WO2022001295A1 - 云环境管理方法、云环境管理平台及存储介质 - Google Patents

云环境管理方法、云环境管理平台及存储介质 Download PDF

Info

Publication number
WO2022001295A1
WO2022001295A1 PCT/CN2021/087767 CN2021087767W WO2022001295A1 WO 2022001295 A1 WO2022001295 A1 WO 2022001295A1 CN 2021087767 W CN2021087767 W CN 2021087767W WO 2022001295 A1 WO2022001295 A1 WO 2022001295A1
Authority
WO
WIPO (PCT)
Prior art keywords
cloud environment
idle
usage
state
cloud
Prior art date
Application number
PCT/CN2021/087767
Other languages
English (en)
French (fr)
Inventor
顾谊
陈飞飞
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 WO2022001295A1 publication Critical patent/WO2022001295A1/zh

Links

Images

Classifications

    • 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]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5011—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
    • G06F9/5016—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals the resource being the memory
    • 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]
    • 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]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5011—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
    • G06F9/5022—Mechanisms to release resources
    • 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]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • 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
    • 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
    • H04L43/0876—Network utilisation, e.g. volume of load or congestion level
    • 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
    • G06F2209/00—Indexing scheme relating to G06F9/00
    • G06F2209/50—Indexing scheme relating to G06F9/50
    • G06F2209/5011—Pool

Definitions

  • Embodiments of the present disclosure relate to, but are not limited to, the field of computers.
  • an embodiment of the present disclosure provides a cloud environment management method, including: acquiring usage data of the cloud environment, where the usage data can represent the current usage of the cloud environment; determining a usage index of the cloud environment according to the usage data of the cloud environment; The usage index of the cloud environment determines the current state of the cloud environment; if the cloud environment is currently in an idle state, the cloud environment is recycled.
  • an embodiment of the present disclosure also provides a cloud environment management platform, where the cloud environment management platform includes a processor, a memory, and a communication bus; the communication bus is used to implement connection and communication between the processor and the memory; the processor is used to execute the memory One or more programs stored in the cloud environment to implement the steps of the above cloud environment management method.
  • an embodiment of the present disclosure further provides a storage medium, where the storage medium stores a cloud environment recycling program, and the cloud environment recycling program can be executed by one or more processors to implement the steps of the above cloud environment management method.
  • FIG. 1 is a flowchart of a cloud environment management method provided by an embodiment of the present disclosure
  • Fig. 2 is a kind of flow chart of evaluating the current state of cloud environment by the cloud environment management platform provided by the embodiment of the present disclosure
  • FIG. 3 is a flow chart of setting a state evaluation policy according to a policy setting instruction input by an administrator by a cloud environment management platform provided by an embodiment of the present disclosure
  • FIG 5 is another interactive interface of the cloud environment management platform according to an embodiment of the present disclosure.
  • FIG. 6 is a flowchart of a cloud environment management method provided by an embodiment of the present disclosure.
  • FIG. 7 is a schematic diagram of the principle of performing state evaluation based on two state evaluation strategies according to an embodiment of the present disclosure
  • FIG. 8 is a schematic diagram of a hardware structure of a cloud environment management platform provided by an embodiment of the present disclosure.
  • FIG. 1 is a flowchart of a cloud environment management method provided by an embodiment of the present disclosure.
  • the cloud environment management method includes steps S102 to S108.
  • step S102 the cloud environment management platform obtains usage data of the cloud environment.
  • the usage data acquired by the cloud environment management platform is data that can represent the current usage situation of the corresponding cloud environment.
  • the occupancy of CPU (Central Processing Unit) resources in the cloud environment can represent the current usage of the cloud environment to a certain extent.
  • the cloud environment The occupancy rate of the cloud environment memory resources in the data obtained by the platform is 32%, and the occupancy rate of the cloud environment memory resources at the current moment is 58%, indicating that the cloud environment is used from the time t1 to the current time.
  • the network traffic of the cloud environment can also represent the usage of the cloud environment.
  • the situation that the cloud environment is logged in by the user can also reflect the situation that the cloud environment is used by the user to a certain extent. Therefore, in this embodiment, the usage data acquired by the cloud environment management platform that can represent the current usage of the cloud environment includes at least one of the above-mentioned data.
  • the cloud environment management platform may deploy an information collection agent in the corresponding cloud environment, and let the information collection agent help the collection in the cloud environment to characterize the CPU resource occupation, memory resource occupation, and storage resource occupation of the cloud environment. Use data of at least one of several types, and feed back the collected data to the cloud environment management platform.
  • the cloud environment management platform may obtain the communication address of the corresponding cloud environment, for example, the IP (Internet Protocol) address of the cloud environment, and the user name and login password for logging in to the corresponding cloud environment, and then follow the cloud The communication address, user name and login password of the environment log in to the corresponding cloud environment, and deploy the information collection agent in the corresponding cloud environment.
  • IP Internet Protocol
  • the usage data to be acquired by the cloud environment management platform includes data that can reflect the login status of the cloud environment, data collection in the cloud environment may not be required.
  • the cloud environment management platform will record the user's login information to the cloud environment, so the cloud environment can directly obtain this type of usage data without an information collection agent.
  • step S104 the cloud environment management platform determines the usage index of the cloud environment according to the usage data of the cloud environment.
  • the cloud environment management platform can process the usage data, and then abstract the usage indicators of the cloud environment.
  • the usage data acquired by the cloud environment management platform include: Some invalid data or some obviously wrong data, therefore, the cloud environment management platform will clean and filter the obtained usage data, filter out invalid data or obviously wrong data, and then analyze and process the remaining data , which abstracts the usage metrics of the cloud environment.
  • the usage indicators of the cloud environment include at least one of CPU utilization, memory utilization, total network traffic, login frequency, and recent access duration.
  • login frequency refers to the number or frequency of logins to the cloud environment within the most recent preset duration
  • cent access duration refers to the time difference between the time when the cloud environment was last accessed and the current time.
  • the usage data obtained in advance by the cloud environment should include data representing the CPU resource occupancy of the cloud environment;
  • the usage data pre-obtained by the cloud environment should include data representing the memory resource usage of the cloud environment;
  • the cloud environment pre-obtains The received usage data should include data that can reflect network traffic in the cloud environment. It can be seen that if the usage indicators of the cloud environment include at least one of CPU utilization, memory utilization, and the sum of network traffic, the cloud environment management platform needs to deploy an information collection agent in the cloud environment before acquiring the usage data of the cloud environment. If the usage data is required to be processed to obtain the login frequency and/or the recent access duration of the cloud environment, the usage data that should be acquired in advance by the cloud environment includes data representing the login status of the cloud environment.
  • step S106 the cloud environment management platform determines the current state of the cloud environment according to the usage index of the cloud environment.
  • the cloud environment management platform After the cloud environment management platform determines the usage index of the cloud environment, it can evaluate the current state of the cloud environment according to the usage index.
  • the usage state of the cloud environment includes an idle state and a non-idle state.
  • the so-called non-idle state is actually the use state. It is worth noting that, in this embodiment, if a cloud environment is judged to be in an idle state, it does not mean that the cloud environment is in an absolute idle state. The environment is idle.
  • the cloud environment platform may set a threshold for the usage indicator, which may also be called an idle standard value, and then use the value of the usage indicator of the cloud environment as the If the current idle value of the cloud environment meets the requirement of the idle standard value, it can be determined that the cloud environment is in a recyclable idle state. In other words, as long as the value of the corresponding usage indicator of the cloud environment meets the requirement of the indicator threshold, it can be determined that the cloud environment is in an idle state.
  • a threshold for the usage indicator which may also be called an idle standard value
  • the cloud environment management platform may convert all usage indicators into percentages: for example, set a maximum value for different usage indicators, and then determine the ratio of the cloud environment usage indicator value to the maximum value, In this way all the values of the used indicator are converted to a value between 0-1.
  • the values of all the usage indicators can be converted into a range of 1-100 or 1-1000, and the size of each usage indicator can be measured according to the percentage system or the thousand-point system.
  • the weight of different usage indicators not exactly the same.
  • the administrator sets a requirement that as long as the CPU utilization, memory utilization, and login frequency meet the requirements, the cloud environment can be determined to be in an idle state, but it does not care about the duration of recent access and the sum of network traffic.
  • the weight of the recent access duration and the sum of network traffic in evaluating the cloud environment status is 0.
  • the cloud environment management platform may refer to the process shown in FIG. 2 to evaluate the current state of the cloud environment, and the process may include steps S202 to S208.
  • step S202 the current idle value of the cloud environment is determined according to the usage index of the cloud environment and the state evaluation strategy.
  • the status evaluation strategy includes the weight of each usage indicator in determining the idle value of the cloud environment.
  • the idle value of the cloud environment can be determined according to the weight occupied by each index:
  • n is the total number of evaluations cloud current state by using the index
  • W i is the value of the i-th use of indicators is converted to uniform weights and measures
  • P i is the i th use of indicators weight
  • S is the idle cloud environment value.
  • step S204 it is determined whether the current idle value of the cloud environment reaches the idle standard value corresponding to the state evaluation strategy.
  • step S206 If the judgment result is yes, go to step S206, otherwise go to step S208.
  • Each state evaluation strategy has a corresponding idle standard value. If the idle value of a cloud environment under the state evaluation strategy reaches the idle standard value, it means that the cloud environment is in an idle state; otherwise, it means that the cloud environment is currently in a non-idle state. state.
  • step S206 it is determined that the cloud environment is currently in an idle state.
  • step S208 it is determined that the cloud environment is currently in a non-idle state.
  • the cloud environment may use only one state evaluation strategy to evaluate the current state of the cloud environment, but in other examples of this embodiment, the cloud environment may simultaneously use at least two state evaluation strategies to evaluate The current state of the cloud environment.
  • the indicators used in each state evaluation strategy are not exactly the same, but they may partially overlap.
  • the weights of these usage indicators in the two state evaluation strategies can also be different.
  • the cloud environment management platform adopts a state evaluation policy a and a state evaluation policy b to evaluate the state of the cloud environment, wherein the usage indicators concerned by the state evaluation policy a include CPU utilization, memory utilization, and recent access
  • the usage indicators of state evaluation strategy b are CPU utilization and total network traffic. Both state evaluation policies have CPU utilization, but the weights of CPU utilization in the two state evaluation policies are different.
  • the state evaluation policies there is an "OR" logical relationship between the state evaluation policies, that is, as long as the idle value of the cloud environment under any one of the state evaluation policies reaches the idle standard value corresponding to the state evaluation policy, it means that the cloud environment is idle. Status, currently available for recycling.
  • there is an "and" logical relationship between each state evaluation strategy that is, the cloud environment only has the idle value under one of the state evaluation strategies to reach the corresponding idle standard value, which does not mean that the cloud environment is idle.
  • the cloud environment is in an idle state only if the idle values of the cloud environment under all state evaluation policies reach the corresponding idle standard values.
  • the state evaluation policy may be set by an administrator.
  • the cloud environment management platform may include human-computer interaction devices (display devices and input devices). The input device inputs a policy setting instruction to the cloud environment management platform, and after receiving the policy setting instruction input by the administrator, the cloud environment management platform determines a state evaluation policy according to the instruction.
  • FIG. 3 is a flow chart of the cloud environment management platform provided by the embodiment of the present disclosure for setting a state evaluation policy according to a policy setting instruction input by an administrator. As shown in FIG. 3 , the flow may include steps S302 and S304.
  • step S302 the cloud environment management platform receives the policy setting instruction through the input device.
  • the policy setting instruction includes a level requirement of each usage indicator of the cloud environment that can be recycled, and a weight occupied by each usage indicator in determining the idle value of the cloud environment.
  • the cloud environment management platform has a friendly human-computer interaction interface, which can guide users when the administrator issues policy setting instructions.
  • the cloud environment management platform provides optional usage indicators through the display device. When setting a status assessment strategy, managers can select the usage that the status assessment strategy focuses on from these usage indicators. indicators, and then indicate the weight of the selected indicators and the idle level requirements in the interactive interface shown in Figure 5.
  • step S304 the cloud environment management platform determines a state evaluation policy according to the policy setting instruction.
  • the cloud environment management platform After the cloud environment management platform receives the user's policy setting instruction, it can parse the policy setting instruction, and then generate a state evaluation policy.
  • the administrator has set more than one state evaluation policy, he should also specify to the cloud environment management platform the logical relationship between the state evaluation policies when they are used to evaluate the cloud environment state.
  • step S108 if the cloud environment is currently in an idle state, the cloud environment management platform recycles the cloud environment.
  • the cloud environment management platform can recycle the cloud environment. It should be understood that reclaiming the cloud environment is actually reclaiming all the resources of the cloud environment together: cloud environment management The platform releases all the resources in the cloud environment, and then puts the recovered resources into the resource pool corresponding to the cloud platform for allocation when receiving cloud environment applications later.
  • the cloud environment management platform determines that a certain cloud environment is currently in a non-idle state, it can predict the time when the cloud environment enters the idle state, that is, predict the time when the cloud environment enters the recyclable state. time.
  • the cloud environment management platform determines that the idle value of the cloud environment is smaller than the standard value of the corresponding state evaluation policy, the current idle value of the cloud environment may be recorded and stored. In this way, when the cloud environment management platform needs to predict the recovery time of a certain cloud environment, the prediction can be made according to the recorded historical idle values.
  • the cloud environment management platform when predicting the cloud environment recovery time, it may be performed in combination with a prediction model.
  • the prediction model may be composed of one or more prediction functions.
  • the cloud environment management platform inputs the historical idle value of the cloud environment into the prediction function, and then The recycle time of the cloud environment can be obtained.
  • a prediction mode includes two or more prediction functions at the same time, and these prediction functions can be used independently. Therefore, the cloud environment management platform predicts the recovery time of a cloud environment Before, it is necessary to select one of the prediction functions of the prediction model as the final prediction function, and then the final prediction function can be used to determine the recovery time of the cloud environment.
  • the usage data of the cloud environment is obtained, then the usage index of the cloud environment is determined according to the usage data of the cloud environment, and the current state of the cloud environment is determined according to the usage index of the cloud environment. If it is determined that the cloud environment is currently in an idle state according to the usage indicator of the cloud environment, the cloud environment can be recycled. In this way, the recycled cloud environment can be applied by other users for other R&D activities.
  • the idle cloud environment is recovered as a whole, the reuse of the idle cloud environment is realized, and the infinite recycling of cloud environment resources can be realized under the condition of limited actual resources, which is beneficial to Improve the utilization of cloud environment resources and provide guarantee for enterprise R&D innovation.
  • FIG. 6 is a flowchart of a cloud environment management method provided by an embodiment of the present disclosure. As shown in FIG. 6 , the cloud environment management method may include steps S602 to S618.
  • step S602 the usage data of the cloud environment is collected.
  • the usage data collected by the cloud environment management platform must at least ensure that the following usage indicators of the cloud environment can be subsequently determined: CPU utilization, memory utilization, total network traffic, login frequency, and recent access duration.
  • the usage data collected by the cloud environment management platform also includes data that can reflect the utilization rate of storage resources in the cloud environment, or also includes data that can reflect the usage of IO (input and output) interfaces of the cloud environment. data.
  • step S604 the usage data of the cloud environment is processed to obtain the usage index of the cloud environment.
  • the cloud environment management platform cleans the collected usage data, filters out invalid data or obviously wrong data, and then analyzes and processes the remaining usage data to abstract the usage indicators of the cloud environment.
  • step S608 the idle value of the cloud environment under each state evaluation strategy is determined according to the usage index of the cloud environment.
  • the cloud environment management platform divides the idle levels of each usage indicator according to the value of each usage indicator. For example, see Table 2:
  • the idle level of the usage indicator if it is lower than 20% (and greater than or equal to 10%), the idle level of the usage indicator is low, and if it is lower than 10% (and greater than or equal to 5%) The idle level is medium, and below 5%, the idle level of the usage indicator is considered to be high.
  • the division of the idle level of other usage indicators is similar. For example, for the login frequency, if the number of logins in the most recent preset time period is less than 10, the idle level of the usage indicator is considered to be high. If the number of logins is between 10 and 20 In between, it is considered that the idle level of the usage indicator has reached the medium level... I won't go into details here.
  • the status evaluation strategy a only pays attention to the three usage indicators of CPU utilization, memory utilization and recent access duration.
  • the idle level requirements for these three usage indicators are high, medium and low respectively, and the weights are are are 0.9, 0.9 and 0.6.
  • the status evaluation strategy b focuses on four usage indicators: memory utilization, total network traffic, login frequency, and recent access duration.
  • the idle level requirements for these four indicators are medium, high, high, and low, respectively, with weights of 0.8, 0.4 and 0.6, 0.1.
  • the total score under a state evaluation strategy is 100 points. If there are n usage indicators, the corresponding score for each usage indicator is 100/n. For example, in the state evaluation strategy a, each usage The scores of the indicators are all 100/3, while in the state evaluation strategy b, the score of each used indicator is 25. Under a state evaluation strategy, if a usage indicator of the cloud environment meets the idle state requirement of the state evaluation strategy for the usage indicator, the cloud environment can obtain the score corresponding to the usage indicator; otherwise, the cloud environment should use The score corresponding to the indicator is 0. For example, under the state evaluation strategy b, if the recent access duration of a cloud environment reaches the lower level required by the state evaluation strategy b, the cloud environment can get 25 points corresponding to the cloud environment.
  • the idle value of a cloud environment under a state evaluation strategy can be determined according to the following formula:
  • S is the idle value of the cloud environment under the state evaluation strategy
  • n is the total number of indicators used under the state evaluation strategy
  • V i is the score corresponding to the ith usage indicator. It can be understood that the value of V i either 100 / n, it is either 0; P i of the i-th weight used indicators.
  • step S610 it is judged whether the cloud environment has an idle value under a certain state evaluation strategy, and whether the idle value reaches the idle standard value corresponding to the state evaluation strategy.
  • the idle value of the cloud environment E1 under the state evaluation strategy a is greater than the idle standard value of the state evaluation strategy a, so the cloud environment E1 is in an idle state.
  • the idle values of the cloud environment E2 under the state evaluation strategies a and b are both smaller than the corresponding idle standard values. Therefore, the cloud environment E2 is in a non-idle state.
  • FIG. 7 shows a schematic diagram of a principle of evaluating the cloud environment state under the state evaluation strategy a and the state evaluation strategy b based on the usage index k1 , the usage index k2 , the usage index k3 and the usage index k4 , respectively.
  • the idle level of each usage indicator is determined according to Table 2, and then the cloud is determined according to the state evaluation strategy a and the state evaluation strategy b respectively.
  • the idle value of the environment, and finally the state of the cloud environment is determined based on the logical relationship of the two state evaluation strategies.
  • step S612 the cloud environment is recovered.
  • the cloud environment management platform releases all the resources of the cloud environment E1 and recycles the cloud environment E1 as a whole.
  • step S614 the degree of fitting difference of each prediction function in the prediction model is determined according to the historical idle value of the cloud environment.
  • the cloud environment management platform Since the cloud environment E2 cannot be reclaimed temporarily, the cloud environment management platform records the idle values of the cloud environment E2 under the state evaluation strategies a and b respectively, and then predicts the reclamation time of the cloud environment E2 according to all the recorded historical idle values. .
  • the prediction model M is:
  • Cloud environment history of each management platform may cloud E2 idle value y t1, y t2, y t3 ?? y tm are inputted predictive model M for each prediction function among each respectively a prediction function value corresponding to idle, then determined for each prediction function corresponding to the difference of a c fitting according to the following equation:
  • y t1 , y t2 , y t3 ... y tm are the idle values of cloud environment E2 at time t1, time t2, time t3... tn time m times respectively, and time tm is the closest to the current time.
  • a c is the fitting difference of the prediction function, y i is the idle value at the ith moment; y t is the predicted idle value obtained by inputting the historical idle value into the prediction function, and u is the accuracy factor.
  • step S616 a prediction function with the smallest fitting difference is selected as the final prediction function.
  • one of the prediction functions may be selected as the final prediction function according to the fitting difference degree.
  • the cloud environment management platform may select the fitting difference The prediction function with the smallest degree is used as the final prediction function.
  • the cloud environment management platform may select a prediction function whose fitting difference degree is smaller than a preset threshold as the final prediction function. If there are two or more, the cloud environment management platform can also randomly select one of these prediction functions as the final prediction function.
  • step S618 the recovery time of the cloud environment is predicted according to the final prediction function and the idle value of the cloud environment history.
  • the cloud environment management platform can determine the recovery of the cloud environment E2 based on the final prediction function, the recorded historical idle value of the cloud environment E2 under the state evaluation strategies a and b, and the logical relationship between the state evaluation strategies a and b. time.
  • the cloud environment platform management solution provided in this embodiment determines the state of the cloud environment by collecting information representing the resource usage of the cloud environment and analyzing the information, thereby realizing the recovery of idle cloud environments and reducing enterprise costs.
  • the waste of cloud environment resources can improve the utilization rate and turnover rate of resources, and help enterprises to improve the efficiency of R&D.
  • the state evaluation strategy and the idle standard value of the state evaluation strategy in this embodiment support dynamic adjustment to meet different cloud environment recycling scenarios of enterprises, and can accurately customize the dynamic recycling of various cloud environments, so as to more efficiently identify and intelligently Recycle cloud environment resources to maximize resource utilization.
  • the cloud environment management platform can also predict the recycling time of the cloud environment that cannot be recycled at present, so as to provide more reference information for the allocation and recycling of the cloud environment, and further optimize the management of the cloud environment.
  • embodiments of the present disclosure provide a storage medium comprising an easy-to-use storage medium implemented in any method or technology for storing information, such as computer-readable instructions, data structures, computer program modules, or other data. Volatile or non-volatile, removable or non-removable media.
  • Storage media include but are not limited to RAM (Random Access Memory), ROM (Read-Only Memory, read only memory), EEPROM (Electrically Erasable Programmable read only memory, electrified Erasable Programmable Read Only Memory), flash memory or other memory technology, CD-ROM (Compact Disc Read-Only Memory), Digital Versatile Disc (DVD) or other optical disk storage, magnetic cartridges, magnetic tape, magnetic disk storage or other magnetic storage devices, or available with Any other medium that stores the desired information and can be accessed by a computer.
  • RAM Random Access Memory
  • ROM Read-Only Memory
  • EEPROM Electrically Erasable Programmable read only memory
  • flash memory or other memory technology
  • CD-ROM Compact Disc Read-Only Memory
  • DVD Digital Versatile Disc
  • the storage medium may store one or more computer programs that can be read, compiled and executed by one or more processors.
  • the storage medium may store a cloud environment management program, a cloud environment management program The process of implementing any one of the cloud environment management methods described in the foregoing embodiments can be executed by one or more processors.
  • an embodiment of the present disclosure also provides a computer program product, including a computer-readable device, on which the computer program shown above is stored.
  • the computer-readable device may include the computer-readable storage medium as described above.
  • the computer program product includes a cloud environment management platform, as shown in FIG. 8 : the cloud environment management platform 8 includes a processor 81, a memory 82, and a communication bus 83 for connecting the processor 81 and the memory 82, wherein the memory 82 can be The aforementioned storage medium stores the cloud environment management program.
  • the processor 81 can read the cloud environment management program, compile and execute the process of implementing the cloud environment management method introduced in the foregoing embodiment: the processor 81 first obtains the usage data of the cloud environment, and the usage data can represent the current usage of the cloud environment . Then, the usage index of the cloud environment is determined according to the usage data of the cloud environment, and then the current state of the cloud environment is determined according to the usage index of the cloud environment; if the cloud environment is currently in an idle state, the processor 81 recycles the cloud environment.
  • the usage indicators of the cloud environment include at least one of CPU utilization, memory utilization, total network traffic, login frequency, and recent access duration, and the login frequency is preset by the cloud environment recently.
  • the number or frequency of logins within the duration, and the most recent access duration is the time difference between the time when the cloud environment was last accessed and the current time.
  • the processor 81 If the usage indicators of the cloud environment include at least one of CPU utilization, memory utilization, and the sum of network traffic, the processor 81 also needs to obtain the corresponding communication address, username and Login password, and then deploy the information collection agent in the cloud environment according to the communication address, user name and login password. When acquiring the usage data of the cloud environment, the processor 81 may receive the usage data of the cloud environment sent by the information collection agent.
  • the processor 81 when the processor 81 determines the current state of the cloud environment according to the use index of the cloud environment, it can determine the current idle value of the cloud environment according to the use index of the cloud environment and the state evaluation strategy. The higher the probability that the cloud environment is reclaimed, the status evaluation strategy includes the weight of each usage indicator in determining the idle value of the cloud environment. If the current idle value of the cloud environment reaches the idle standard value corresponding to the state evaluation policy, the processor 81 determines that the cloud environment is currently in an idle state.
  • the processor 81 before determining the current state of the cloud environment according to the usage index of the cloud environment, the processor 81 further receives a policy setting instruction through an input device, and determines a state evaluation policy according to the policy setting instruction.
  • the policy setting instruction includes the level requirements of each usage indicator of the cloud environment that can be recycled, and the weight occupied by each usage indicator in determining the idle value of the cloud environment.
  • the processor 81 if the cloud environment is currently in a non-idle state, the processor 81 further predicts the recovery time of the cloud environment according to the prediction model combined with the historical idle value of the cloud environment.
  • the prediction model includes at least two prediction functions.
  • the processor 81 predicts the recovery time of the cloud environment according to the prediction model and the historical idle value of the cloud environment, it can first determine each The fitting difference degree corresponding to the prediction function is predicted, and then a prediction function is selected as the final prediction function for the cloud environment according to the fitting difference degree; then the recovery time of the cloud environment is predicted according to the final prediction function and the historical idle value of the cloud environment.
  • the state evaluation strategy is based on the usage index of the cloud environment and the state.
  • the evaluation policy determines the current idle value of the cloud environment under the state evaluation policy, and determines whether the idle value of the cloud environment under the state evaluation policy reaches the idle standard value corresponding to the state evaluation policy. If the idle value of the cloud environment under at least one state evaluation policy reaches the corresponding idle standard value, the processor 81 determines that the cloud environment is currently in an idle state.
  • the cloud environment management platform and storage medium provided by the embodiments of the present disclosure obtain the usage data of the cloud environment, then determine the usage index of the cloud environment according to the usage data of the cloud environment, and determine the current state of the cloud environment according to the usage index of the cloud environment. If it is determined that the cloud environment is currently in an idle state according to the usage indicator of the cloud environment, the cloud environment can be recycled. In this way, the recycled cloud environment can be applied by other users for other R&D activities.
  • the idle cloud environment is recovered as a whole, the reuse of the idle cloud environment is realized, and the infinite recycling of cloud environment resources can be realized under the condition of limited actual resources, which is beneficial to Improve the utilization of cloud environment resources and provide guarantee for enterprise R&D innovation.
  • the functional modules/units in the system, and the device can be implemented as software (which can be implemented by computer program codes executable by a computing device). ), firmware, hardware, and their appropriate combination.
  • the division between functional modules/units mentioned in the above description does not necessarily correspond to the division of physical components; for example, one physical component may have multiple functions, or one function or step may be composed of several physical components Components execute cooperatively.
  • Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor or microprocessor, or as hardware, or as an integrated circuit, such as an application specific integrated circuit .
  • communication media typically embodies computer readable instructions, data structures, computer program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and can include any information delivery, as is well known to those of ordinary skill in the art medium. Therefore, the present disclosure is not limited to any particular combination of hardware and software.

Landscapes

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

Abstract

一种云环境管理方法、云环境管理平台及存储介质。所述方法包括：云环境管理平台获取云环境的使用数据（S102）；云环境管理平台根据云环境的使用数据确定云环境的使用指标（S104）；云环境管理平台根据云环境的使用指标确定云环境当前的状态（S106）；若云环境当前处于闲置状态，则云环境管理平台对云环境进行回收（S108）。

Description

云环境管理方法、云环境管理平台及存储介质 技术领域
本公开实施例涉及但不限于计算机领域。
背景技术
近些年来，云计算技术快速发展，在同一云平台之上提供标准的云环境，并在此基础上实现自动化的应用部署，让用户能以申请的方式按需获取云环境，这一应用模式迅速被人们认可和采用。目前已经可以为用户提供基于裸金属层、IaaS(Infrastructure as a Service，基础设施即服务)层、PaaS(Platform-as-a-Service，平台即服务)层、SaaS(Software-as-a-Service，软件即服务)层的应用环境自动化部署。
发明内容
在一方面，本公开实施例提供一种云环境管理方法，包括：获取云环境的使用数据，使用数据能够表征云环境当前的使用情况；根据云环境的使用数据确定云环境的使用指标；根据云环境的使用指标确定云环境当前的状态；若云环境当前处于闲置状态，则对云环境进行回收。
另一方面，本公开实施例还提供一种云环境管理平台，云环境管理平台包括处理器、存储器及通信总线；通信总线用于实现处理器和存储器之间的连接通信；处理器用于执行存储器中存储的一个或者多个程序，以实现上述云环境管理方法的步骤。
另一方面，本公开实施例还提供一种存储介质，该存储介质存储有云环境回收程序，云环境回收程序可被一个或者多个处理器执行，以实现上述云环境管理方法的步骤。
附图说明
图1为本公开实施例提供的云环境管理方法的一种流程图；
图2为本公开实施例提供的云环境管理平台对云环境当前的状 态进行评估的一种流程图；
图3为本公开实施例提供的云环境管理平台根据管理人员输入的策略设置指令设置状态评估策略的一种流程图；
图4为根据本公开实施例的云环境管理平台的一种交互界面；
图5为根据本公开实施例的云环境管理平台的另一种交互界面；
图6为本公开实施例提供的云环境管理方法的一种流程图；
图7为根据本公开实施例的基于两种状态评估策略进行状态评估的原理示意图；
图8为本公开实施例提供的云环境管理平台的一种硬件结构示意图。
具体实施方式
为了使本公开的目的、技术方案及优点更加清楚明白，下面通过具体实施方式结合附图对本公开实施例作进一步详细说明。应当理解，此处所描述的具体实施例仅仅用以解释本公开，并不用于限定本公开。
在很多企业，资源集中管理之后，研发用户只需要通过统一的云管理平台获取云环境就能进行各种研发活动，这大幅提升了研发效率。但云平台下资源有限，不可能无限制地进行云环境分配，在这种情况下如何保障研发活动的进行就成了亟待解决的问题。
在一方面，为了解决相关技术中因为云平台下资源有限，导致云环境分配完以后，后续研发活动难以获取到需要的云环境，影响企业研发创新效果的问题，本公开实施例提供一种云环境管理方法。图1为本公开实施例提供的云环境管理方法的一种流程图。参见图1，所述云环境管理方法包括步骤S102至S108。
在步骤S102，云环境管理平台获取云环境的使用数据。
本实施例中，云环境管理平台获取的使用数据是能够表征对应云环境当前被使用情况的数据。可以理解的是，云环境CPU(中央处理单元)资源的占用情况、内存资源占用情况以及存储资源的占用 情况可以在一定程度上表征该云环境当前的使用情况，例如，在t1时刻，云环境平台获取的数据中云环境内存资源的占用率为32％，而在当前时刻该云环境内存资源的占用率为58％，则说明从t1时刻到当前时刻，该云环境是有被使用的。另外一方面，云环境的网络流量也可以表征该云环境被使用的情况。除此以外，云环境被用户登录的情况也能在一定程度上体现出该云环境被用户使用的情况。所以，在本实施例中，云环境管理平台获取的能够表征云环境当前使用情况的使用数据包括上述几种数据中的至少一种。
如果云环境管理平台需要获取的使用数据包括能够体现云环境CPU资源的占用情况、内存资源占用情况以及存储资源的占用情况几种中至少一种的数据，或者，包括能够体现云环境网络流量的数据，则需要在对应的云环境中进行数据采集。在本实施例的一些示例当中，云环境管理平台可以部署信息采集代理在对应的云环境中，让信息采集代理在云环境中帮助采集能够表征云环境CPU资源占用、内存资源占用、存储资源占用几种中至少一种的使用数据，并将采集到的数据反馈给云环境管理平台。在本实施例的一种示例当中，云环境管理平台可以获取到对应云环境的通信地址，例如，云环境的IP(互联网协议)地址，以及登录对应云环境的用户名和登录密码，然后按照云环境的通信地址、用户名和登录密码登录到对应的云环境上，在对应的云环境中部署信息采集代理。
如果云环境管理平台需要获取的使用数据包括能够体现云环境被登录情况的数据，则可以不用到云环境中进行数据采集。在一些示例当中，云环境管理平台原本就会记录用户登录云环境的信息，所以，云环境可以在没有信息采集代理的情况下直接获取到这一类的使用数据。
在步骤S104，云环境管理平台根据云环境的使用数据确定云环境的使用指标。
云环境管理平台在获取到云环境的使用数据之后，可以对这些使用数据进行处理，然后抽象得到云环境的使用指标，在本实施例的一些示例当中云环境管理平台获取到的使用数据中包括一些无效的 数据或者是一些明显错误的数据，所以，云环境管理平台会对获取到的使用数据进行清洗筛除处理，筛除掉无效数据或明显错误的数据，然后对剩余的数据进行分析处理，抽象出云环境的使用指标。
在本实施例的一些示例当中，云环境的使用指标包括CPU利用率、内存利用率、网络流量总和、登录频次、最近访问时长几种中的至少一种。其中，“登录频次”是指云环境在最近预设时长内被登录的次数或频率，“最近访问时长”是指该云环境最近一次被访问的时刻距离当前时刻的时间差值。
毫无疑义的是，如果要求使用数据经过处理之后得到云环境的CPU利用率，则云环境预先获取到的使用数据中应该包括表征云环境CPU资源占用情况的数据；如果要求使用数据经过处理之后得到云环境的内存利用率，则云环境预先获取到的使用数据中应该包括表征云环境内存资源占用情况的数据；如果要求使用数据经过处理之后得到云环境的网络流量总和，则云环境预先获取到的使用数据中应该包括能够体现云环境网络流量的数据。可见，如果云环境的使用指标包括CPU利用率、内存利用率、网络流量总和中的至少一种，则云环境管理平台获取云环境的使用数据之前，需要先在云环境中部署信息采集代理。如果要求使用数据经过处理之后得到云环境的登录频次和/或最近访问时长，则云环境应该预先获取到的使用数据包括表征云环境登录情况的数据。
在步骤S106，云环境管理平台根据云环境的使用指标确定云环境当前的状态。
云环境管理平台确定出云环境的使用指标之后，可以根据使用指标评估云环境当前的状态，在本实施例中，云环境的使用状态包括闲置状态与非闲置状态。所谓非闲置状态，实际上也就是使用状态。值得注意的是，本实施例中，如果一个云环境被判断处于闲置状态，并不意味着该云环境处于绝对的闲置状态，这只能说明在云环境管理平台所采用的评估标准下该云环境处于闲置状态。
如果在评估云环境的状态时仅采用一种使用指标进行评估，则云环境平台可以针对该使用指标设置一个阈值，也可称为闲置标准值， 然后，将云环境该使用指标的值作为该云环境当前的闲置值达到闲置标准值的要求，就可以判定该云环境处于可以回收的闲置状态。换言之，只要云环境对应使用指标的值达到指标阈值的要求，则可以判定该云环境处于闲置状态。
不过，在更多的示例当中，评估云环境状态时会结合两种甚至两种以上的使用指标，但这些使用指标可能具有不同的度量衡，所以，云环境管理平台在评估一个云环境的状态时，需要将这些不同度量衡的使用指标转换到统一的标准下参与评估，这实际上是对云环境不同使用指标进行归一化处理过程。在本实施例的一些示例当中，云环境管理平台可以将所有的使用指标均转换为百分比：例如，为不同的使用指标设置一个最大值，然后确定云环境使用指标值与该最大值的比值，通过这种方式将所有使用指标的值均转换为0-1之间的值。当然，采用类似的方式也可以将所有使用指标的值均转换为1-100或1-1000之间，按照百分制或千分制来衡量各使用指标的大小。
应当明白的是，在不同的情境下，或者是对于不同的企业而言，对于各个使用指标的看重程度也不同，所以，在一些示例当中，在评估云环境状态时，不同的使用指标的权重不完全相同。例如，在一种示例当中，管理人员设置要求，只要CPU利用率、内存利用率以及登录频次符合要求，则可以判定云环境处于闲置状态，而对于最近访问时长以及网络流量总和则不关心。或者也可以认为，最近访问时长以及网络流量总和在评估云环境状态时所占的权重为0。
在本实施例的一些示例当中，云环境管理平台可以参照图2示出的流程来对云环境当前的状态进行评估，该流程可以包括步骤S202至S208。
在步骤S202，根据云环境的使用指标以及状态评估策略确定云环境当前的闲置值。
一个云环境的闲置值越高，则该云环境处于闲置状态的可能性越大，其被回收的可能性就越大。状态评估策略中包括各使用指标在确定云环境闲置值时所占的权重。在本实施例的一个示例当中，云环境管理平台将云环境各使用指标转换为同一的度量衡之后，可以根据 各指标所占的权重确定云环境的闲置值：
Figure PCTCN2021087767-appb-000001
其中，n为评估云环境当前状态所用的使用指标的总数，W i为第i个使用指标被转换到统一度量衡后的值，P i为第i个使用指标的权重，S是云环境的闲置值。
可以理解的是，在计算云环境闲置值的时候也可以采用其他方式，例如参照图6描述的实施例中提供的方式。
在步骤S204，判断云环境当前的闲置值是否达到状态评估策略对应的闲置标准值。
若判断结果为是，则进入步骤S206，否则进入步骤S208。
每一个状态评估策略都有一个对应的闲置标准值，如果一个云环境在该状态评估策略下的闲置值达到闲置标准值，则说明该云环境处于闲置状态，否则说明该云环境当前处于非闲置状态。
在步骤S206，确定云环境当前处于闲置状态。
在步骤S208，确定云环境当前处于非闲置状态。
在本实施例的一些示例当中，云环境可以仅采用一个状态评估策略来评估云环境当前的状态，但在本实施例的另外一些示例当中，云环境可以同时采用至少两个状态评估策略来评估云环境当前的状态。各状态评估策略中的使用指标不完全相同，但可以由部分重叠。当然，即便某两个状态评估策略中包括相同的部分使用指标，不过这些使用指标在两个状态评估策略中的权重也可以不同。例如，在一种示例当中，云环境管理平台采用了状态评估策略a和状态评估策略b来评估云环境的状态，其中状态评估策略a关注的使用指标有CPU利用率、内存利用率以及最近访问时长，而状态评估策略b关注的使用指标有CPU利用率、网络流量总和，这两个状态评估策略中均有CPU利用率，不过CPU利用率在两个状态评估策略中的权重不相同。
在一些情况下，各状态评估策略间是“或”逻辑关系，即只要云环境在其中任意一个状态评估策略下的闲置值达到该状态评估策略对应的闲置标准值，则说明该云环境处于闲置状态，当前可被回收。 在另一些情况下，各状态评估策略间是“和”逻辑关系，也即，云环境仅在其中某一个状态评估策略下的闲置值达到对应的闲置标准值，并不能说明该云环境处于闲置状态，而是要云环境在所有状态评估策略下的闲置值均达到对应的闲置标准值才能判定该云环境处于闲置状态。
在本实施例中，状态评估策略可以是由管理人员设置的，例如，在本实施例的一些示例当中，云环境管理平台可以包括人机交互设备(显示设备与输入设备)，管理人员可以通过输入设备向云环境管理平台输入策略设置指令，云环境管理平台接收到管理人员输入的策略设置指令后，根据指令确定出状态评估策略。图3为本公开实施例提供的云环境管理平台根据管理人员输入的策略设置指令设置状态评估策略的一种流程图，如图3所示，该流程可以包括步骤S302和S304。
在步骤S302，云环境管理平台通过输入设备接收策略设置指令。
在本实施例的一些示例当中，策略设置指令中包括能被回收的云环境的各使用指标的等级要求，以及各使用指标在确定云环境闲置值时所占的权重。应当理解的是，在一些情况下中，云环境管理平台具有友好的人机交互界面，可以在管理人员下发策略设置指令的时候对用户进行引导，例如，在图4示出的云环境管理平台的一种交互界面中，云环境管理平台通过显示设备给出可供选择的使用指标，管理人员在设置一个状态评估策略时，可以从这些使用指标中选择出该状态评估策略所关注的使用指标，然后在图5所示的交互界面中指出选中使用指标的权重以及闲置等级要求。
在步骤S304，云环境管理平台根据策略设置指令确定状态评估策略。
云环境管理平台接收到用户的策略设置指令后，可以对策略设置指令进行解析，然后生成状态评估策略。
另外，如果管理人员设置了不只一个状态评估策略，则其还应当向云环境管理平台指定各状态评估策略在用于评估云环境状态时彼此间的逻辑关系。
返回参照图1，在步骤S108，若云环境当前处于闲置状态，则 云环境管理平台对云环境进行回收。
在确定一个云环境处于闲置状态后，云环境管理平台可以对该云环境进行回收，应该理解的是，对该云环境进行回收实际上就是对该云环境的所有资源一同进行回收：云环境管理平台释放该云环境中的所有资源，然后将回收得到的资源放入云平台对应的资源池中，以供后续接收到云环境申请的时候进行分配。
在本实施例的一些示例当中，如果云环境管理平台确定某一云环境当前处于非闲置状态，则可以对该云环境进入闲置状态的时间进行预测，即预测该云环境进入可回收状态的回收时间。
应当理解的是，任意一个云环境从被分配使用开始，到进入使用高峰期以及到最终结束使用，这个使用过程中用户对该云环境的使用情况通常是符合一定规律的。所以，基于一个云环境历史的闲置值可以预测该云环境进入闲置状态的时间。例如，在一些示例当中，当云环境管理平台确定出云环境的闲置值小于对应状态评估策略的标准值时，可以将该云环境当前的闲置值进行记录存储。这样，当云环境管理平台需要预测某一云环境的回收时间时，可以根据记录的各历史闲置值进行预测。在本实施例的一些示例当中，预测云环境回收时间的时候，可以结合预测模型进行。预测模型可以由一个或多个预测函数组成，在本实施例的一些示例当中，一个预测模型中仅有一个预测函数，则云环境管理平台将云环境的历史闲置值输入该预测函数中，就可以得到该云环境的回收时间。在本实施例的另外一些示例当中，一个预测模式中同时包括两个或两个以上的预测函数，这些预测函数可以单独使用，所以，云环境管理平台在对某一云环境的回收时间进行预测前，需要先从预测模型的各预测函数中选择一个作为最终预测函数，然后才能利用最终预测函数来确定云环境的回收时间。
根据本公开实施例提供的云环境管理方法，通过获取云环境的使用数据，然后根据云环境的使用数据确定云环境的使用指标，并根据云环境的使用指标确定云环境当前的状态。如果根据云环境的使用指标确定云环境当前处于闲置状态，则可以对该云环境进行回收。这样，回收的云环境又可以被其他用户申请用于其他研发活动。通过本 公开实施例提供的云环境管理方法，将闲置的云环境进行整体回收，实现了闲置云环境的再利用，能够在实际资源有限的情况下，实现云环境资源的无限循环利用，有利于提升云环境资源的利用率，为企业研发创新提供保障。
为了使本领域技术人员对本公开实施例所提供的方案的优点与细节更加了解，下面将在前述实施例的基础上结合示例继续对云环境管理方案进行说明。图6为本公开实施例提供的云环境管理方法的一种流程图。如图6所述，所述云环境管理方法可以包括步骤S602至S618。
在步骤S602，采集云环境的使用数据。
在本实施例中，云环境管理平台采集的使用数据至少要保证后续能够确定出云环境的以下几种使用指标：CPU利用率、内存利用率、网络流量总和、登录频次、最近访问时长。当然，在本实施例的其他一些示例当中，云环境管理平台采集的使用数据中还包括能够体现云环境存储资源利用率的数据，或者还包括能够体现云环境IO(输入输出)接口使用情况的数据。
在步骤S604，对云环境的使用数据进行处理得到云环境的使用指标。
云环境管理平台对采集到的使用数据进行清洗处理，筛除掉无效数据或明显错误的数据，然后对剩余的使用数据进行分析处理，抽象出云环境的使用指标。
假定存在E1与E2两个云环境，这两个云环境的使用指标分别如表1所示：
表1
Figure PCTCN2021087767-appb-000002
在步骤S608，根据云环境的使用指标确定云环境在各个状态评 估策略下的闲置值。
在本实施例中，假定管理人员设置了两个逻辑关系为“或”的状态评估策略来对云环境的闲置状态进行评估，分别为状态评估策略a与状态评估策略b。这两个状态评估策略对应的闲置标准值均为55。在本实施例的一些示例当中，云环境管理平台按照各使用指标值的大小对各使用指标进行了闲置等级的划分，例如请参见表2：
表2
Figure PCTCN2021087767-appb-000003
根据表2可知，对于CPU利用率这一使用指标，低于20％(且大于等于10％)就算该使用指标的闲置等级为低，低于10％(且大于等于5％)该使用指标的闲置等级就为中，而低于5％后，则认为该使用指标的闲置等级为高。对于其他使用指标闲置等级的划分也是类似，例如，对于登录频次，如果在最近预设时长内的登录次数小于10次，则认为该使用指标的闲置等级达到高，如果登录次数介于10-20之间，则认为该使用指标的闲置等级达到了中等……这里不再赘述。
表3中分别示出了两个状态评估策略的内容：
表3
Figure PCTCN2021087767-appb-000004
从表3中可以看出，状态评估策略a仅关注CPU利用率、内存利用率以及最近访问时长三个使用指标，对这三个使用指标的闲置等级要求分别是高、中、低，权重分别为0.9、0.9以及0.6。而状态评估策略b关注内存利用率、网络流量总和、登录频次以及最近访问时长四个使用指标，对这四个使用指标的闲置等级要求分别是中、高、高、低，权重分别为0.8、0.4以及0.6、0.1。
在本实施例中，一个状态评估策略下的总分为100分，如果有n个使用指标，则每个使用指标对应分值均为100/n，例如，状态评估策略a中，每个使用指标的分值均为100/3，而状态评估策略b中，每个使用指标的分值均为25。在一个状态评估策略下，如果云环境的一个使用指标达到了该状态评估策略对该使用指标的闲置状态要求，则该云环境就可以得到该使用指标对应的分值，否则，云环境该使用指标对应的分数就为0。例如，在状态评估策略b下，如果一个云环境的最近访问时长达到了状态评估策略b所要求的低等，则该云环境就可以得到该云环境对应的25分。
本实施例中一个云环境在一个状态评估策略下的闲置值可以根据以下公式确定：
Figure PCTCN2021087767-appb-000005
其中，S为云环境在该状态评估策略下的闲置值，n为该状态评估策略下使用指标的总数，V i为第i个使用指标对应的得分，可以理解的是，V i的取值要么是100/n，要么是0；P i为第i个使用指标的权重。
下面针对表1中两个云环境E1和E2分别计算它们在两个状态评估策略下的闲置值：
首先，对于云环境E1，其在状态评估策略a下的闲置值
Figure PCTCN2021087767-appb-000006
其在状态评估策略b下的闲置值
Figure PCTCN2021087767-appb-000007
对于云环境E2，其在状态评估策略a下的闲置值
Figure PCTCN2021087767-appb-000008
其在状态评估策略b下的闲置值
Figure PCTCN2021087767-appb-000009
在步骤S610，判断云环境是否存在某一状态评估策略下的闲置值是否达到该状态评估策略对应的闲置标准值。
若判断结果为是，则进入S612，否则继续执行S614。
可见，云环境E1在状态评估策略a下的闲置值大于状态评估策略a的闲置标准值，因此云环境E1处于闲置状态。而云环境E2在状态评估策略a和b下的闲置值均小于对应的闲置标准值，因此，云环境E2处于非闲置状态下。
图7中示出了基于使用指标k1、使用指标k2、使用指标k3以及使用指标k4分别在状态评估策略a和状态评估策略b下评估云环境状态的一种原理示意图。基于云环境的使用数据确定出使用指标k1、使用指标k2、使用指标k3以及使用指标k4后，按照表2确定各使用指标的闲置等级，然后分别按照状态评估策略a和状态评估策略b确定云环境的闲置值，最后基于两个状态评估策略的逻辑关系确定云环境的状态。
在步骤S612，对云环境进行回收。
经过判断可以确定云环境E1需要被回收，因此云环境管理平台释放云环境E1的所有资源，对云环境E1进行整体回收。
在步骤S614，根据云环境的历史闲置值确定预测模型中各预测函数的拟合差异度。
由于云环境E2暂时不能回收，所以，云环境管理平台记录下该云环境E2分别在状态评估策略a和b下的闲置值，然后根据记下的所有历史闲置值预测该云环境E2的回收时间。
在本实施例中，预测模型M为：
Figure PCTCN2021087767-appb-000010
云环境管理平台可以将云环境E2的各历史闲置值y t1、y t2、y t3……y tm分别输入预测模型M的各预测函数当中，分别得到每一个预测函数对应的预测闲置值，然后按照如下公式确定每个预测函数对应的拟合差异度A c：
Figure PCTCN2021087767-appb-000011
其中，y t1、y t2、y t3……y tm分别是云环境E2在t1时刻、t2时刻、t3时刻……tn时刻m个时刻的闲置值，tm时刻距离当前最近。A c为预测函数的拟合差异度，y i为第i个时刻的闲置值；y t为将历史闲置值输入预测函数得到的预测闲置值，u为准确度因子。
在步骤S616，选择拟合差异度最小的一个预测函数作为最终预测函数。
确定出各预测函数的拟合差异度A c之后，可以根据拟合差异度从各预测函数中选择一个作为最终预测函数，在本实施例的一些示例当中，云环境管理平台可以选择拟合差异度最小的一个预测函数作为最终预测函数。不过，在本实施例的另外一些示例当中，云环境管理平台可以选择拟合差异度小于预设阈值的预测函数作为最终预测函数，如果预测模型M中拟合差异度小于预设阈值的预测函数有两个或两个以上，则云环境管理平台也可以随机从这些预测函数中选择一个作为最终预测函数。
在步骤S618，根据最终预测函数与云环境历史的闲置值预测云环境的回收时间。
选择出最终预测函数后，云环境管理平台可以基于最终预测函数、记录的云环境E2在状态评估策略a、b下的历史闲置值以及状态评估策略a、b的逻辑关系确定云环境E2的回收时间。
本实施例提供的云环境平台管理方案，通过采集表征云环境的资源使用情况的信息，并对这些信息进行分析，确定了云环境的状态，从而实现了对闲置云环境的回收，减少了企业云环境资源的浪费，提升资源的利用率、周转率，助力企业研发提效。此外，本实施例中的状态评估策略、状态评估策略的闲置标准值支持动态调整以满足企业不同的云环境回收场景，可以做到精准定制各类云环境的动态回收，从而更高效识别和智能回收云环境资源，最大力度提升资源利用率。
同时，云环境管理平台还可以对当前不能回收的云环境的回收时间进行预测，以便为云环境的分配、回收等提供更多可供参考的信息，进一步优化云环境的管理。
另一方面，本公开实施例提供了一种存储介质，该存储介质包括在用于存储信息(诸如计算机可读指令、数据结构、计算机程序模块或其他数据)的任何方法或技术中实施的易失性或非易失性、可移除或不可移除的介质。存储介质包括但不限于RAM(Random Access Memory，随机存取存储器),ROM(Read-Only Memory，只读存储器),EEPROM(Electrically Erasable Programmable read only memory，带电可擦可编程只读存储器)、闪存或其他存储器技术、CD-ROM(Compact Disc Read-Only Memory，光盘只读存储器)，数字多功能盘(DVD)或其他光盘存储、磁盒、磁带、磁盘存储或其他磁存储装置、或者可以用于存储期望的信息并且可以被计算机访问的任何其他的介质。
该存储介质中可以存储有一个或多个可供一个或多个处理器读取、编译并执行的计算机程序，在本实施例中，该存储介质可以存储有云环境管理程序，云环境管理程序可供一个或多个处理器执行实现前述实施例介绍的任意一种云环境管理方法的流程。
另一方面，本公开实施例还提供了一种计算机程序产品，包括计算机可读装置，该计算机可读装置上存储有如上所示的计算机程序。本实施例中该计算机可读装置可包括如上所示的计算机可读存储介质。例如，该计算机程序产品包括云环境管理平台，如图8所示：云 环境管理平台8包括处理器81、存储器82以及用于连接处理器81与存储器82的通信总线83，其中存储器82可以为前述存储有云环境管理程序的存储介质。处理器81可以读取云环境管理程序，进行编译并执行实现前述实施例中介绍的云环境管理方法的流程：处理器81先获取云环境的使用数据，使用数据能够表征云环境当前的使用情况。然后根据云环境的使用数据确定云环境的使用指标，再根据云环境的使用指标确定云环境当前的状态；若云环境当前处于闲置状态，则处理器81对云环境进行回收。
在本实施例的一些示例当中，云环境的使用指标包括CPU利用率、内存利用率、网络流量总和、登录频次、最近访问时长几种中的至少一种，登录频次为云环境在最近预设时长内被登录的次数或频率，最近访问时长为云环境最近一次被访问的时刻距离当前时刻的时间差值。
若云环境的使用指标包括CPU利用率、内存利用率、网络流量总和中的至少一种，则处理器81获取云环境的使用数据之前，还需要先获取云环境对应的通信地址、用户名与登录密码，然后根据通信地址、用户名与登录密码在云环境中部署信息采集代理。处理器81获取云环境的使用数据时，可以接收信息采集代理发送的云环境的使用数据。
在本实施例的一些示例当中，处理器81根据云环境的使用指标确定云环境当前的状态时，可以根据云环境的使用指标以及状态评估策略确定云环境当前的闲置值，闲置值越高，则表征云环境被回收的可能性越高，状态评估策略中包括各使用指标在确定云环境闲置值时所占的权重。如果云环境当前的闲置值达到状态评估策略对应的闲置标准值，则处理器81确定云环境当前处于闲置状态。
在本实施例的一些示例当中，处理器81根据云环境的使用指标确定云环境当前的状态之前，还会先通过输入设备接收策略设置指令，并根据策略设置指令确定状态评估策略。其中，策略设置指令中包括能被回收的云环境的各使用指标的等级要求，以及各使用指标在确定云环境闲置值时所占的权重。
在本实施例的一些示例当中，若云环境当前处于非闲置状态，则处理器81还会根据预测模型结合云环境历史的闲置值预测云环境的回收时间。
例如，在一些示例当中，预测模型中包括至少两个预测函数，处理器81根据预测模型结合云环境历史的闲置值预测云环境的回收时间时，可以先根据云环境历史的闲置值确定每个预测函数对应的拟合差异度，然后根据拟合差异度选择一个预测函数作为针对云环境的最终预测函数；再根据最终预测函数与云环境历史的闲置值预测云环境的回收时间。
在本实施例的一些示例当中，状态评估策略至少有两个，处理器81根据云环境的使用指标确定云环境当前的状态时，对于每一个状态评估策略，根据云环境的使用指标以及该状态评估策略确定云环境当前在该状态评估策略下的闲置值，并确定云环境在该状态评估策略下的闲置值是否达到该状态评估策略对应的闲置标准值。如果若云环境在至少一个状态评估策略下的闲置值达到对应的闲置标准值，则处理器81确定云环境当前处于闲置状态。
本公开实施例提供的云环境管理平台及存储介质，通过获取云环境的使用数据，然后根据云环境的使用数据确定云环境的使用指标，并根据云环境的使用指标确定云环境当前的状态。如果根据云环境的使用指标确定云环境当前处于闲置状态，则可以对该云环境进行回收。这样，回收的云环境又可以被其他用户申请用于其他研发活动。通过本公开实施例提供的云环境管理方法，将闲置的云环境进行整体回收，实现了闲置云环境的再利用，能够在实际资源有限的情况下，实现云环境资源的无限循环利用，有利于提升云环境资源的利用率，为企业研发创新提供保障。
可见，本领域的技术人员应该明白，上文中所公开方法中的全部或某些步骤、系统、装置中的功能模块/单元可以被实施为软件(可以用计算装置可执行的计算机程序代码来实现)、固件、硬件及其适当的组合。在硬件实施方式中，在以上描述中提及的功能模块/单元之间的划分不一定对应于物理组件的划分；例如，一个物理组件可以 具有多个功能，或者一个功能或步骤可以由若干物理组件合作执行。某些物理组件或所有物理组件可以被实施为由处理器，如中央处理器、数字信号处理器或微处理器执行的软件，或者被实施为硬件，或者被实施为集成电路，如专用集成电路。
此外，本领域普通技术人员公知的是，通信介质通常包含计算机可读指令、数据结构、计算机程序模块或者诸如载波或其他传输机制之类的调制数据信号中的其他数据，并且可包括任何信息递送介质。所以，本公开不限制于任何特定的硬件和软件结合。
以上内容是结合示例性的实施方式对本公开实施例所作的进一步详细说明，不能认定本公开的具体实施只局限于这些说明。对于本公开所属技术领域的普通技术人员来说，在不脱离本公开构思的前提下，还可以做出若干简单推演或替换，都应当视为属于本公开的保护范围。

Claims (10)

  1. 一种云环境管理方法，包括：
    获取云环境的使用数据，所述使用数据能够表征所述云环境当前的使用情况；
    根据所述云环境的使用数据确定所述云环境的使用指标；
    根据所述云环境的使用指标确定所述云环境当前的状态；
    若所述云环境当前处于闲置状态，则对所述云环境进行回收。
  2. 如权利要求1所述的云环境管理方法，其中，所述云环境的使用指标包括中央处理器(CPU)利用率、内存利用率、网络流量总和、登录频次、最近访问时长几种中的至少一种，所述登录频次为所述云环境在最近预设时长内被登录的次数或频率，所述最近访问时长为所述云环境最近一次被访问的时刻距离当前时刻的时间差值。
  3. 如权利要求2所述的云环境管理方法，其中，若所述云环境的使用指标包括CPU利用率、内存利用率、网络流量总和中的至少一种，则所述获取云环境的使用数据之前，还包括：
    获取所述云环境对应的通信地址、用户名与登录密码；
    根据所述通信地址、用户名与登录密码在所述云环境中部署信息采集代理；
    所述获取云环境的使用数据包括：
    接收所述信息采集代理发送的所述云环境的使用数据。
  4. 如权利要求1-3任一项所述的云环境管理方法，其中，所述根据所述云环境的使用指标确定所述云环境当前的状态包括：
    根据所述云环境的使用指标以及状态评估策略确定所述云环境当前的闲置值，所述闲置值越高，则表征所述云环境被回收的可能性越高，所述状态评估策略中包括各使用指标在确定所述云环境闲置值时所占的权重；
    若所述云环境当前的闲置值达到所述状态评估策略对应的闲置标准值，则确定所述云环境当前处于闲置状态。
  5. 如权利要求4所述的云环境管理方法，其中，所述根据所述云 环境的使用指标确定所述云环境当前的状态之前，还包括：
    通过输入设备接收策略设置指令，所述策略设置指令中包括能被回收的云环境的各所述使用指标的闲置等级要求，以及各所述使用指标在确定所述云环境闲置值时所占的权重；
    根据所述策略设置指令确定所述状态评估策略。
  6. 如权利要求4所述的云环境管理方法，其中，若所述云环境当前处于非闲置状态，则所述云环境管理方法还包括：
    根据预测模型结合所述云环境历史的闲置值预测所述云环境的回收时间。
  7. 如权利要求6所述的云环境管理方法，其中，所述预测模型中包括至少两个预测函数，所述根据预测模型结合所述云环境历史的闲置值预测所述云环境的回收时间包括：
    根据所述云环境历史的闲置值确定每个预测函数对应的拟合差异度；
    根据所述拟合差异度选择一个预测函数作为针对所述云环境的最终预测函数；
    根据所述最终预测函数与所述云环境历史的闲置值预测所述云环境的回收时间。
  8. 如权利要求4所述的云环境管理方法，其中，所述状态评估策略至少有两个，所述根据所述云环境的使用指标确定所述云环境当前的状态包括：
    对于每一个状态评估策略，根据所述云环境的使用指标以及所述状态评估策略确定所述云环境当前在该状态评估策略下的闲置值，并确定所述云环境在该状态评估策略下的闲置值是否达到该状态评估策略对应的闲置标准值；
    若所述云环境在至少一个状态评估策略下的闲置值达到对应的闲置标准值，则确定所述云环境当前处于闲置状态。
  9. 一种云环境管理平台，包括处理器、存储器及通信总线；
    所述通信总线配置为实现处理器和存储器之间的连接通信；
    所述处理器配置为执行存储器中存储的一个或者多个程序，以 实现如权利要求1至8中任一项所述的云环境管理方法的步骤。
  10. 一种存储介质，其上存储有云环境回收程序，所述云环境回收程序可被一个或者多个处理器执行，以实现如权利要求1至8中任一项所述的云环境管理方法的步骤。
PCT/CN2021/087767 2020-06-30 2021-04-16 云环境管理方法、云环境管理平台及存储介质 WO2022001295A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010617016.2A CN113867926A (zh) 2020-06-30 2020-06-30 一种云环境管理方法、云环境管理平台及存储介质
CN202010617016.2 2020-06-30

Publications (1)

Publication Number Publication Date
WO2022001295A1 true WO2022001295A1 (zh) 2022-01-06

Family

ID=78981611

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/087767 WO2022001295A1 (zh) 2020-06-30 2021-04-16 云环境管理方法、云环境管理平台及存储介质

Country Status (2)

Country Link
CN (1) CN113867926A (zh)
WO (1) WO2022001295A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115242602B (zh) * 2022-07-20 2023-11-14 北京天一恩华科技股份有限公司 网络设备生命周期管理方法、装置、设备及介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102831016A (zh) * 2012-08-01 2012-12-19 浪潮(北京)电子信息产业有限公司 一种云计算的物理机回收方法及其装置
CN106020969A (zh) * 2016-05-05 2016-10-12 云神科技投资股份有限公司 一种高性能云计算混合计算系统及其方法
CN106528266A (zh) * 2016-10-28 2017-03-22 柏域信息科技（上海）有限公司 一种云计算系统中的资源动态调整方法及装置
US20180267833A1 (en) * 2017-03-16 2018-09-20 Quanta Computer Inc. Management systems of cloud resources and management methods thereof
CN109491808A (zh) * 2018-11-09 2019-03-19 国网山东省电力公司信息通信公司 一种云服务管理系统

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107239337A (zh) * 2016-03-28 2017-10-10 北京智梵网络科技有限公司 虚拟化资源的分配和调度方法及系统
CN110505262B (zh) * 2018-05-18 2022-04-29 深信服科技股份有限公司 云环境下的动态微分段方法、系统、云服务器及存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102831016A (zh) * 2012-08-01 2012-12-19 浪潮(北京)电子信息产业有限公司 一种云计算的物理机回收方法及其装置
CN106020969A (zh) * 2016-05-05 2016-10-12 云神科技投资股份有限公司 一种高性能云计算混合计算系统及其方法
CN106528266A (zh) * 2016-10-28 2017-03-22 柏域信息科技（上海）有限公司 一种云计算系统中的资源动态调整方法及装置
US20180267833A1 (en) * 2017-03-16 2018-09-20 Quanta Computer Inc. Management systems of cloud resources and management methods thereof
CN109491808A (zh) * 2018-11-09 2019-03-19 国网山东省电力公司信息通信公司 一种云服务管理系统

Also Published As

Publication number Publication date
CN113867926A (zh) 2021-12-31

Similar Documents

Publication Publication Date Title
US10896055B2 (en) Capacity risk management for virtual machines
US20200174842A1 (en) Reward-based admission controller for resource requests in the cloud
US10838839B2 (en) Optimizing adaptive monitoring in resource constrained environments
KR101287066B1 (ko) 소셜미디어 클라우드 서비스 지원을 위한 지능형 가상화 시스템 및 그 방법
US8762525B2 (en) Managing risk in resource over-committed systems
CN113010576A (zh) 云计算系统容量评估的方法、装置、设备和存储介质
CN111381970B (zh) 集群任务的资源分配方法及装置、计算机装置及存储介质
CN108616553B (zh) 云计算资源池进行资源调度的方法及装置
CN105022668B (zh) 一种作业调度方法及系统
WO2022001295A1 (zh) 云环境管理方法、云环境管理平台及存储介质
KR20200122589A (ko) 클라우드 리소스 할당 장치 및 방법
CN114490078A (zh) 一种微服务的动态缩扩容方法、装置及设备
CN114844791A (zh) 基于大数据的云服务自动管理分配方法、系统及存储介质
CN113378498B (zh) 一种任务分配方法及装置
CN113886086A (zh) 云平台计算资源分配方法、系统、终端及存储介质
US20240020168A1 (en) Resource Scheduling Method and Apparatus, Computer Device, System, and Storage Medium
CN109992408B (zh) 一种资源分配方法、装置、电子设备和存储介质
CN117435337A (zh) 资源配置方法、装置、电子设备及存储介质
CN110750350A (zh) 一种大资源调度方法、系统、装置及可读存储介质
CN110851286B (zh) 线程管理方法、装置、电子设备及存储介质
CN110932926B (zh) 一种容器集群的监控方法、系统及装置
JP2023545316A (ja) システム性能に対する複数のアプリケーションの影響の決定
CN113127178B (zh) 资源抢占方法及装置、计算机可读存储介质、电子设备
CN112148469B (zh) 管理资源的方法、装置及计算机存储介质
CN118469229A (zh) 设备资源分配方法、装置、存储介质及电子设备

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: 21834489

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205 DATED 16/05/2023)

122 Ep: pct application non-entry in european phase

Ref document number: 21834489

Country of ref document: EP

Kind code of ref document: A1