims

权 利 要 求 书
1、 一种用于云计算的策略调度系统, 所述系统包括策略配置模块、 监 控模块、 策略调度模块以及策略执行模块, 其中,
所述策略配置模块设置为:配置针对云计算资源池中目标虚拟机的策略; 所述监控模块设置为: 从策略配置模块获知所述目标虚拟机, 监控所述 目标虚拟机的性能参数, 并将该性能参数发送至策略调度模块;
所述策略调度模块设置为: 判断来自所述监控模块的目标虚拟机性能参 数是否满足与该目标虚拟机对应的策略的触发条件, 并在目标虚拟机性能参 数满足与该目标虚拟机对应的策略的触发条件时, 向策略执行模块发出请求 消息;
所述策略执行模块设置为: 接收到来自策略调度模块的请求消息后, 执 行策略配置模块配置的满足所述触发条件的策略。
2、 如权利要求 1所述的系统, 其中,
所述策略配置模块配置的策略包括以下一种或多种: 扩展虚拟机、 收缩 虚拟机、 虚拟机告警记录和虚拟机告警通知;
所述策略配置模块配置的策略组成包括: 目标虚拟机、 策略触发条件和 策略动作。
3、 所述权利要求 1所述的系统, 其中,
所述策略配置模块还设置为: 检测配置的策略是否发生冲突, 以及在检 测出配置的策略没有冲突的条件下向所述监控模块发送策略正常指示;
所述监控模块还设置为:在接收到来自策略配置模块的策略正常指示后, 从策略配置模块获知目标虚拟机。
4、 如权利要求 2所述的系统, 其中,
所述策略配置模块还设置为: 配置的策略组成还包括: 连续满足本策略 触发条件的次数;
所述策略调度模块还设置为: 在目标虚拟机性能参数满足所述触发条件 的情况下, 统计连续满足本策略触发条件的次数, 判断连续满足本策略触发 条件的次数是否达到预设的阔值, 在连续满足本策略触发条件的次数达到预 设的阔值的情况下, 才向执行所述本策略的策略执行模块发出请求消息。
5、 如权利要求 2或 4所述的系统, 其中,
所述策略配置模块设置为: 配置多个目标虚拟机对应同一条策略, 所述 多个目标虚拟机组成目标虚拟机列表;
所述监控模块设置为: 每隔一段预设周期监控目标虚拟机列表中包含的 虚拟机的性能参数;
所述策略执行模块还设置为: 在所述策略配置模块为所述多个目标虚拟 机配置的策略为收缩虚拟机时, 在根据配置模块配置的策略中包含的策略动 作执行收缩虚拟机策略后, 还通知所述策略配置模块将收缩的虚拟机从所述 多个目标虚拟机组成的目标虚拟机列表中删除; 以及在所述策略配置模块为 所述多个目标虚拟机配置策略为扩展虚拟机时, 在根据配置模块配置的策略 中包含的策略动作执行扩展虚拟机策略后, 还通知所述策略配置模块将扩展 的虚拟机添加至所述多个目标虚拟机组成的目标虚拟机列表。
6、 如权利要求 5所述的系统, 其中,
所述策略执行模块还设置为: 在所述策略配置模块为所述多个虚拟机配 置的策略为收缩虚拟机时, 在根据配置模块配置的策略中包含的策略动作执 行收缩虚拟机策略前, 检测所述目标虚拟机列表中目标虚拟机的个数, 仅在 所述个数大于设定的最低配置目标虚拟机个数时, 才继续执行收缩虚拟机策 略。
7、 一种用于云计算的策略调度方法, 所述方法包括以下步骤: 配置针对云计算资源池中目标虚拟机的策略;
监控所述目标虚拟机的性能参数;
判断所述目标虚拟机的性能参数是否满足与该目标虚拟机对应的策略的 触发条件, 在所述目标虚拟机的性能参数满足与该目标虚拟机对应的策略的 触发条件时, 执行满足所述触发条件的策略。
8、 如权利要求 7所述的方法, 其中,
所述策略包括以下一种或多种: 扩展虚拟机、 收缩虚拟机、 虚拟机告警 记录和虚拟机告警通知;
所述策略组成包括: 目标虚拟机、 策略触发条件和策略动作。
9、 如权利要求 7所述的方法, 其中, 所述方法还包括:
配置针对目标虚拟机的策略后, 检测配置的策略是否发生冲突, 在检测 出配置的策略没有冲突的条件下, 监控目标虚拟机的性能参数。
10、 如权利要求 8所述的方法, 其中,
在目标虚拟机性能参数满足所述触发条件的情况下, 统计连续满足本策 略触发条件的次数, 判断连续满足本策略触发条件的次数是否达到预设的阔 值, 在连续满足本策略触发条件的次数达到预设的阔值的情况下, 才执行本 策略。
11、 如权利要求 8或 10所述的方法, 其中,
配置多个目标虚拟机对应同一条策略, 所述多个目标虚拟机组成目标虚 拟机列表;
每隔一段预设周期监控目标虚拟机列表中包含的目标虚拟机的性能参 数;
当为所述多个目标虚拟机配置的策略为收缩虚拟机时, 在执行收缩虚拟 机策略后, 还将收缩的虚拟机从所述多个目标虚拟机组成的目标虚拟机列表 中删除;
当为所述多个目标虚拟机配置策略为扩展虚拟机时, 在执行扩展虚拟机 策略后, 还将扩展的虚拟机添加至所述多个目标虚拟机组成的目标虚拟机列 表。
12、 如权利要求 11所述的方法, 其中,
当为所述多个虚拟机配置的策略为收缩虚拟机时, 在执行收缩虚拟机策 略前, 检测所述目标虚拟机列表中目标虚拟机的个数, 仅在所述个数大于设 定的最低配置目标虚拟机个数时, 才继续执行收缩虚拟机策略。
PCT/CN2012/076114 2012-01-19 2012-05-25 一种用于云计算的策略调度系统和方法 WO2013107141A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201210017636.8A CN102609309B (zh) 2012-01-19 2012-01-19 一种用于云计算的策略调度系统和方法
CN201210017636.8 2012-01-19

Publications (1)

Publication Number Publication Date
WO2013107141A1 true WO2013107141A1 (zh) 2013-07-25

Family

ID=46526703

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2012/076114 WO2013107141A1 (zh) 2012-01-19 2012-05-25 一种用于云计算的策略调度系统和方法

Country Status (2)

Country Link
CN (1) CN102609309B (zh)
WO (1) WO2013107141A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016101597A1 (zh) * 2014-12-23 2016-06-30 华为技术有限公司 一种虚拟机vm的伸缩方法和设备
CN106095581A (zh) * 2016-06-18 2016-11-09 南京采薇且歌信息科技有限公司 一种私有云条件下的网络存储虚拟化调度方法

Families Citing this family (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103164271A (zh) * 2011-12-12 2013-06-19 台达电子工业股份有限公司 虚拟机器的搬移位置计算程序的触发方法
CN102932413B (zh) * 2012-09-26 2016-03-30 华为软件技术有限公司 一种计算资源分配方法、云管理平台节点和计算资源集群
CN103581183B (zh) * 2013-10-30 2017-01-04 华为技术有限公司 一种虚拟化安全隔离方法与装置
CN104796274B (zh) * 2014-01-21 2019-05-03 华为技术有限公司 运维操作的匹配方法及装置
CN104951351B (zh) * 2014-03-28 2018-12-18 株式会社日立制作所 虚拟机调度器、调度需求管理器及方法和虚拟机调度系统
MX364510B (es) * 2014-04-14 2019-04-29 Huawei Tech Co Ltd Método y aparato para configurar solución de redundancia en arquitectura de cómputo en nube.
CN105357097A (zh) * 2014-08-19 2016-02-24 中兴通讯股份有限公司 虚拟网络的调节方法及系统
TWI519943B (zh) * 2014-10-24 2016-02-01 Virtual machine automatic expansion system and method
CN104954478A (zh) * 2015-06-23 2015-09-30 普元信息技术股份有限公司 云计算平台中实现服务器自动纵向伸缩的系统及方法
CN105159759B (zh) * 2015-08-31 2019-01-08 华为软件技术有限公司 应用实例部署方法和装置
CN105490846B (zh) * 2015-12-07 2019-11-05 中国电子科技集团公司第三十二研究所 服务器虚拟化管理系统及方法
CN107807874A (zh) * 2017-10-19 2018-03-16 国云科技股份有限公司 一种虚拟机行为监测及锁定方法
CN110427356B (zh) * 2018-04-26 2021-08-13 中移(苏州)软件技术有限公司 一种参数配置方法和设备

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102158535A (zh) * 2011-02-10 2011-08-17 浪潮(北京)电子信息产业有限公司 一种云计算操作系统
CN102164184A (zh) * 2011-04-22 2011-08-24 广州杰赛科技股份有限公司 云计算网络中计算机实体的接入和管理方法及云计算网络
CN102262567A (zh) * 2010-05-24 2011-11-30 中兴通讯股份有限公司 虚拟机调度决策的系统、平台及方法

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101753461B (zh) * 2010-01-14 2012-07-25 中国建设银行股份有限公司 实现负载均衡的方法、负载均衡服务器以及集群系统
CN101938416B (zh) * 2010-09-01 2012-08-08 华南理工大学 一种基于动态重配置虚拟资源的云计算资源调度方法
CN101986274B (zh) * 2010-11-11 2012-10-03 东软集团股份有限公司 一种私有云环境下资源调配系统及资源调配方法
CN102082692B (zh) * 2011-01-24 2012-10-17 华为技术有限公司 基于网络数据流向的虚拟机迁移方法、设备和集群系统
CN102081554A (zh) * 2011-01-30 2011-06-01 浪潮(北京)电子信息产业有限公司 云计算操作系统及其内核控制系统及方法

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102262567A (zh) * 2010-05-24 2011-11-30 中兴通讯股份有限公司 虚拟机调度决策的系统、平台及方法
CN102158535A (zh) * 2011-02-10 2011-08-17 浪潮(北京)电子信息产业有限公司 一种云计算操作系统
CN102164184A (zh) * 2011-04-22 2011-08-24 广州杰赛科技股份有限公司 云计算网络中计算机实体的接入和管理方法及云计算网络

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016101597A1 (zh) * 2014-12-23 2016-06-30 华为技术有限公司 一种虚拟机vm的伸缩方法和设备
US10789084B2 (en) 2014-12-23 2020-09-29 Huawei Technologies Co., Ltd. Virtual machine scaling method and device
CN106095581A (zh) * 2016-06-18 2016-11-09 南京采薇且歌信息科技有限公司 一种私有云条件下的网络存储虚拟化调度方法
CN106095581B (zh) * 2016-06-18 2022-07-26 南京采薇且歌信息科技有限公司 一种私有云条件下的网络存储虚拟化调度方法

Also Published As

Publication number Publication date
CN102609309B (zh) 2018-08-21
CN102609309A (zh) 2012-07-25

Similar Documents

Publication Publication Date Title
WO2013107141A1 (zh) 一种用于云计算的策略调度系统和方法
EP3340057B1 (en) Container monitoring method and apparatus
US9104480B2 (en) Monitoring and managing memory thresholds for application request threads
CN108039964B (zh) 基于网络功能虚拟化的故障处理方法及装置、系统
WO2019237593A1 (zh) 任务处理方法、装置、计算机设备及存储介质
US7587718B1 (en) Method and apparatus for enforcing a resource-usage policy in a compute farm
TWI559153B (zh) 分散式計算架構
JP2015510201A (ja) クラウドネットワークにおける迅速な災害回復準備のための方法および装置
JP4811830B1 (ja) コンピュータリソース制御システム
WO2016058318A1 (zh) 虚拟机vm资源弹性伸缩处理方法、装置及系统
CA2975242A1 (en) System, virtualization control apparatus, method for controlling a virtualization control apparatus, and program
US10305974B2 (en) Ranking system
US20190260634A1 (en) Service state transition method and apparatus
KR101371068B1 (ko) 클라우드 컴퓨팅 자원 관리를 위한 모니터링 메트릭을 활용한 트리거링 방법 및 그 시스템
CN106357757B (zh) 一种监控系统中的被监控装置及监控方法
EP3576359B1 (en) Method and device for volume expansion of virtual network function
Lung et al. Improving software performance and reliability in a distributed and concurrent environment with an architecture-based self-adaptive framework
US20220182851A1 (en) Communication Method and Apparatus for Plurality of Administrative Domains
US20170111240A1 (en) Service Elastic Method and Apparatus in Cloud Computing
WO2017019018A1 (en) Virtualized network function autoscaling based on lightweight threshold crossing notifications
WO2012087104A1 (en) Intelligent load handling in cloud infrastructure using trend analysis
CN105635231A (zh) 一种分布式系统的调用方法和装置
US9317355B2 (en) Dynamically determining an external systems management application to report system errors
JP2015121963A (ja) 情報処理システム、監視方法、及び、プログラム
WO2017002812A1 (ja) 仮想化インフラストラクチャ管理装置、仮想ネットワークファンクション管理装置、仮想マシンの管理方法及びプログラム

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: 12866147

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 12866147

Country of ref document: EP

Kind code of ref document: A1