CN106533723A

CN106533723A - Virtual resource scheduling method, device and system

Info

Publication number: CN106533723A
Application number: CN201510580104.9A
Authority: CN
Inventors: 于跃波; 蒋天超; 欧阳新志; 左奇; 楚俊生
Original assignee: ZTE Corp
Current assignee: Nanjing ZTE New Software Co Ltd
Priority date: 2015-09-11
Filing date: 2015-09-11
Publication date: 2017-03-22
Anticipated expiration: 2035-09-11
Also published as: WO2017041556A1; CN106533723B

Abstract

The invention provides a virtual resource scheduling method, device and system. The method comprises the following steps: acquiring performance data relevant to a virtualized network function (VNF); determining a scheduling strategy of virtual resources of the VNF according to the performance data; and executing a scheduling operation on the virtual resources according to the scheduling strategy. Through adoption of the virtual resource scheduling method, device and system, the problem of incomplete performance data during dynamic scaling of an existing VNF is solved; resource scaling can be performed according to service index data and system performance data provided by the VNF; and the performance and the stability of a VNF service system are ensured.

Description

Schedule virtual resources method, apparatus and system

Technical field

The present invention relates to the communications field, in particular to a kind of schedule virtual resources method, apparatus and system.

Background technology

At present in the communications industry, government, information technology (IT) department of enterprises and institutions, with carrying for the level of informatization Height, the scale of computing resource are more and more huger, directly miscellaneous service application are deployed on hardware host, not only workload It is increasing, manage also more and more difficult.And the appearance of cloud computing so that what various applications can be unified is deployed on cloud, The difference of various bottom hardwares is shielded not only, management is also increasingly simpler, and virtual management system can be according to industry The ruuning situation of business application, is dynamically controlled to virtual resource.Particularly in telecommunications industry, the special network equipment High cost, safeguards complicated, high energy consumption, and the business function that equipment is provided is single, and by the strength of cloud computing, then may be used To realize the flexible configuration of resource, every cost is reduced.

Current dynamic dispatching method, is typically based on operating system index, for example, the CPU of operating system, internal memory, magnetic The indexs such as disk I/O, the network bandwidth or system visit capacity to control the flexible of virtual resource, between being disadvantageous in that and be all Connect by indexs such as CPU, internal memory, visit capacities reflecting the disposal ability of operating system, and independent use these numbers According to the foundation as scheduling of resource, and the performance of the operation system in the operating system passively can only be perceived；But, although The disposal ability of operating system can reflect the performance of operation system to a certain extent, but this degree of correlation is not Absolute.For example, in the case where the disk resource that operating system distributes to operation system is few, operation system may be The not enough situation of disk resource is occurred in that, but the disk resource of operating system does not occur the situation of deficiency.It can be seen that, Resource scheduling scheme in correlation technique cannot ensure the performance of operation system.

In terms of the dynamic retractility of virtual resource, the virtual resource management system in correlation technique is not interacted with operation system； Although correlation technique solves flexible demand of the system to resource to a certain extent, more complicated for service logic is System, then can not do too many intervention operation, when there is logical relation between business unit, cannot meet to business unit Coordinated operation between business unit.

In telecommunications industry, the network function virtualization (Network proposed according to ETSI (ETSI) Function Virtualization, referred to as NFV) standard, virtualize by basic resource, software defined network (Software Defined Network, referred to as SDN) etc. Intel Virtualization Technology, industrial quarters has been realized in the void of the resources such as Telecommunication Network Element Planization is operated.NFV frameworks can be used as general cloud computing virtual resource management framework, all management based on virtual machine Model can correspond to NFV frameworks.

We under NFV frameworks can pass through virtualized infrastructure manager (Virtualized Infrastructure Manager, referred to as VIM) system performance information is obtained, report interface to give virtualization network work(by NFV invocation performances Can manager (Virtualized Network Function Manager, referred to as VNFM), VNFM is according to collecting Virtualization network function (Virtualized Network Function, referred to as VNF) performance data come do business tune Degree.

Inventor is had found through research, as the data that VIM reports VNFM can only obtain the other property of operating system grade Energy the index such as memory usage of CPU, part system, magnetic disc i/o, for the resource of VNF application system aspects Performance index information can not be obtained, such as disk utilization rate, the systematic memory usage of institute, bandwidth utilization rate, VNF The Special use rate such as the portfolio of respective network elements, percent of call completed；So, if the performance data obtained according to VIM is making Judge for VNF scheduling, equally exist the problem of the performance that cannot ensure operation system.

Schedule virtual resources for individually carrying out VNF according to the other performance indications of operating system grade in correlation technique cannot The problem of the performance of operation system is ensured, effective solution is not yet proposed at present.

The content of the invention

The invention provides a kind of schedule virtual resources method, apparatus and system, at least to solve individually according to operating system The data of rank carry out the problem that the schedule virtual resources of VNF cannot ensure the performance of operation system.

According to an aspect of the invention, there is provided a kind of schedule virtual resources method, including：Obtain and virtualization network The related performance data of function VNF；According to the performance data, the scheduling strategy of the virtual resource of the VNF is determined； According to the scheduling strategy, scheduling operation is performed to the virtual resource.

Alternatively, the performance data includes at least one of：System performance information, operational indicator data；Wherein, The system performance information is the quantized data for weighing the performance of the virtual resource of the deployment VNF；The business Achievement data is the quantized data for weighing the traffic performance of the operation system of the VNF.

Alternatively, the system performance information includes：The system performance information for being gathered and reported by the VNF；By void The system performance information that planization infrastructure manager VIM is gathered and reported.

Alternatively, according to the performance data, determine that the scheduling strategy of the virtual resource of the VNF includes：Converge The system performance information for always being gathered by the VNF and being reported and the system performance information for being gathered and reported by the VIM； The system performance information according to the operational indicator data and after collecting, determines the institute of the virtual resource of the VNF State scheduling strategy.

Alternatively, collect the system performance information that gathered and reported by the VNF and gathered and reported by the VIM System performance information includes：For the system performance information that can be collected from the VNF and from the VIM with from The VNF or the system performance information from VIM collections are defined；Or for from the VNF and from described The system performance information that VIM can be collected carries out aggregation process by Weighted Average Algorithm.

Alternatively, the operational indicator data include at least one of：Business Processing amount within the unit interval, and/ Or, other performance datas related to traffic performance；The system performance information includes at least one of：Central authorities are processed Device cpu busy percentage, memory usage, disk write/read performance indications, remaining disk space, the network bandwidth and utilize Rate, offered load, network interface card loss rate.

Alternatively, before the performance data related to the VNF is obtained, methods described also includes：Transmission is ordered Information is read to the VNF, wherein, the subscription information is used for subscribing to one or more of the performance data, and arranges The report cycle of the performance data.

Alternatively, the subscription information, is additionally operable to arrange the triggering thresholding for reporting the performance data and duration.

Alternatively, the subscription information is being sent to the VNF, methods described also includes：The VNF is received The subscription information；The duration for exceeding the triggering thresholding in the performance data exceedes the situation of the duration Under, the VNF reports the performance data.

Alternatively, the VNF is reported on the report cycle and virtualized infrastructure manager VIM of the performance data The report cycle of reporting system performance data is identical.

Alternatively, performing scheduling operation to the virtual resource includes：Send pre-scheduling to notify to the VNF, wherein, The pre-scheduling is notified for indicating to perform scheduling operation to the virtual resource；Response is confirmed pre-scheduling is received In the case of, scheduling operation is performed to the virtual resource, wherein, the pre-scheduling confirms response for indicating the VNF The Business Processing related to the scheduling operation of the virtual resource is completed, it is allowed to which virtualization system performs resource scaling operation.

Alternatively, after scheduling operation is performed to the virtual resource, methods described also includes：Send scheduling to complete really Recognize notice to the VNF, wherein, the scheduling complete acknowledgement notification for indicate completed to the virtual resource tune Degree operation.

Alternatively, include in the scheduling operation：In the case of increasing virtual resource or change virtual resource, to institute Before stating virtual resource execution scheduling operation, methods described also includes：Receive on virtualized infrastructure manager VIM The physical machine system performance information of the physical machine of report；Selecting the physical machine system performance information to meet the scheduling operation will The first physical machine asked is used as newly-increased virtual resource or the host of the virtual resource of change.

Alternatively, according to the performance data, determine that the scheduling strategy of the virtual resource of the VNF includes：Root According to the performance data, the performance indications of the VNF are calculated；Inquiry preset schedule plan corresponding with the performance indications Slightly, wherein, the preset schedule strategy is used for indicating in the case of the performance indications, the scheduling operation that should be performed.

Alternatively, the performance indications are calculated by following equation：The performance indicationsIts In, w_iRepresent pre-set level weight, the h of i-th performance data of the VNF_iRepresent i-th property of the VNF The value of energy data, s_iRepresent the value of i-th performance data in the VNF oeprations at full load.

According to another aspect of the present invention, a kind of schedule virtual resources device is additionally provided, including：Acquisition module, uses In the acquisition performance data related to virtualization network function VNF；Determining module, for according to the performance data, Determine the scheduling strategy of the virtual resource of the VNF；Performing module, for according to the scheduling strategy, to the void Intend resource and perform scheduling operation.

Alternatively, the determining module includes：Collection unit, for collecting the system for being gathered and reported by the VNF Performance data and the system performance information for being gathered and reported by the VIM；Determining unit, for referring to according to the business The system performance information after marking data and collecting, determines the scheduling strategy of the virtual resource of the VNF.

Alternatively, the collection unit is used for：For the system that can be collected from the VNF and from the VIM Performance data is defined by the system performance information gathered from the VNF or from the VIM；Or for from described VNF and the system performance information that can be collected from the VIM carry out aggregation process by Weighted Average Algorithm.

Alternatively, described device also includes：First sending module, for sending subscription information to the VNF, wherein, The subscription information is used for subscribing to one or more of the performance data, and arranges the report cycle of the performance data.

Alternatively, the performing module includes：Transmitting element, notifies to the VNF for sending pre-scheduling, wherein, The pre-scheduling is notified for indicating to perform scheduling operation to the virtual resource；Performance element, for receive it is pre- In the case that scheduling confirms response, scheduling operation is performed to the virtual resource, wherein, the pre-scheduling confirms that response is used The Business Processing related to the scheduling operation of the virtual resource is completed in the instruction VNF, it is allowed to virtualization system Perform resource scaling operation.

Alternatively, described device also includes：Second sending module, completes acknowledgement notification to the VNF for sending scheduling, Wherein, the scheduling complete acknowledgement notification for indicate completed to the virtual resource scheduling operation.

Alternatively, described device also includes：Receiver module, for the scheduling operation include increase virtual resource or In the case of change virtual resource, the physical machine system of the physical machine that virtualized infrastructure manager VIM is reported is received Performance data；Selecting module, for selecting the physical machine system performance information to meet the first of the scheduling operation requirement Physical machine is used as newly-increased virtual resource or the host of the virtual resource of change.

Alternatively, the determining module includes：Computing unit, for according to the performance data, calculating the VNF Performance indications；Query unit, it is for inquiring about preset schedule strategy corresponding with the performance indications, wherein, described pre- If scheduling strategy is used for indicating in the case of the performance indications, the scheduling operation that should be performed.

Alternatively, the computing unit calculates the performance indications by following equation：The performance indications Wherein, w_iRepresent pre-set level weight, the h of i-th performance data of the VNF_iRepresent the VNF The value of i-th performance data, s_iRepresent the value of i-th performance data in the VNF oeprations at full load.

According to another aspect of the present invention, a kind of schedule virtual resources system is additionally provided, the system includes：Virtually Change infrastructure manager VNFM and virtualization network function VNF, wherein, the VNFM is connected with the VNF Connect, the VNFM includes above-mentioned schedule virtual resources device.

Alternatively, the system also includes：Virtualized infrastructure manager VIM, wherein, the VIM with it is described VNFM connects, and the VIM is used for gathering and reporting the system performance information of the VNF.

By the present invention, using the acquisition performance data related to virtualization network function (VNF)；According to performance data, Determine the scheduling strategy of the virtual resource of VNF；According to scheduling strategy, the mode of scheduling operation is performed to virtual resource, The schedule virtual resources that solving individually carries out VNF according to the other performance indications of operating system grade cannot ensure operation system Performance problem, ensured the performance and stability of operation system.

Description of the drawings

Accompanying drawing described herein is used for providing a further understanding of the present invention, constitutes the part of the application, the present invention Schematic description and description be used for explain the present invention, do not constitute inappropriate limitation of the present invention.In the accompanying drawings：

Fig. 1 is the flow chart of schedule virtual resources method according to embodiments of the present invention；

Fig. 2 is the structured flowchart of schedule virtual resources device according to embodiments of the present invention；

Fig. 3 is the preferred structure schematic diagram one of schedule virtual resources device according to embodiments of the present invention；

Fig. 4 is the preferred structure schematic diagram two of schedule virtual resources device according to embodiments of the present invention；

Fig. 5 is the preferred structure schematic diagram three of schedule virtual resources device according to embodiments of the present invention；

Fig. 6 is the preferred structure schematic diagram four of schedule virtual resources device according to embodiments of the present invention；

Fig. 7 is the preferred structure schematic diagram five of schedule virtual resources device according to embodiments of the present invention；

Fig. 8 is the preferred structure schematic diagram six of schedule virtual resources device according to embodiments of the present invention；

Fig. 9 is the structural representation of NFV frameworks according to the preferred embodiment of the invention；

Figure 10 is the flow chart of schedule virtual resources method according to the preferred embodiment of the invention；

Figure 11 is the schedule virtual resources method in general virtual cloud management system according to the preferred embodiment of the invention Flow chart.

Specific embodiment

Below with reference to accompanying drawing and in conjunction with the embodiments describing the present invention in detail.It should be noted that in the feelings not conflicted Under condition, the feature in embodiment and embodiment in the application can be mutually combined.

It should be noted that description and claims of this specification and the term " first " in above-mentioned accompanying drawing, " second " Etc. being for distinguishing similar object, without for describing specific order or precedence.

A kind of schedule virtual resources method is provided in the present embodiment, and Fig. 1 is virtual resource according to embodiments of the present invention The flow chart of dispatching method, as shown in figure 1, the flow process comprises the steps：

Step S102, obtains the performance data related to virtualization network function VNF；

Step S104, according to performance data, determines the scheduling strategy of the virtual resource of VNF；

Step S106, according to scheduling strategy, performs scheduling operation to virtual resource.

By above-mentioned steps, scheduling strategy is determined according to the performance data related to VNF, and then according to the scheduling strategy Perform the scheduling operation of the virtual resource to VNF；For example, performance data can be system performance information, or Operational indicator data, also, system performance information can also can be obtained from VIM from VNF direct access, with VNF Related performance data can reflect the disposal ability of operation system, therefore, VNF is solved according to single type data Carry out VNF schedule virtual resources cannot ensure operation system performance problem, ensured operation system performance and Stability.

Alternatively, above-mentioned steps can be performed by VNFM or NFVO, or are divided on VNFM and NFVO Do not perform a part therein and another part is operated.In addition, being performed by other management entities in NFV frameworks above-mentioned Step also can be what is be conceived to.

Alternatively, performance data includes but is not limited at least one of：System performance information, operational indicator data；Its In, system performance information is the quantized data for weighing the performance of the virtual resource of deployment VNF；Operational indicator data It is the quantized data for weighing the traffic performance of the operation system of VNF.

Alternatively, system performance information includes：The system performance information for being gathered and reported by VNF；By virtualization basis The system performance information that infrastructure manager VIM is gathered and reported.

As system performance information can be gathered from VNF, it is also possible to gather from VIM, therefore, in step S104, The system performance information for being gathered and reported by VNF and the system performance information for being gathered and reported by VIM can be collected；Enter And the system performance information according to operational indicator data and after collecting, determine the scheduling strategy of the virtual resource of VNF.

Furthermore, it is contemplated that the system performance information collected from VNF and the system performance information collected from VIM may It is the system performance information of same type, therefore, when carrying out system performance information and collecting, can at least adopt following two The mode of kind：

Mode 1, for the system performance information that can be collected from VNF and from VIM is with from VNF or from VIM The system performance information of collection is defined；

Mode 2, for the system performance information that can be collected from VNF and from VIM is carried out by Weighted Average Algorithm Aggregation process.

Above-mentioned system performance information and operational indicator data are used for the service process performance of the operation system for representing VNF, Can represent that the parameter of service process performance has a lot, and according to the type of service of VNF, the Business Processing of concern The parameter of energy is also possible to difference.For example, operational indicator data include but is not limited at least one of：In the unit interval Interior Business Processing amount, and/or, other performance datas related to traffic performance；System performance information includes but does not limit In at least one of：Central processor CPU utilization rate, memory usage, disk write/read performance indications, remain Residual magnetism disk space, network bandwidth utilization factor, offered load, network interface card loss rate.These operational indicator data are to quantify Data.

Alternatively, in step s 102, the operational indicator data and system performance information for getting from VNF are by VNF Periodically or irregularly report；Although VIM can also reporting system performance data, the report cycle of VIM VNF cannot be controlled；And pass through VNF reporting services achievement data and system performance information can then accomplish that business refers to Report while mark data and system performance information, for example, in the same report cycle of VNF, reporting service refers to simultaneously Mark data and system performance information, can be achieved with reporting while operational indicator data and system performance information, thus are avoided that Operational indicator data and system performance information is obtained not by VIM that VNFM or NFVO is obtained by VNF Caused by synchronous, scheduling strategy selects the problem of deviation.

Alternatively, the report cycle of system performance information is identical with the report cycle of operational indicator data, or operational indicator Integral multiple of the report cycle of data for system performance information report cycle.

So, before the system performance information and operational indicator data of virtualization network function VNF is obtained, can be with Subscription information is sent to VNF, wherein, subscription information is used for subscribing to one or more of performance data, and arranges performance number According to report cycle.By the way, personalized subscription can be carried out to the performance data for needing concern.

Alternatively, subscription information is additionally operable to arrange triggering thresholding and the duration of reporting performance data.

Alternatively, after the triggering thresholding and duration for being provided with VNF reporting performance datas, alternatively, sending out Subscription information is sent to VNF, VNF receives subscription information；The duration for exceeding triggering thresholding in performance data surpasses In the case of crossing duration, VNF reporting performance datas.By the way, it is to avoid real-time report performance data is led A large amount of occupancy of the system resource of cause.

Alternatively, when carrying out the scheduling operation of virtual resource in step s 106, pre-scheduling can be sent and is notified to VNF, Wherein, pre-scheduling is notified for indicating to perform scheduling operation to virtual resource；And response is confirmed pre-scheduling is received In the case of, scheduling operation is performed to virtual resource, wherein, pre-scheduling confirms response for indicating that VNF has been completed and void Intend the related Business Processing of the scheduling operation of resource, it is allowed to which virtualization system performs resource scaling operation.By the way, There is provided a kind of mode notified before scheduling, the normal execution of business on VNF can be ensured, it is to avoid due to void Intend service disconnection or exception caused by scheduling of resource.

Alternatively, after scheduling operation being performed to virtual resource, can also send scheduling and complete acknowledgement notification to VNF, Wherein, scheduling complete acknowledgement notification for indicate completed the scheduling operation to virtual resource.By the way so that VNF can perceive the performance of schedule virtual resources, and then carry out subsequent treatment, and for example, VNF is being received Scheduling starts with newly assigned virtual resource and performs business operation after completing acknowledgement notification.

Alternatively, the performance indications of physical machine can be reported by VIM, in some increase virtual resource (for example, empty machines The operations such as upper electricity, newly-increased void machine) or change virtual resource (such as empty machine migration) in the case of, then need to consider thing The select permeability of reason machine.In general, the physical machine for selecting at least will can provide enough virtual resources on empty machine Electricity, newly-increased void machine or empty machine migration are used.For example, before step S106, VNFM can receive virtualization base The physical machine system performance information of the physical machine that Infrastructure manager VIM is reported；VNFM selects physical machine systematic function Data meet the first physical machine of scheduling operation requirement as newly-increased virtual resource or the host of the virtual resource of change Machine.Furthermore, it is possible to the physical machine in scheduling operation selecting resource utilization low is used as newly-increased virtual resource or change Virtual resource host.

Alternatively, in step S104, it is determined that during scheduling strategy, general way is computational using weighting scheme The weighted value of energy data, and then, the scheduling strategy corresponding to this weighted value is selected according to the weighted value for calculating.Property The weight of energy data can be configured as needed, for example, in the case where the service feature of certain VNF is more paid close attention to, Can the weight of this operational indicator data be arranged some larger.

Alternatively, the business of VNF can in step S104, according to operational indicator data and system performance information, be calculated Performance indications；Inquiry preset schedule strategy corresponding with service feature index, wherein, preset schedule strategy is used for indicating In the case of service feature index, the scheduling operation that should be performed.These scheduling operations include：It is electric on empty machine, it is lower electricity, stretch Contracting CPU, flexible internal memory, extension disk, newly-built empty machine, deletion void machine, the empty machine of migration etc..Preset schedule strategy can be with Configured according to service needed respectively.

It is alternatively possible to calculate service feature index by following equation：Service feature index Wherein, w_iRepresent pre-set level weight, the h of i-th achievement data of VNF_iRepresent i-th achievement data of VNF Value, s_iRepresent the value of i-th achievement data in VNF oeprations at full load；Achievement data includes：Operational indicator data And system performance information.

Through the above description of the embodiments, those skilled in the art is can be understood that according to above-described embodiment Method can add the mode of required general hardware platform to realize by software, naturally it is also possible to by hardware, but a lot In the case of the former is more preferably embodiment.Based on such understanding, technical scheme is substantially in other words to existing There is the part that technology contributes embody in the form of software product, the computer software product is stored in one In storage medium (such as ROM/RAM, magnetic disc, CD), use so that a station terminal equipment (can including some instructions Being mobile phone, computer, server, or network equipment etc.) method that performs each embodiment of the invention.

Additionally provide a kind of schedule virtual resources device in the present embodiment, the device is used for realizing above-described embodiment and preferably Embodiment, had carried out repeating no more for explanation.As used below, term " module " can realize predetermined work( The combination of the software and/or hardware of energy.Although the device described by following examples is preferably with software realizing, Hardware, or the realization of the combination of software and hardware is also may and to be contemplated.

Fig. 2 is the structured flowchart of schedule virtual resources device according to embodiments of the present invention, as shown in Fig. 2 the device bag Include：Acquisition module 22, determining module 24 and performing module 26, wherein, acquisition module 22, for obtaining and virtualizing Network function VNF related performance data；Determining module 24, coupled to acquisition module 22, for according to performance data, Determine the scheduling strategy of the virtual resource of VNF；Performing module 26, coupled to determining module 24, for according to scheduling plan Slightly, scheduling operation is performed to virtual resource.

Alternatively, performance data includes at least one of：System performance information, operational indicator data；Wherein, system Performance data is the quantized data for weighing the performance of the virtual resource of deployment VNF；Operational indicator data are for weighing The quantized data of the traffic performance of the operation system of amount VNF.

Fig. 3 is the alternative construction block diagram one of schedule virtual resources device according to embodiments of the present invention, as shown in figure 3, can Selection of land, determining module 24 include：Collection unit 242, for collecting the system performance information for being gathered and reported by VNF With the system performance information gathered and reported by VIM；Determining unit 244, coupled to collection unit 242, for basis Operational indicator data and the system performance information after collecting, determine the scheduling strategy of the virtual resource of VNF.

Alternatively, collection unit is used for：For the system performance information that can be collected from VNF and from VIM with from VNF or the system performance information from VIM collections are defined；Or for can collect from VNF and from VIM System performance information carry out aggregation process by Weighted Average Algorithm.

Alternatively, operational indicator data include at least one of：Business Processing amount within the unit interval, and/or, Other performance datas related to traffic performance；System performance information includes at least one of：Central processor CPU Utilization rate, memory usage, disk write/read performance indications, remaining disk space, network bandwidth utilization factor, net Network load, network interface card loss rate.

Fig. 4 is the alternative construction block diagram two of schedule virtual resources device according to embodiments of the present invention, as shown in figure 4, can Selection of land, device also include：First sending module 42, coupled to VNF, for sending subscription information to VNF, wherein, Subscription information is used for subscribing to one or more of performance data, and arranges the report cycle of performance data.

Alternatively, subscription information, is additionally operable to arrange triggering thresholding and the duration of reporting performance data.

Alternatively, the report cycle of VNF reporting performance datas and virtualized infrastructure manager VIM reporting systems The report cycle of energy data is identical.

Fig. 5 is the alternative construction block diagram three of schedule virtual resources device according to embodiments of the present invention, as shown in figure 5, can Selection of land, performing module 26 include：Transmitting element 262, coupled to VNF, notifies to VNF for sending pre-scheduling, Wherein, pre-scheduling is notified for indicating to perform scheduling operation to virtual resource；Performance element 264, coupled to VNF, For in the case where pre-scheduling confirmation response is received, performing scheduling operation to virtual resource, wherein, pre-scheduling confirms Respond for indicating that VNF has completed the Business Processing related to the scheduling operation of virtual resource, it is allowed to which virtualization system is held Row resource scaling operation.

Fig. 6 is the alternative construction block diagram four of schedule virtual resources device according to embodiments of the present invention, as shown in fig. 6, can Selection of land, device also include：Second sending module 62, coupled to performing module 26 and VNF, completes for sending scheduling Acknowledgement notification to VNF, wherein, scheduling complete acknowledgement notification for indicate completed the scheduling operation to virtual resource.

Fig. 7 is the alternative construction block diagram five of schedule virtual resources device according to embodiments of the present invention, as shown in fig. 7, can Selection of land, device also include：Receiver module 72, coupled to VIM, for scheduling operation include increasing virtual resource or In the case of person's change virtual resource, the physical machine system of the physical machine that virtualized infrastructure manager VIM is reported is received System performance data；Selecting module 74, coupled to receiver module 72 and performing module 26, for selecting physical machine systematicness Energy data meet the first physical machine of scheduling operation requirement as newly-increased virtual resource or the place of the virtual resource of change Main frame.

Fig. 8 is the alternative construction block diagram six of schedule virtual resources device according to embodiments of the present invention, as shown in figure 8, can Selection of land, determining module 24 include：Computing unit 246, for according to performance data, calculating the performance indications of VNF； Query unit 248, coupled to computing unit 246, for inquiring about preset schedule strategy corresponding with performance indications, wherein, Preset schedule strategy is used for instruction in the case of performance indications, the scheduling operation that should be performed.

Alternatively, computing unit passes through following equation calculation of performance indicators：Performance indicationsIts In, w_iRepresent pre-set level weight, the h of i-th performance data of VNF_iRepresent i-th performance data of VNF Value, s_iRepresent the value of i-th performance data in VNF oeprations at full load.

It should be noted that above-mentioned modules can be by software or hardware to realize, for the latter, Ke Yitong Cross in the following manner realization, but not limited to this：Above-mentioned module is respectively positioned in same processor；Or, above-mentioned module distinguishes position In multiple processors.

The present embodiment additionally provides a kind of schedule virtual resources system, and the system is used for realizing above-mentioned schedule virtual resources method. The system includes：Virtualized infrastructure manager (VNFM) and virtualization network function (VNF), wherein, VNFM It is connected with VNF, VNFM includes above-mentioned schedule virtual resources device.

Alternatively, system also includes：Virtualized infrastructure manager (VIM), wherein, VIM is connected with VNFM, VIM is used for gathering and reporting the system performance information of VNF.

Embodiments of the invention additionally provide a kind of software, and the software is used for performing in above-described embodiment and preferred embodiment The technical scheme of description.

Embodiments of the invention additionally provide a kind of storage medium.In the present embodiment, above-mentioned storage medium can be set It is used for performing the program code of following steps for storage：

Alternatively, in the present embodiment, above-mentioned storage medium can be including but not limited to：USB flash disk, read-only storage (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as For RAM), portable hard drive, magnetic disc or CD etc. are various can be with the medium of store program codes.

Alternatively, the specific example in the present embodiment may be referred to showing described in above-described embodiment and optional embodiment Example, the present embodiment will not be described here.

In order that the description of the embodiment of the present invention is clearer, it is described with reference to preferred embodiment and illustrates.

The preferred embodiment of the present invention has VNF reporting systems performance and operational indicator data there is provided one kind, is virtually provided The resource regulating method of source dynamic retractility；Integrated system performance data and operational indicator data, determine the virtual money of VNF The scheduling strategy in source；According to scheduling strategy, the mode of scheduling operation is performed to virtual resource, solved individually according to operation The performance indications of system level, or the schedule virtual resources for carrying out VNF separately through service feature index cannot ensure industry The problem of the performance of business system, has ensured the performance of operation system.

In NFV frameworks, VIM can report performance data (CPU, internal memory, the magnetic of VNF underlying operating system levels Disk I/O) VNFM is given, but some application layer performance data cannot be obtained from VIM, such as disk remaining space, net Network handling capacity, network delay, network usage and VNF own service achievement datas；And VNF performance indications The report cycle of data carrys out dynamic setting by VNF, and the report cycle of VIM is typically all the unified time cycle for determining. In a preferred embodiment of the invention, VNF can tuning performance achievement data as needed report cycle, and for adjusting Degree decision business performance data and system performance information can be reported by VNF, this ensures that there data Unification on time dimension so that the configuration of traffic scheduling more flexibility.In the preferred embodiment, in scheduling The performance data of the physical machine that can also be reported with reference to VIM, selects more reasonably host when being conducive to flexible Operation VNF.

The preferred embodiment of the present invention by taking NFV frameworks as an example is described and illustrates.Fig. 9 is preferable to carry out according to the present invention The structural representation of the NFV frameworks of example, as shown in figure 9, the NFV frameworks include：

Network function virtualization architecture (Network Functions Virtualization Infrastructure, referred to as NFVI, comprising deployment VNF apply required software, hardware (hardware, abbreviation HW) environment), VIM, VNFM, VNF, can also include：Network function virtualization telegon (Network Functions Virtualization Orchestrator, Referred to as NFVO)；NFVO can replace VNFM to perform scheduling operation.

In a preferred embodiment of the invention：

VIM includes following functions：

1. virtual resource can be managed.

2. the expanding-contracting action of VNF can be performed.

3. the system performance information of host and client computer is gathered；

VNF includes following functions：

1. the measurement period of corresponding system performance data and operational indicator data, report cycle are configured.

2. in the measurement period for pre-setting, according to the characteristic of VNF, correlation method is write, generate the industry in this cycle Business achievement data and system performance information, system performance information and operational indicator data are included but is not limited to：CPU/ internal memories/ Disk occupancy, while it is also conceivable to magnetic disc i/o and network bandwidth utilization rate and network throughput, and special VNF The property index of network element.

3. can be according to the dynamic retractility operational order of VNFM or NVFO, to VNF process.Such as：Work as pass The business of the empty machine when closing a virtual machine, can be terminated, the business of this virtual machine is transferred to into other not by VNFM Close empty machine to be processed or start new VNF to continue with business in other nodes, it is ensured that the flatness of business.

4. ensure to meet operation system minimum resources demand, when certain empty machine is closed in the instruction that traffic scheduling center is transmitted, When but the number of resources for causing operation system to be possessed being less than minimum resources number after closing the empty machine, can refuse to close the void Machine.

5. the virtual resource dynamic retractility result that can be transmitted according to VNFM or NFVO, is processed to business.Such as： When a new void machine is extended, new empty machine has been started up finishing, and operation system gets VFNM or NFVO and passes Come resource scaling operation result, then operation system can initialize the business of the empty machine so that the empty machine can be with other Empty machine processing business together.

VNFM or NFVO include following functions：

1.VNFM can preserve system performance information and the operational indicator data that operation system is reported.

2.VNFM receives operational indicator data and the system performance information that VNF is reported.

3.VNFM or NFVO can create the plan for carrying out dynamic retractility according to system performance information and operational indicator data Omit execution mechanism.

4. transmission dynamic retractility instruction can be given.

5. virtual resource dynamic retractility result can determine.

6. the system performance information of host that VIM reports and client computer can be received.

Based on above-mentioned NFV frameworks, what the preferred embodiment of the present invention was proposed carries out dynamic retractility according to VNF reported datas Dispatching method comprises the steps：

Step one, the business datum generation module of each business unit of operation system, according to operational indicator generation method, Operational indicator data are generated in real time, and these data have reacted the traffic handing capacity of operation system.It is provided simultaneously with generating system System performance data ability, reports VNFM.

Step 2, the operational indicator data acquisition module of operation system are carried out to the operational indicator data of each business unit Aggregation process generates reported data, passes to VNFM.

Whether the scheduling strategy that step 3, VNFM or NFVO are configured in judging system meets trigger condition, if full Foot, then do preoperative Notification Validation flow process according to interface interchange VNF.Scheduling strategy includes that the operational indicator for arranging is worked as Amount and system energy data target, system performance information index emphasis rely on CPU/ internal memories/disk occupancy, while considering magnetic Disk I/O and network bandwidth utilization rate and network throughput.

Step 4, after VNF receives the notice before scaling operation, carries out the pretreatment before scaling operation inside VNF.

Step 5, after VNF completes the pretreatment before scaling operation, and the confirmation before VNFM or NFVO are operated.

After step 6, NFVO or VNFM receive confirmation, notify that VIM carries out resource scaling operation.

Step 7, after VIM completes resource scaling operation, VNFM or NFVO notifies that operation system completes to stretch

Step 8, VNF receive resource and stretch after the notice being disposed, will it is flexible after change in resources situation, be saved in In operation system.

In a preferred embodiment of the invention, active by business unit come produce need concern system performance information and business Achievement data, for the condition of resource dynamic retractility triggering.Configuration system strategy integrated service and performance of both system Index is carried out synthetic determination and is stretched, while reporting the data such as magnetic disc i/o and network throughput and utilization rate to be in VNF System early warning is stretched, it is ensured that system each latitude it is reliable and stable.

Additionally, the flexible method for carrying out in front and back notifying, confirming of virtual resource is additionally provided in a preferred embodiment of the invention, So that operation system, can be with the flexible impact to system of more flexible control virtual resource so that after resource is flexible, be System can be more stable operation.

The scheme provided using the preferred embodiment of the present invention, can solve traditional according to the indirectly index such as CPU, internal memory Carry out the deficiency of dynamic adjustment virtual resource, can more reflect the close relation of business and resource, so really could reflect The disposal ability of VNF.It can be seen that, compared with prior art, the perfect operation system of this preferred embodiment to own service at Reason ability is controlled, and has reached the effect of dynamic retractility resource, has saved manpower, improves the stable place of operation system The ability of reason business.

The preferred embodiment of the present invention is described and is illustrated with two examples below.

Preferred embodiment 1

Figure 10 is the flow chart of schedule virtual resources method according to the preferred embodiment of the invention, as shown in Figure 10, the stream Journey comprises the steps：

Step S1001, arranges the acquisition time interval of VNF service data acquisition modules.

Step S1002, arranges the time interval that VNF operational indicator data are reported.

The scheduling strategy of VNFM can also be set here simultaneously, be follow-up flexible as decision-making setting.Strategy setting is It is controlled according to the equivalent percent that operational indicator is calculated, while the system performance information of VNF itself is taken into account, In order to ensure the reliability service of VNF, system disk I/O detections can be set and judged and network throughput judgement, to guarantee The reliability of VNF hardware.

Step S1003, is deployed in the operational indicator data acquisition module of each VNF, according to the service needed of itself, The system performance information and operational indicator initial data of collection itself.These system performance informations and achievement data can include But it is not limited to：In unit interval, Business Processing amount, operation system CPU, the utilization rate of internal memory, magnetic disc i/o, disk are empty Between remaining size, the network bandwidth, offered load, loss rate, the property index of special VNF etc..

Step S1004, VNF obtain the performance indications and operating system aspect of business datum according to the sampling period for arranging Performance data, keeps unified the acquisition granularity, and by these reporting performance datas to VNFM.VNFM needs to provide industry Business reporting performance data interface and VNF system performance information interfaces, VNF system performance informations are included and are not limited to and business System correlation：CPU, internal memory, magnetic disc i/o, disk remaining space, bandwidth are handled up and the data such as network usage. With reference to business own characteristic, final data requires that carrying out acquisition reports.VNFM will provide general tolerance and report interface.It is special It is not the data such as disk remaining space of some applications that cannot be obtained from VIM, network usage etc..

The operational indicator initial data of each VNF node is carried out aggregation process by step S1005, VNFM, according to calculation Method is converted to the operational indicator data that scheduling needs.The calculation of this operational indicator data can have various ways, under Face enumerates one：

In a service data acquisition cycle, according to：(1) number n of VNF business datums index item, (2) step Business initial data h of the VNF nodes of S1003 collections, weight w of (3) each operational indicator raw value, (4) Value s during each index item oeprations at full load of VNF, is calculated, and obtains operational indicator data, and computing formula can be with It is equation below：

Step S1006, VIM call the interface periodic report host machine system energy data with VNFM to make in VNFM Refer to for scheduling resource.Do flexible decision-making to use to follow-up VNFM, in telescopic process, host such as can be selected to provide Source utilization rate it is low as target machine, decision-making performs the policy development according to VNFM.

Step S1007, VNFM are scheduled the judgement of strategy according to the operational indicator data for reporting.Scheduling strategy VNFM The strategy of the triggering virtual resource scaling operation of middle configuration, which carries out the flexible behaviour of resource according to the value of miscellaneous service achievement data The scheduling of work.

Wherein, scaling operation can be the establishment of VNF nodes, delete, and migration is hung up, wake up etc., can basis Needs are configured.

Such as two traffic scheduling strategies of system configuration：

Strategy 1, when Business Processing amount more than system processing power 60%, will increase VNF1-2.

Strategy 2, when Business Processing amount less than system processing power 40%, will delete VNF1-2.

So when the operational indicator data for reporting are more than 60%, for example, value is reported to be 61%, then scheduling of resource module will be touched Send out the strategy of newly-increased VNF1-2.

When reporting value to be less than 40%, the strategy of VNF1-2 power supplys is deleted in triggering.

Simultaneity factor can be with configuration service performance scheduling strategy, and such as CPU, internal memory, magnetic disc i/o are when certain interval Time is migrated according to service feature.This strategy needs and traffic scheduling strategy is comprehensively arranged, because different business can Can be different to CPU and internal memory, the requirement of magnetic disc i/o, some business are such as message-oriented middlewares, and internal memory is used The calculating service require, having is required to CPU, and some file storage services are had high demands to internal memory and magnetic disc i/o. Business scenario will be considered when this Provisioning Policy.

To CPU, magnetic disc i/o, network throughput and network usage situation carry out system health judgement, if do not met Expection can migrate VNF and run to other purposes main frame, it is ensured that business normal reliable.

Step S1008, VNFM call notification interface before the business scaling operation of VNF, send logical before scaling operation Know.

Step S1009, VNF according to receive it is flexible before notify, do that the strategy inside VNF is pre-processed or business is cut The operation such as change, it is ensured that it is flexible perform in do not affect original VNF institutes functional.

For example, when strategy is for deleting VNF1-2, the business on VNF1-2 is transferred to VNF1-1 and is carried out by VNFM Process, and judge whether VNF1-2 can delete, if it is then configuring the related service of VNF1-2 in systems Information is time-out process.If VNF1-2 currently processed business does not terminate, this scaling operation terminates, and treats next time Scaling operation is triggered, and when going to this step, then judge to process.

Step S1010, VNF confirm interface before calling the scaling operation that VNFM is provided, and confirmation can be held by VNFM Row scaling operation.

Step S1011, after VNFM is confirmed before getting the scaling operation of step S1010, calls VIM to carry out resource Scaling operation.

Step S1012, VIM perform scaling operation.As needed to delete VNF1-2 in execution step S1009, at this moment VNF1-2 is deleted.

Step S1013, VNFM call the resource of the VNF nodes of correlation to stretch after the execution of resource scaling operation terminates Notification interface after contracting operation, notifies that the scaling operation of VNF is over.

Step S1014, related VNF voluntarily process related procedure after notice is received as needed.

For example, if step S1009 is for increasing VNF1-2, step S1013 will notify other related VNF nodes, The information of the VNF1-2 to increasing newly is processed in this step related VNF, and subsequently dependency number is being forwarded to which According to having carried out data distribution.If step S1009 is to delete VNF1-2, same other VNF after the notice is received, The information of the VNF1-2 to deleting is processed, it is to avoid be sent to data.

It should be noted that above-mentioned process is persistently carried out, system is periodically detected the trigger condition of scheduling operation, Once the system performance information for reporting and operational indicator data meet the trigger condition of scheduling strategy, tactful configuration is carried out Scaling operation.

Preferred embodiment 2

Below for general virtualization cloud management, the preferred embodiment of the present invention is described.

Figure 11 is the flow process for virtualizing the schedule virtual resources method in cloud management system according to the preferred embodiment of the invention Figure, as shown in figure 11, the method comprises the steps：

Step S1101, arranges operational indicator data collection interval in operation system.

Step S1102, at traffic scheduling center, arranges the time interval of reporting service achievement data.

Step S1103, business datum generation module of the operational indicator data by the service node being deployed on each empty machine, According to the Business Processing amount of itself, the system performance information and operational indicator initial data of itself is generated.

These system performance informations and achievement data can be including but not limited to：Business Processing amount, business system in unit interval System CPU, the utilization rate of internal memory, disk I/O, the remaining size of disk, network bandwidth etc..

Step S1104, the operational indicator data acquisition module of operation system, according to the time interval that step S1101 is arranged, The operational indicator initial data that collection step S1103 each service node is generated, and each service node initial data is entered Row aggregation process, is encapsulated as the operational indicator data for needing to report.

The calculation of this operational indicator data can have various ways, be set forth below one：

In a service data acquisition cycle, according to：(1) quantity n of the service node of normal operation, (2) business Disposal ability s during node single node oepration at full load, the original service number of each service node of (3) step 3 collection According to h, weight w (acquiescence can be 1) of (4) each service node, calculated, obtained operational indicator data, Computing formula can be equation below：

Step S1105, traffic scheduling center call the operational indicator data that schedule virtual resources system is provided to report interface, The operational indicator data that step S1104 is generated are reported to schedule virtual resources system.

Step S1106, the scheduling of resource module of schedule virtual resources system are scheduled according to the operational indicator data for reporting The judgement of strategy.Scheduling strategy be virtualize scheduling system in configure triggering virtual resource scaling operation strategy, its root The scheduling of resource scaling operation is carried out according to the value of miscellaneous service achievement data.

Scaling operation can be electric, flexible CPU electric on empty machine, lower, flexible internal memory, extend disk, newly-built empty machine, deletes Except empty machine, empty machine of migration etc. can be configured as needed.

Such as two scheduling strategies of system configuration：

Strategy 1, when Business Processing amount more than system processing power 60%, by 1 power supply opening of empty machine.

Strategy 2, when Business Processing amount less than system processing power 40%, by 1 power-off of empty machine.

So when the operational indicator data for reporting are more than 60%, for example, value is reported to be 61%, then scheduling of resource module will be touched Send out the strategy for opening 1 power supply of empty machine.

When reporting value to be less than 40%, the strategy of 1 power supply of empty machine is closed in triggering.

It should be noted that if i.e. operation system does not have reporting service achievement data, step S1106 is also to perform, This is the task that timing is performed, and this task has been provided simultaneously with the data that judgement business is reported, and judges that virtual resource is adjusted The empty machine performance data of degree system itself collection, determines the execution of scheduling strategy.I.e., it is possible to only judge what business was reported Operational indicator data, it is also possible to only judge the empty machine performance data of empty level resource scheduling system itself collection, it is also possible to two Individual data are all judged.

In addition, strategy related in step S1106 can be created to the real needs of virtual resource according to operation system, For example：

1. can create one it is tactful, when empty machine is migrated, force the migration of empty machine can only be in the main frame for having compatibility with which Under migrated.Compatibility between empty machine and main frame, meet empty machine different applications is had different I/O speed, The specially treated during demands such as CPU computing capabilitys, the network bandwidth, storage.

2. can create one it is tactful, the empty machine with exclusiveness can not be run under same main frame.Exclusiveness can expire The redundant processing unit of sufficient identical services, it is impossible to while when being deployed in the demand of same main frame.

3. when off-peak hours sections such as nights, it is impossible to which the empty machine of lower electricity is moved on same main frame, and will be main by calling interface Machine power-off, when regular time periods on daytime, then will call on main frame electrical interface will electricity on main frame, and be deployed in originally these masters Empty machine migration on machine is returned, and reaches energy-saving purpose with this.

Additionally, strategy can also have a lot, here will not enumerate.

Step S1107, schedule virtual resources system call the business operation notification interface at traffic scheduling center, send flexible Preoperative notice.

Step S1108, operation system processing module according to traffic scheduling center receive it is flexible before notice, to corresponding Business is judged, and is processed.

For example, when electrically operated under strategy is for empty machine 1, the business on empty machine 1 is transferred to which by operation system processing module He is processed normal void machine, and judges whether current void machine 1 can process this scaling operation, if it is then The related service information for configuring empty machine 1 in systems is time-out process.If the currently processed business of empty machine 1 does not terminate, Then this scaling operation terminates, and treats next scaling operation triggering, and when going to this step, then judge to process.

Step S1109, after operation system processing module completes the business processing flow variation for causing because resource is stretched, Interface is confirmed before the scaling operation for call empty level scheduling system to provide by traffic scheduling center, confirmation can be by virtual resource Scheduling system performs scaling operation.

Step S1110, after schedule virtual resources system is confirmed before getting the scaling operation of step S1109, notifies resource Management module carries out resource scaling operation.

Step S1111, resource management module perform scaling operation.As needed 1 time electricity behaviour of empty machine in execution step S1108 Make, at this moment empty machine 1 is by lower electricity.

Step S1112, schedule virtual resources system call traffic scheduling center after the execution of resource scaling operation terminates Notification interface after resource scaling operation, notifies that the scaling operation of virtual resource is over.

Step S1113, notifies after the scaling operation that operation system processing module is received according to traffic scheduling center, to industry Business system carries out follow-up process.

For example, if step S1108 is the power on operation of empty machine 1, directly can return successfully in step S1108, Do not process, business is processed in step step S1113.In step S1113, operation system is to 1 top of empty machine The application of administration is initialized, and adds the business information of empty machine 1 in operation system, after completing other configurations work, Business operation is distributed to empty machine 1 to be processed.

It should be noted that above-mentioned process is persistently carried out, system is periodically detected the trigger condition of scheduling operation, Once the operational indicator data for reporting meet the trigger condition of scheduling strategy, the scaling operation of strategy configuration is carried out.

Obviously, those skilled in the art should be understood that each module or each step of the above-mentioned present invention can be with general Realizing, they can concentrate on single computing device computing device, or be distributed in multiple computing devices and constituted Network on, alternatively, they can be realized with the executable program code of computing device, it is thus possible to by they Storage in the storage device by computing device performing, and in some cases, can be holding different from order herein The shown or described step of row, or they are fabricated to each integrated circuit modules respectively, or will be many in them Individual module or step are fabricated to single integrated circuit module to realize.So, the present invention is not restricted to any specific hardware Combine with software.

The preferred embodiments of the present invention are the foregoing is only, the present invention is not limited to, for the technology of this area For personnel, the present invention can have various modifications and variations.It is all within the spirit and principles in the present invention, made it is any Modification, equivalent, improvement etc., should be included within the scope of the present invention.

Claims

1. a kind of schedule virtual resources method, it is characterised in that include：

Obtain the performance data related to virtualization network function VNF；

According to the performance data, the scheduling strategy of the virtual resource of the VNF is determined；

According to the scheduling strategy, scheduling operation is performed to the virtual resource.

2. method according to claim 1, it is characterised in that

The performance data includes at least one of：System performance information, operational indicator data；

Wherein, the system performance information is the quantization number for weighing the performance of the virtual resource of the deployment VNF According to；The operational indicator data are the quantized datas for weighing the traffic performance of the operation system of the VNF.

3. method according to claim 2, it is characterised in that the system performance information includes：

The system performance information for being gathered and reported by the VNF；

The system performance information for being gathered and being reported by virtualized infrastructure manager VIM.

4. method according to claim 3, it is characterised in that according to the performance data, determine the void of the VNF The scheduling strategy for intending resource includes：

Collect the system performance information for being gathered and reported by the VNF and the systematicness for being gathered and reported by the VIM Can data；

The system performance information according to the operational indicator data and after collecting, determines the virtual money of the VNF The scheduling strategy in source.

5. method according to claim 4, it is characterised in that collect the systematic function for being gathered and reported by the VNF Data and the system performance information for being gathered and reported by the VIM include：

For the system performance information that can be collected from the VNF and from the VIM with from the VNF or It is defined from the system performance information of VIM collections；Or

For the system performance information that can be collected from the VNF and from the VIM is by Weighted Average Algorithm Carry out aggregation process.

6. method according to any one of claim 1 to 5, it is characterised in that

The operational indicator data include at least one of：Business Processing amount within the unit interval, and/or, Other performance datas related to traffic performance；

The system performance information includes at least one of：Central processor CPU utilization rate, memory usage, Disk writes/reads performance indications, remaining disk space, network bandwidth utilization factor, offered load, network interface card bag and loses Mistake rate.

7. method according to claim 1, it is characterised in that obtaining the performance data related to the VNF Before, methods described also includes：

Subscription information is sent to the VNF, wherein, the subscription information is used for subscribing to one of the performance data Or multiple, and the report cycle of the performance data is set.

8. method according to claim 7, it is characterised in that

The subscription information, is additionally operable to arrange the triggering thresholding for reporting the performance data and duration.

9. method according to claim 8, it is characterised in that sending the subscription information to the VNF, Methods described also includes：

The VNF receives the subscription information；

In the case where the duration that the performance data exceedes the triggering thresholding exceedes the duration, institute State VNF and report the performance data.

10. method according to claim 7, it is characterised in that

The VNF reports the report cycle and virtualized infrastructure manager VIM reporting systems of the performance data The report cycle of performance data is identical.

11. methods according to claim 1, it is characterised in that performing scheduling operation to the virtual resource includes：

Send pre-scheduling and notify that wherein, the pre-scheduling is notified will be to the virtual money for instruction to the VNF Source performs scheduling operation；

In the case where pre-scheduling confirmation response is received, scheduling operation, wherein, institute are performed to the virtual resource Stating pre-scheduling confirms response for indicating that the VNF has completed the industry related to the scheduling operation of the virtual resource Business is processed, it is allowed to which virtualization system performs resource scaling operation.

12. methods according to claim 1, it is characterised in that after scheduling operation is performed to the virtual resource, Methods described also includes：

Send scheduling and complete acknowledgement notification to the VNF, wherein, the scheduling complete acknowledgement notification for indicate Complete the scheduling operation to the virtual resource.

13. methods according to claim 1, it is characterised in that include in the scheduling operation：Increase virtual resource or In the case of person's change virtual resource, before scheduling operation is performed to the virtual resource, methods described also includes：

Receive the physical machine system performance information of the physical machine that virtualized infrastructure manager VIM is reported；

The physical machine system performance information is selected to meet the first physical machine of the scheduling operation requirement as newly-increased Virtual resource or change virtual resource host.

14. methods according to any one of claim 1 to 13, it is characterised in that according to the performance data, really The scheduling strategy of the virtual resource of the fixed VNF includes：

According to the performance data, the performance indications of the VNF are calculated；

Inquiry preset schedule strategy corresponding with the performance indications, wherein, the preset schedule strategy is used for indicating In the case of the performance indications, the scheduling operation that should be performed.

15. methods according to claim 14, it is characterised in that the performance indications are calculated by following equation：

The performance indications

S = Σ_{i = 1}^{n} (w_{i} * h_{i} / s_{i});

Wherein, w_iRepresent pre-set level weight, the h of i-th performance data of the VNF_iRepresent the VNF's The value of i-th performance data, s_iRepresent the value of i-th performance data in the VNF oeprations at full load.

16. a kind of schedule virtual resources devices, it is characterised in that include：

Acquisition module, for obtaining the performance data related to network function VNF is virtualized；

Determining module, for according to the performance data, determining the scheduling strategy of the virtual resource of the VNF；

Performing module, for according to the scheduling strategy, performing scheduling operation to the virtual resource.

17. devices according to claim 16, it is characterised in that

18. devices according to claim 17, it is characterised in that the system performance information includes：

19. devices according to claim 18, it is characterised in that the determining module includes：

Collection unit, for collecting the system performance information that gathered and reported by the VNF and being gathered by the VIM And the system performance information for reporting；

Determining unit, for the system performance information according to the operational indicator data and after collecting, determines institute State the scheduling strategy of the virtual resource of VNF.

20. devices according to claim 19, it is characterised in that the collection unit is used for：

21. devices according to any one of claim 16 to 20, it is characterised in that

22. devices according to claim 16, it is characterised in that described device also includes：

First sending module, for sending subscription information to the VNF, wherein, the subscription information is used for subscribing to One or more of the performance data, and the report cycle of the performance data is set.

23. devices according to claim 22, it is characterised in that

24. devices according to claim 22, it is characterised in that

25. devices according to claim 16, it is characterised in that the performing module includes：

For sending pre-scheduling, transmitting element, notifies that wherein, the pre-scheduling is notified for indicating to the VNF Scheduling operation will be performed to the virtual resource；

Performance element, in the case where pre-scheduling confirmation response is received, performing scheduling to the virtual resource Operation, wherein, the pre-scheduling confirms response for indicating that the VNF has completed the scheduling with the virtual resource The related Business Processing of operation, it is allowed to which virtualization system performs resource scaling operation.

26. devices according to claim 16, it is characterised in that described device also includes：

Second sending module, completes acknowledgement notification to the VNF for sending scheduling, and wherein, the scheduling is completed Acknowledgement notification is used for indicating to have completed the scheduling operation to the virtual resource.

27. devices according to claim 16, it is characterised in that described device also includes：

Receiver module, in the case of including increasing virtual resource or change virtual resource in the scheduling operation, Receive the physical machine system performance information of the physical machine that virtualized infrastructure manager VIM is reported；

Selecting module, for selecting the physical machine system performance information to meet the first thing of the scheduling operation requirement Reason machine is used as newly-increased virtual resource or the host of the virtual resource of change.

28. devices according to any one of claim 16 to 27, it is characterised in that the determining module includes：

Computing unit, for according to the performance data, calculating the performance indications of the VNF；

Query unit, for inquiring about preset schedule strategy corresponding with the performance indications, wherein, the default tune Degree strategy is used for instruction in the case of the performance indications, the scheduling operation that should be performed.

29. devices according to claim 28, it is characterised in that the computing unit calculates described by following equation Performance indications：

The performance indications

S = Σ_{i = 1}^{n} (w_{i} * h_{i} / s_{i});

30. a kind of schedule virtual resources systems, it is characterised in that the system includes：Virtualized infrastructure manager VNFM With virtualization network function VNF, wherein,

The VNFM is connected with the VNF, and the VNFM is included such as any one of claim 16 to 29 Described schedule virtual resources device.

31. schedule virtual resources systems according to claim 30, it is characterised in that the system also includes：Virtually Change infrastructure manager VIM, wherein,

The VIM is connected with the VNFM, and the VIM is used for gathering and reporting the systematic function of the VNF Data.