Cloud platform resource scheduling method and device, equipment, medium and product

Abstract

The invention relates to the technical field of computers, and discloses a cloud platform resource scheduling method, device and equipment, a medium and a product, the method is applied to a data center comprising a plurality of power systems, each power system comprises a host machine, the data center comprises a plurality of host sets, and each host set comprises a cloud host; the method comprises the following steps: identifying an access host unit, and judging the type of the access host unit to create a target cloud host; under the condition that the access main unit is the power anti-affinity main unit, obtaining a target power system in the data center, and storing host machines in the target power system into an alternative list; the target power system is a power system capable of placing a target cloud host; and for each host machine in the alternative list, obtaining a first score and a second score to determine the power system bound with the target cloud host. The host resource utilization rate can be improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a cloud platform resource scheduling method, device, equipment, medium and product. Background Art

[0002] A cloud host group is a logical division of cloud hosts. The elastic cloud hosts in a cloud host group follow a unified policy.

[0003] In the related technology, cloud host groups have four strategies: forced affinity, forced anti-affinity, affinity, and anti-affinity. Affinity means that it can achieve nearby deployment, enhance network capabilities, achieve nearby routing in communication, and reduce network losses. Anti-affinity means that for high reliability considerations, instances are dispersed as much as possible. The binding of cloud hosts to hosts in the host group is random. If a host fails, the cloud hosts bound to the host will also fail, and the high availability requirements cannot be met.

[0004] In view of this, a cloud platform resource scheduling method is needed that can improve the utilization of host resources. Summary of the invention

[0005] In view of this, the present invention provides a cloud platform resource scheduling method, which can improve the utilization rate of host resources.

[0006] In a first aspect, the present invention provides a cloud platform resource scheduling method, which is applied to a data center including several power systems, the power system including host machines, the data center including several host groups, and the host group including cloud hosts; the method includes: identifying an access host group and determining the type of the access host group to create a target cloud host; in the case where the access host group is a power anti-affinity host group, obtaining a target power system in the data center, and storing the host machines in the target power system in an alternative list; the target power system is a power system in which the target cloud host can be placed; for each host machine in the alternative list, obtaining a first score and a second score to determine the power system bound to the target cloud host; the first score is a score for the target power system, and the second score is a score for the host machines in the target power system.

[0007] In this embodiment, the target cloud host is created by identifying the access host group and determining the type of the access host group; when the access host group is a power anti-affinity group, the power system that can place the target cloud host is obtained, and the host machines in the power system are stored in the candidate list, and then each host machine in the candidate list is scored, and the power system bound to the target cloud host is determined based on the scoring results. In the above scheme, the power system that cannot place the target cloud host is filtered out, and then the target cloud host is placed based on the scoring results, which can achieve efficient use of resources and avoid idle or overused resources.

[0008] In an optional embodiment, identifying an access host group and determining the type of the access host group includes: comparing power links of each host in the access host group, and determining the type of the access host group based on the comparison result; when the power links are independent of each other, the access host group is a power anti-affinity host group.

[0009] In this embodiment, by comparing the power links of each host in the access host group and judging the type of the access host group according to the comparison result, the power anti-affinity host group can be screened out, so as to facilitate the subsequent placement of the target cloud host.

[0010] In an optional embodiment, obtaining the target power system in the data center includes: determining whether there is a second cloud host in each power system in the data center, the second cloud host being a cloud host of the same type as the target cloud host; if there is no second cloud host in the power system, the power system is not the target power system; if there is no second cloud host in the power system, the power system is the target power system.

[0011] In this embodiment, by judging whether there is a cloud host of the same type as the target cloud host in the power system, it is determined whether the power system can place the target cloud host, so as to ensure the rationality of resource allocation and avoid waste of resources.

[0012] In an optional embodiment, obtaining a first score includes: obtaining energy consumption information of each power system in a data center to determine an average energy consumption, a maximum energy consumption and a first energy consumption value of the data center; the first energy consumption value is energy consumption information of a power system corresponding to a host machine in an alternative list; based on the average energy consumption, the maximum energy consumption and the first energy consumption value of the data center, obtaining a first score for the host machine in the alternative list.

[0013] In this embodiment, based on the energy consumption information of each power system in the data center, the average energy consumption, maximum energy consumption and first energy consumption value of the data center are determined to obtain the first score of the host machine in the alternative list. This can avoid the situation where the power system in the data center is under excessive energy consumption pressure, thereby improving the stability of the method.

[0014] In an optional implementation, obtaining the second score includes: obtaining an evaluation index of the host machine, wherein the evaluation index includes at least one of resource utilization and failure frequency; based on the evaluation index, collecting evaluation information of the host machine to determine the second score of the host machine.

[0015] In this embodiment, the second score of the host is determined according to the evaluation index of the host; the evaluation index includes at least one of resource utilization and failure frequency. The working status of the host can be obtained to improve the rationality of subsequent cloud host placement.

[0016] In an optional embodiment, obtaining a first score and a second score to determine a power system bound to a target cloud host includes: obtaining a first score, a second score, and a weight sequence of a host machine; matching the weight sequence with the first score and the second score; obtaining a third score based on the first score, the second score, and the weight sequence; and determining the power system bound to the target cloud host based on the third score.

[0017] In this embodiment, the first score and the second score are weighted and fused to obtain a third score. The host machine can be quantitatively evaluated to determine the power system bound to the target cloud host. The correct scheduling and reasonable placement of the cloud host can be achieved.

[0018] In a second aspect, the present invention provides a cloud platform resource scheduling device, which includes: an identification module for identifying an access host group and determining the type of the access host group to create a target cloud host; a screening module for obtaining a target power system in a data center when the access host group is a power anti-affinity host group, and storing the host machines in the target power system in an alternative list; the target power system is a power system in which the target cloud host can be placed; a scoring module for obtaining a first score and a second score for each host machine in the alternative list to determine the power system bound to the target cloud host; the first score is a score for the target power system, and the second score is a score for the host machine in the target power system.

[0019] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor are communicatively connected to each other, computer instructions are stored in the memory, and the processor executes the cloud platform resource scheduling method of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0020] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the cloud platform resource scheduling method of the first aspect or any corresponding embodiment thereof.

[0021] In a fifth aspect, the present invention provides a computer program product, comprising computer instructions, which are used to enable a computer to execute the cloud platform resource scheduling method of the above-mentioned first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0023] Figure 1 is a flow chart of a cloud platform resource scheduling method according to an embodiment of the present invention;

[0024] Figure 2 is a flow chart of another cloud platform resource scheduling method according to an embodiment of the present invention;

[0025] Figure 3 is a schematic diagram of a model architecture of a power anti-affinity scheduling architecture model according to an embodiment of the present invention;

[0026] Figure 4 is a module schematic diagram of a power system host management module according to an embodiment of the present invention;

[0027] Figure 5 is a cloud host placement result diagram according to the cloud platform resource scheduling method according to an embodiment of the present invention;

[0028] Figure 6 is a structural block diagram of a cloud platform resource scheduling device according to an embodiment of the present invention;

[0029] Figure 7 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0031] There are four cloud host group policies: forced affinity, forced anti-affinity, affinity, and anti-affinity. In the field of communications, affinity means that it can achieve nearby deployment, enhance network capabilities to achieve nearby routing in communications, and reduce network losses. Anti-affinity means that for high reliability considerations, instances are dispersed as much as possible.

[0032] In the related art, the binding of a cloud host to a host machine in a host group is random. If a host machine fails, the cloud host bound to the host machine will also fail, and the high availability requirement cannot be met.

[0033] The embodiment of the present invention provides a cloud platform resource scheduling method, which creates a target cloud host by identifying an access host group and determining the type of the access host group. Then, the power systems in the data center are screened, and then the target host machines in the power systems where the target cloud host can be placed are stored in a candidate list. And according to the scores of each host machine in the candidate list, the power system bound to the target cloud host is determined. It can ensure the reasonable allocation of resources, improve the stability of the system, and optimize the management of the data center.

[0034] According to an embodiment of the present invention, an embodiment of a cloud platform resource scheduling method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0035] In this embodiment, a cloud platform resource scheduling method is provided, which is applied to a data center including several power systems, the power systems including host machines, the data center including several host groups, and the host groups including cloud hosts. Figure 1 is a flow chart of a cloud platform resource scheduling method according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0036] Step S101, identifying an access host group and determining the type of the access host group to create a target cloud host.

[0037] The data center includes several power systems. To facilitate the management of the power systems, each power system can be numbered, and each power system has a unique number. This facilitates the unified management of the data center. In some optional implementations, the power systems can be numbered according to the operation status or geographical location of the power systems. Several host groups in the data center can also be numbered.

[0038] The number of each host group and the type information of the host group can be stored in the data center; after identifying the access host group, the number of the access host group can be obtained to determine the type of the access host group. The workload distribution of the access host group can also be viewed through monitoring software. In some optional implementations, a message queue service with multiple copies is input to the access host group. If they are all concentrated in one or a few hosts, the access host group can be determined to be a power-affinity host group; if they are evenly distributed on different hosts, the access host group is determined to be a power-anti-affinity host group.

[0039] If the access host group is a power anti-affinity host group, the target cloud host can be created. If the access host group is not a power anti-affinity host group, the anti-affinity rule is not met and the target cloud host cannot be created.

[0040] Step S102, when the access host group is a power anti-affinity host group, obtain the target power system in the data center, and store the host in the target power system into a candidate list; the target power system is a power system that can place the target cloud host.

[0041] When obtaining the target power system in the data center, it is first necessary to obtain the business applications of the cloud host and the functions of the cloud platform where the cloud host is located. Then, based on the business applications and cloud platform functions, it is determined one by one whether the target cloud host can be placed in the power system.

[0042] In a practical application, the anti-affinity rule is that cloud hosts of the same type cannot be placed in the same power system. Therefore, when there are cloud hosts of the same type as the target cloud host in the power system, the target cloud host cannot be placed in the power system.

[0043] The power system where the target cloud host can be placed is taken as the target power system, and then the host in the target power system is included in the candidate list. This means that the target cloud host can select a host from the candidate list for binding.

[0044] Step S103, for each host in the candidate list, obtain a first score and a second score to determine the power system bound to the target cloud host; the first score is a score for the target power system, and the second score is a score for the host in the target power system.

[0045] Among them, the first score is the score for the target power system. First of all, the first indicator system for the power system score must be clarified. The indicator system can be determined by the staff according to the actual situation. It can include power supply reliability, power quality, power supply capacity, etc. Then, according to the actual operation data of the power system, collect information on the corresponding indicators. It can be collected by installing power detection equipment on the power system to obtain real-time data. Then, according to the established scoring criteria, assign corresponding weights to each indicator and perform quantitative scoring to obtain the first score.

[0046] The second score is a score for the host machine in the target power system, and a second score index system for the host machine can be determined. The second score index system may include the hardware configuration device of the host machine, resource utilization, etc. Then, according to the second score index system, the corresponding indicator information is collected. In some optional embodiments, the indicator information can be collected using relevant monitoring tools. Then, for each indicator, a quantitative score is performed according to the pre-set standards and the actual data collected, and then the second score of the host machine is calculated by weighted summation based on the scores of each indicator and the assigned weights.

[0047] After obtaining the first score and the second score of each host in the candidate list, the first score and the second score can be weighted and fused, and based on the result of the weighted fusion, a host is selected from the candidate list for binding. In some optional implementations, if a power system with low energy consumption is to be selected to place the cloud host, the weight corresponding to the first score can be increased.

[0048] The cloud platform resource scheduling method provided in this embodiment creates a target cloud host by identifying the access host group and determining the type of the access host group; when the access host group is an electric power anti-affinity group, the power system that can place the target cloud host is obtained, and the host machines in the power system are stored in the candidate list, and then each host machine in the candidate list is scored, and the power system bound to the target cloud host is determined based on the scoring results. In the above scheme, the power system that cannot place the target cloud host is filtered out, and the target cloud host is placed based on the scoring results, which can achieve efficient use of resources and avoid idle or overused resources.

[0049] In this embodiment, a cloud platform resource scheduling method is provided, which is applied to a data center including several power systems, the power system including a host machine, the data center including several host groups, and the host group including a cloud host; Figure 2 is a flow chart of a cloud platform resource scheduling method according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:

[0050] Step S201, identifying an access host group and determining the type of the access host group to create a target cloud host.

[0051] For details, please see Figure 1 Step S101 of the illustrated embodiment will not be described in detail here.

[0052] In some optional implementations, the above step S201 includes:

[0053] Step S2011, comparing the power links of each host in the access host group, and judging the type of the access host group based on the comparison result.

[0054] The type of the access host group can be determined by comparing whether the power links of the hosts in the access host group are independent of each other. If the power links are not independent of each other, the access host group is not a power anti-affinity host group.

[0055] By checking the load distribution strategy of the access host group, etc., it can be determined whether to avoid binding the hosts in the same access host group to a common power link. If the power supply of the hosts in the access host group is relatively independent and has low correlation, the access host group is an anti-affinity host group; otherwise, it is an affinity host group.

[0056] Step S2012: When the power links are independent of each other, the access host group is a power anti-affinity host group.

[0057] When the power links are independent of each other, the access host group is a power anti-affinity host group.

[0058] By comparing the power links of each host in the access host group and judging the type of the access host group based on the comparison results, you can filter out the power anti-affinity host group, so as to facilitate the subsequent creation of the target cloud host.

[0059] Step S202, when the access host group is a power anti-affinity host group, obtain the target power system in the data center, and store the host in the target power system into a candidate list; the target power system is a power system that can place the target cloud host.

[0060] For details, please see Figure 1 Step S102 of the illustrated embodiment will not be described in detail here.

[0061] In some optional implementations, the above step S202 includes:

[0062] Step S2021, determining whether there is a second cloud host in each power system in the data center, where the second cloud host is a cloud host of the same type as the target cloud host.

[0063] The basic information of the target cloud host and the cloud hosts in each power system can be obtained first, and then the basic information of the target cloud host and the cloud hosts in each power system can be compared. According to the comparison results, it can be determined whether there is a second cloud host in the power system. The basic information can be determined by the staff according to the actual situation. It can include the network function of the cloud host, the type of operating system, etc.

[0064] Step S2022: When there is no second cloud host in the power system, the power system is not a target power system.

[0065] Step S2023: when there is no second cloud host in the power system, the power system is the target power system.

[0066] There are no two cloud hosts of the same type in the same power system, which can reduce the impact of power failures and optimize the use of power resources.

[0067] By judging whether there is a cloud host of the same type as the target cloud host in the power system, it can be determined whether the power system can place the target cloud host. This can ensure the rationality of resource allocation and avoid waste of resources.

[0068] Step S203, for each host in the candidate list, obtain a first score and a second score to determine the power system bound to the target cloud host; the first score is a score for the target power system, and the second score is a score for the host in the target power system.

[0069] Specifically, the above step S203 includes:

[0070] Step S2031, obtaining a first score, a second score, and a weight sequence of the host machine; the weight sequence matches the first score and the second score.

[0071] In some optional implementations, the above step S2031 includes:

[0072] Step a1, obtaining energy consumption information of each power system in the data center to determine the average energy consumption, maximum energy consumption and first energy consumption value of the data center; the first energy consumption value is the energy consumption information of the power system corresponding to the host machine in the candidate list.

[0073] Energy consumption information of the power system can be obtained by installing energy consumption monitoring equipment at key nodes of each power system in the data center, and then the average energy consumption of the power system in the data center and the maximum energy consumption of the power system in the data center can be obtained.

[0074] Step a2: obtaining a first score of the host machine in the candidate list based on the average energy consumption, the maximum energy consumption and the first energy consumption value of the data center.

[0075] According to the average energy consumption, the maximum energy consumption and the first energy consumption value of the data center, a first score of each host machine in the candidate list is obtained.

[0076] Among them, the first score can be obtained by the following formula:

[0077]

[0078] Among them, W Pi is the first score of the i-th host; P max is the maximum energy consumption; Pi is the energy consumption value of the i-th host, that is, the first energy consumption value; n is the number of power systems in the data center, is the average energy consumption of the power system in the data center.

[0079] According to the energy consumption information of each power system in the data center, the average energy consumption, maximum energy consumption and first energy consumption value of the data center are determined to obtain the first score of the host machine in the alternative list, thereby determining the placement of the target cloud host. This can avoid the situation where the power system in the data center is under excessive energy consumption pressure, thereby improving the stability of the method.

[0080] In some optional implementations, the above step S2031 further includes:

[0081] Step b1, obtaining evaluation indicators of the host machine, wherein the evaluation indicators include at least one of resource utilization and failure frequency.

[0082] The evaluation index of the host machine can be determined by the staff according to the actual situation. The resource utilization rate can include the host machine memory utilization rate, the host machine storage resource utilization rate and the network bandwidth utilization rate.

[0083] Step b2: Based on the evaluation index, collect evaluation information of the host machine to determine a second score of the host machine.

[0084] Based on the evaluation indicators, corresponding evaluation information is collected.

[0085] When the evaluation index includes resource utilization, the information of the host machine under the evaluation index can be collected through a performance monitoring tool.

[0086] When the evaluation index includes the failure frequency, the frequency of failure of the host machine can be determined by tracking and recording the operating status of the host machine.

[0087] Then, the second score of the host machine is calculated by weighting the weights assigned to the evaluation information and taking the weighted sum.

[0088] The second score can be obtained by the following formula:

[0089]

[0090] Among them, W hi is the second score of the ith host; x is the number of evaluation indicators; α ji is the weight of the jth evaluation index of the ith host; W ji is the score of the i-th host on the j-th evaluation indicator.

[0091] The second score of the host is determined according to the evaluation index of the host; the evaluation index includes at least one of resource utilization and failure frequency. The working status of the host can be obtained to improve the rationality of subsequent cloud host placement.

[0092] Step S2032: Obtain a third score based on the first score, the second score, and the weight sequence.

[0093] The weight sequence corresponds to the first score and the second score. It can be determined by the staff according to the actual situation. Based on the first score, the second score and the weight sequence, the third score can be obtained by weighted summation.

[0094] In a practical application, the third score can be obtained by the following formula:

[0095] W i =β1W pi +β2W hi

[0096] Among them, W i is the third score of the i-th host; β1 is the weight of the first score of the i-th host; W pi is the first score of the i-th host; β2 is the weight of the second score of the i-th host; W hi is the second score of the i-th host.

[0097] Step S2033: Determine the power system bound to the target cloud host based on the third score.

[0098] The power system bound to the target cloud host is determined based on the third scores of the hosts in the candidate list. In some optional implementations, the host with the highest third score in the candidate list is selected as the target host, and the target cloud host is bound to the power system where the target host is located.

[0099] The cloud platform resource scheduling method provided in this embodiment performs weighted fusion of the first score and the second score to obtain a third score. The host machine can be quantitatively evaluated to determine the power system bound to the target cloud host. The correct scheduling and reasonable placement of the cloud host can be achieved.

[0100] This embodiment also provides a power anti-affinity scheduling architecture model, such as Figure 3As shown in the figure, the power anti-affinity dispatching architecture includes three parts: CMDB (Configuration Management Database) power system management module, power system host management module, and Iaas (Infrastructure as a Service) cloud platform dispatching module. Among them, the CMDB power system management module is mainly responsible for monitoring and managing the power system of the data center, the power system host management module is mainly responsible for managing the physical machines and cloud hosts within the power system, and can timely perceive and adjust when the physical machines and cloud hosts change. The Iaas cloud platform dispatching module introduces power system filters and power anti-affinity scorers on the basis of the original dispatching occupancy unit, thereby realizing the placement and migration of cloud hosts in the access group and other functions.

[0101] The CMDB power system management module can manage the power system in the data center based on the intelligent operation and maintenance platform and CMDB resource management. The CMDB power system management module includes four functions: power system monitoring, power system division, power system number generation, and power system update. It can manage the life cycle of the power system, generate or update the power system model according to the operation of the power system in the data center, and pass the relevant information of the power system to the power system host management module or the Iaas cloud platform scheduling module for use by other modules. And by dividing the power system, the division and management of different power systems can be realized, creating basic conditions for the realization of power system facilities.

[0102] Among them, the power system monitoring function refers to monitoring and recording the working conditions of the power system in the current data center, so as to manage the power system and guide the dispatch of power anti-affinity. The power system division function refers to dividing the power system according to the power operation conditions of the power system in the data center. The power system can be divided according to the stability of power supply, power load conditions, etc. The power system number generation refers to generating corresponding identifiers based on the power system division results, so as to facilitate the subsequent management of the power system and extract power system information. In the same data center, the identifier of the power system can be unique. The power system update function refers to adding, deleting or changing the power system information according to the actual situation when the relevant information of the power system changes or there is a new power system.

[0103] The power system host management module is used to maintain key information associated with the power system and the host. It includes the management of the host in the power system, the management of the cloud host in the power system, and the management of the relationship between the power system and the host. According to the division of the power system in the CMDB power system management module, the binding relationship between the host and the power system can be established, so as to manage the host in the power system and the cloud host corresponding to the host. The schematic diagram of the power system host management module is shown in the figure. Figure 4 As shown, the power system host management module includes n power systems, namely Power-1 to Power-n. Each power system has m host groups, namely Host1-Hostm, and each host group has several host machines, which can be represented by vm1, vm2, vm3, etc.

[0104] The Iaas cloud platform scheduling module can reasonably allocate computing, storage, network and other resources according to user needs and the real-time status of platform resources to achieve efficient resource utilization; it can also arrange task scheduling and execution order, and process tasks submitted by different users according to task priority to ensure that key businesses are not affected. The Iaas cloud platform scheduling module includes a basic scheduling occupancy unit, a power anti-affinity filter and a power anti-affinity scorer. Among them, the basic scheduling occupancy unit is used to arrange task scheduling and execution order. The power anti-affinity filter is used to filter out power systems that meet the requirements, and the power anti-affinity scorer is used to select which power system of the host machine is more suitable for placing the cloud host.

[0105] The power anti-affinity filter can be used to screen the various power systems in the data center, select the power systems that do not have the same type of cloud hosts, and put the hosts in the corresponding power systems into the alternative list.

[0106] The power anti-affinity scorer can be used to obtain the host machines in the candidate list, score the host machines and the power systems corresponding to the host machines, and select a host machine from the candidate list to bind to the cloud host based on the scoring results.

[0107] In an actual application, the power input form of the resource pool is a dual-channel UPS (Uninterruptible Power Supply), and the CMDB power system management module uses the UPS output as the collection point to monitor the power system energy consumption. According to the actual situation of the power system, six available power systems are divided, and the power system number is generated. The power system number naming method can be: data center-availability zone-power system keyword. Among them, the data center can be the name or number of the data center; the availability zone refers to the different areas divided within the data center; the power system keyword can be the name of the power system.

[0108] In the power system host management module, the binding relationship between the host and the power system is established. At the same time, the association data between the host and the power anti-affinity host group is determined to identify the placement of the cloud host.

[0109] In some optional implementations, a label system can be designed to add a power system number label to each host machine, thereby marking which power system the host machine belongs to, so as to facilitate subsequent management and dispatcher scheduling. In a practical application, the core pseudo code for updating the power system and host machine numbers is as follows:

[0110] 1: H_IP = {,,...}

[0111] 2:ch CMDB Manager<-aRoadEqSimpleCode,bRoadEqSimpleCode from UPS

[0112] 2:ch power_number<-watch Power from CMDB Manager

[0113] 3:cronTab SynchronizeHostPowerNumberFromCMDB(H_IP)

[0114] 4:CheckPowerExists(power_number)

[0115] IF NOT_EXISTS->CreatePower(power_number)

[0116] ELSE:->GetPowerByPowerName(power_number)

[0117] 5:UpdateHostPowerNumber(H_IP)

[0118] The Iaas cloud platform management system of the newly added power anti-affinity filter and power anti-affinity scorer is deployed to the corresponding resource pool, thereby supporting the implementation of power anti-affinity scheduling based on the Iaas cloud platform.

[0119] In an actual application, 10 power anti-affinity host groups are connected, namely Group 1 to Group 10. Each host group has six cloud hosts. Through the cloud platform scheduling method provided by the embodiment of the present invention, the cloud host placement is as follows: Figure 5As shown in the figure, Power1 to Power6 represent six power systems. The six cloud hosts in each power anti-affinity host group are evenly placed in different power systems. Cloud hosts in the same host group do not share the power system, thus achieving high availability at the power system level.

[0120] In this embodiment, a cloud platform resource scheduling device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware for a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0121] This embodiment provides a cloud platform resource scheduling device, such as Figure 6 As shown, including:

[0122] The identification module 301 is used to identify the access host group and determine the type of the access host group to create a target cloud host.

[0123] The screening module 302 is used to obtain the target power system in the data center and store the host machines in the target power system into the candidate list when the access host group is a power anti-affinity host group; the target power system is a power system where the target cloud host can be placed.

[0124] Scoring module 303 is used to obtain a first score and a second score for each host in the candidate list to determine the power system bound to the target cloud host; the first score is the score for the target power system, and the second score is the score for the host in the target power system.

[0125] In some optional implementations, the identification module 301 includes:

[0126] The comparison unit is used to compare the power links of each host in the access host group, and determine the type of the access host group based on the comparison result.

[0127] The host group judgment unit is used to determine whether the access host group is a power anti-affinity host group when the power links are independent of each other.

[0128] In some optional implementations, the screening module 302 further includes:

[0129] A determination unit, used to determine whether there is a second cloud host in each power system in the data center, where the second cloud host is a cloud host of the same type as the target cloud host;

[0130] The first result unit is used to determine that, when the second cloud host does not exist in the power system, the power system is not a target power system.

[0131] The second result unit is used to determine that the power system is a target power system when the second cloud host does not exist in the power system.

[0132] In some optional implementations, the scoring module 303 includes:

[0133] An energy consumption information acquisition unit is used to acquire energy consumption information of each power system in the data center to determine an average energy consumption value, a maximum energy consumption value and a first energy consumption value of the data center; the first energy consumption value is energy consumption information of the power system corresponding to the host machine in the candidate list;

[0134] The first score acquisition unit is used to acquire a first score of a host machine in the candidate list based on an average energy consumption value, a maximum energy consumption value and a first energy consumption value of the data center.

[0135] In some optional implementations, the scoring module 303 includes:

[0136] An evaluation index acquisition unit, used to acquire an evaluation index of the host machine, wherein the evaluation index includes at least one of resource utilization and failure frequency;

[0137] The second score acquisition unit is used to collect evaluation information of the host machine based on the evaluation index to determine a second score of the host machine.

[0138] In some optional implementations, the scoring module 303 includes:

[0139] An information acquisition unit, used to acquire a first score, a second score, and a weight sequence of the host machine; the weight sequence matches the first score and the second score;

[0140] A third score acquisition unit, configured to acquire a third score based on the first score, the second score, and the weight sequence;

[0141] The placement unit is used to determine a power system bound to the target cloud host based on the third score.

[0142] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0143] The cloud platform resource scheduling device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0144] The embodiment of the present invention also provides a computer device having the above Figure 6 The cloud platform resource scheduling device shown.

[0145] See also Figure 7 , Figure 7 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Figure 7 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 7 A processor 10 is taken as an example.

[0146] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0147] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0148] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0149] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0150] The computer device also includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected via a bus or other means. Figure 7 The example of connecting through bus is taken in the following.

[0151] The input device 30 can receive input digital or character information, and generate key signal input related to the user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display and a plasma display. In some optional embodiments, the display device can be a touch screen.

[0152] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0153] A part of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the existence of the computer program instruction in a computer-readable medium includes, but is not limited to, a source file, an executable file, an installation package file, etc., and accordingly, the way in which the computer program instruction is executed by the computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium accessible to the computer.

[0154] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A cloud platform resource scheduling method, characterized in that: The method is applied to a data center including several power systems, wherein the power systems include host machines, and the data center includes several host groups, wherein the host groups include cloud hosts; the method includes: Identify an access host group and determine the type of the access host group to create a target cloud host; In the case where the access host group is a power anti-affinity host group, a target power system in the data center is obtained, and hosts in the target power system are stored in a candidate list; the target power system is a power system in which a target cloud host can be placed; For each host in the candidate list, a first score and a second score are obtained to determine the power system bound to the target cloud host; the first score is a score for the target power system, and the second score is a score for the host in the target power system.

2. The method according to claim 1, characterized in that The identifying the access host group and determining the type of the access host group includes: Comparing the power links of the respective hosts in the access host group, and determining the type of the access host group based on the comparison result; In the case that the power links are independent of each other, the access host group is a power anti-affinity host group.

3. The method according to claim 1, characterized in that The acquiring of the target power system in the data center comprises: Determine whether there is a second cloud host in each power system in the data center, where the second cloud host is a cloud host of the same type as the target cloud host; In the case that the second cloud host does not exist in the power system, the power system is not a target power system; When the second cloud host does not exist in the power system, the power system is a target power system.

4. The method according to claim 1, characterized in that: The obtaining of the first score comprises: Obtaining energy consumption information of each power system in the data center to determine an average energy consumption, a maximum energy consumption, and a first energy consumption value of the data center; the first energy consumption value is energy consumption information of the power system corresponding to the host machine in the candidate list; Based on the average energy consumption, the maximum energy consumption and the first energy consumption value of the data center, a first score of the host machine in the candidate list is obtained.

5. The method according to claim 1, characterized in that The acquisition of the second score includes: Acquire an evaluation index of the host machine, wherein the evaluation index includes at least one of resource utilization and failure frequency; Based on the evaluation index, evaluation information of the host machine is collected to determine a second score of the host machine.

6. The method according to claim 1, characterized in that The obtaining of the first score and the second score to determine the power system bound to the target cloud host includes: Obtaining a first score, a second score, and a weight sequence of the host machine; the weight sequence matches the first score and the second score; Based on the first score, the second score and the weight sequence, obtaining a third score; Based on the third score, a power system bound to the target cloud host is determined.

7. A cloud platform resource scheduling device, characterized in that: The device comprises: An identification module, used to identify an access host group and determine the type of the access host group to create a target cloud host; A screening module, used for obtaining a target power system in a data center and storing the host machines in the target power system into a candidate list when the access host group is a power anti-affinity host group; the target power system is a power system in which a target cloud host can be placed; A scoring module is used to obtain a first score and a second score for each host machine in the candidate list to determine the power system bound to the target cloud host; the first score is a score for the target power system, and the second score is a score for the host machine in the target power system.

8. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the cloud platform resource scheduling method according to any one of claims 1 to 6 by executing the computer instructions.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the cloud platform resource scheduling method according to any one of claims 1 to 6.

10. A computer program product, characterized in that It includes computer instructions, and the computer instructions are used to enable a computer to execute the cloud platform resource scheduling method described in any one of claims 1 to 6.

Images