CN106533723B

CN106533723B - Virtual resource scheduling method, device and system

Info

Publication number: CN106533723B
Application number: CN201510580104.9A
Authority: CN
Inventors: 于跃波; 蒋天超; 欧阳新志; 左奇; 楚俊生
Original assignee: Nanjing ZTE New Software Co Ltd
Current assignee: Nanjing ZTE New Software Co Ltd
Priority date: 2015-09-11
Filing date: 2015-09-11
Publication date: 2020-10-23
Anticipated expiration: 2035-09-11
Also published as: CN106533723A; WO2017041556A1

Abstract

The invention provides a virtual resource scheduling method, a virtual resource scheduling device and a virtual resource scheduling system. Wherein, the method comprises the following steps: obtaining performance data related to a Virtualized Network Function (VNF); determining a scheduling policy of virtual resources of the VNF according to the performance data; and executing scheduling operation on the virtual resource according to the scheduling strategy. The invention solves the problem that the performance data is incomplete when the VNF is dynamically stretched, can stretch resources according to the service index data and the system performance data provided by the VNF, and ensures the performance and the stability of a VNF service system.

Description

Virtual resource scheduling method, device and system

Technical Field

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

Background

At present, in the Information Technology (IT) departments of communication industries, governments and enterprises and public institutions, along with the improvement of informatization degree, the scale of computing resources is more and more huge, various business applications are directly deployed on hardware hosts, the workload is more and more large, and the management is more and more difficult. Due to the cloud computing, various applications can be uniformly deployed on the cloud, so that the difference of various bottom-layer hardware is shielded, the management is simpler and simpler, and the virtualization management system can dynamically control the virtual resources according to the running condition of the service application. Especially in the telecommunication industry, the cost of special network equipment is high, the maintenance is complex, the energy consumption is high, the service function provided by the equipment is single, and by means of the power of cloud computing, the flexible configuration of resources can be realized, and the cost is reduced.

The existing dynamic scheduling method generally controls the expansion of virtual resources according to the indexes of an operating system, such as the indexes of a CPU, a memory, a disk I/O, a network bandwidth or a system access amount and the like of the operating system, and has the disadvantages that the processing capacity of the operating system is only reflected indirectly through the indexes of the CPU, the memory, the access amount and the like, the data are independently used as the basis of resource scheduling, and the performance of a service system on the operating system can only be passively sensed; however, although the processing power of the operating system can reflect the performance of the business system to some extent, this degree of correlation is not absolute. For example, in the case that the disk resources allocated to the service system by the operating system are few, the service system may already have insufficient disk resources, but the disk resources of the operating system are not yet insufficient. Therefore, the resource scheduling scheme in the related art cannot guarantee the performance of the service system.

In the aspect of dynamic expansion and contraction of virtual resources, a virtual resource management system in the related art is not interacted with a service system; although the related technology solves the requirement of the system for the expansion and contraction of resources to a certain extent, for a system with complex service logic, too many intervention operations cannot be performed on the service units, and when the service units have a logical relationship, the coordinated operation among the service units cannot be satisfied.

In the telecommunication industry, according to the Network Function Virtualization (NFV) standard proposed by the European Telecommunications Standards Institute (ETSI), the industry has implemented the Virtualization operation of resources such as telecommunication Network elements by means of Virtualization technologies such as basic resource Virtualization and Software Defined Networking (SDN). The NFV architecture can be used as a general cloud computing virtual resource management architecture, and all virtual machine-based management models can be mapped to the NFV architecture.

Under the NFV architecture, system performance data can be acquired through a Virtualized Infrastructure Manager (VIM), and a performance reporting interface is called by the NFV to a Virtualized Network Function Manager (VNFM), and the VNFM performs service scheduling according to the collected performance data of the Virtualized Network Function (VNF).

The inventor finds that, as the data reported to the VNFM by the VIM can only obtain performance indexes of an operating system level, such as CPU, memory usage of partial systems, and disk I/O, resource performance index information of the VNF application system level cannot be obtained, such as disk usage, memory usage of all systems, bandwidth usage, traffic volume of a corresponding network element of the VNF, call completing rate, and other special usage rates; then, if the performance data obtained by the VIM is used as the VNF scheduling decision, there is also a problem that the performance of the business system cannot be guaranteed.

Aiming at the problem that the performance of a service system cannot be guaranteed by performing virtual resource scheduling of a VNF according to performance indexes of an operating system level in the related art, an effective solution is not provided at present.

Disclosure of Invention

The invention provides a virtual resource scheduling method, a virtual resource scheduling device and a virtual resource scheduling system, which are used for at least solving the problem that the performance of a service system cannot be guaranteed by performing virtual resource scheduling of a VNF according to data of an operating system level.

According to an aspect of the present invention, a virtual resource scheduling method is provided, including: obtaining performance data related to a virtualized network function, VNF; determining a scheduling policy of virtual resources of the VNF according to the performance data; and executing scheduling operation on the virtual resource according to the scheduling strategy.

Optionally, the performance data comprises at least one of: system performance data, service index data; wherein the system performance data is quantitative data for measuring performance of virtual resources deploying the VNF; the service index data is quantitative data for measuring service characteristics of a service system of the VNF.

Optionally, the system performance data comprises: system performance data collected and reported by the VNF; the system performance data collected and reported by the virtualized infrastructure manager VIM.

Optionally, determining the scheduling policy of the virtual resource of the VNF according to the performance data includes: summarizing the system performance data collected and reported by the VNF and the system performance data collected and reported by the VIM; and determining the scheduling strategy of the virtual resources of the VNF according to the service index data and the summarized system performance data.

Optionally, the aggregating the system performance data collected and reported by the VNF and the system performance data collected and reported by the VIM includes: for system performance data that is collectable from both the VNF and from the VIM, subject to system performance data that is collected from either the VNF or from the VIM; or summarizing the system performance data collected from both the VNF and the VIM according to a weighted average algorithm.

Optionally, the service indicator data includes at least one of: traffic throughput per unit time, and/or other performance data related to traffic characteristics; the system performance data includes at least one of: CPU utilization, memory utilization, disk write/read performance index, remaining disk space, network bandwidth utilization, network load, network card packet loss.

Optionally, before obtaining the performance data related to the VNF, the method further comprises: and sending subscription information to the VNF, wherein the subscription information is used for subscribing one or more of the performance data and setting a reporting period of the performance data.

Optionally, the subscription information is further configured to set a trigger threshold and a duration for reporting the performance data.

Optionally, after sending the subscription information to the VNF, the method further includes: the VNF receiving the subscription information; and reporting the performance data by the VNF under the condition that the duration time of the performance data exceeding the trigger threshold exceeds the duration time.

Optionally, a reporting period of reporting the performance data by the VNF is the same as a reporting period of reporting system performance data by the virtualized infrastructure manager VIM.

Optionally, the performing a scheduling operation on the virtual resource includes: sending a pre-scheduling notification to the VNF, wherein the pre-scheduling notification is used for indicating that a scheduling operation is to be performed on the virtual resource; and executing scheduling operation on the virtual resource under the condition of receiving a pre-scheduling confirmation response, wherein the pre-scheduling confirmation response is used for indicating that the VNF has completed business processing related to the scheduling operation of the virtual resource and allowing a virtualization system to execute resource scaling operation.

Optionally, after performing the scheduling operation on the virtual resource, the method further includes: sending a scheduling completion confirmation notification to the VNF, wherein the scheduling completion confirmation notification is used for indicating that the scheduling operation on the virtual resource is completed.

Optionally, the scheduling operation includes: in the case of adding a virtual resource or changing a virtual resource, before performing a scheduling operation on the virtual resource, the method further includes: receiving physical machine system performance data of a physical machine reported by a Virtual Infrastructure Manager (VIM); and selecting the first physical machine of which the system performance data of the physical machine meets the scheduling operation requirement as a host machine of the newly added virtual resource or the changed virtual resource.

Optionally, determining the scheduling policy of the virtual resource of the VNF according to the performance data includes: calculating a performance index of the VNF according to the performance data; and inquiring a preset scheduling strategy corresponding to the performance index, wherein the preset scheduling strategy is used for indicating scheduling operation to be executed under the condition of the performance index.

Optionally, the performance indicator is calculated by the following formula: the performance index

Wherein, w_iA preset index weight, h, representing the ith performance data of the VNF_iA value, s, representing the ith capability data of the VNF_iA value representing the ith capability data when the VNF is fully operational.

According to another aspect of the present invention, there is also provided a virtual resource scheduling apparatus, including: an acquisition module for acquiring performance data related to a virtualized network function, VNF; a determining module configured to determine a scheduling policy of a virtual resource of the VNF according to the performance data; and the execution module is used for executing scheduling operation on the virtual resource according to the scheduling strategy.

Optionally, the determining module includes: a collecting unit, configured to collect and report the system performance data collected and reported by the VNF and the system performance data collected and reported by the VIM; a determining unit, configured to determine the scheduling policy of the virtual resource of the VNF according to the service index data and the summarized system performance data.

Optionally, the summarizing unit is configured to: for system performance data that is collectable from both the VNF and from the VIM, subject to system performance data that is collected from either the VNF or from the VIM; or summarizing the system performance data collected from both the VNF and the VIM according to a weighted average algorithm.

Optionally, the apparatus further comprises: a first sending module, configured to send subscription information to the VNF, where the subscription information is used to subscribe to one or more of the performance data, and set a reporting period of the performance data.

Optionally, the execution module includes: a sending unit, configured to send a pre-scheduling notification to the VNF, where the pre-scheduling notification is used to indicate that a scheduling operation is to be performed on the virtual resource; an execution unit, configured to execute a scheduling operation on the virtual resource when a pre-scheduling confirmation response is received, where the pre-scheduling confirmation response is used to indicate that the VNF has completed a service process related to the scheduling operation of the virtual resource, and allow a virtualization system to perform a resource scaling operation.

Optionally, the apparatus further comprises: a second sending module, configured to send a scheduling completion acknowledgement notification to the VNF, where the scheduling completion acknowledgement notification is used to indicate that scheduling operations on the virtual resource have been completed.

Optionally, the apparatus further comprises: a receiving module, configured to receive physical machine system performance data of a physical machine, which is reported by a virtualized infrastructure manager VIM, when the scheduling operation includes increasing a virtual resource or changing a virtual resource; and the selection module is used for selecting the first physical machine of which the system performance data of the physical machine meets the scheduling operation requirement as a host machine of the newly added virtual resource or the changed virtual resource.

Optionally, the determining module includes: a calculating unit, configured to calculate a performance index of the VNF according to the performance data; and the query unit is used for querying a preset scheduling strategy corresponding to the performance index, wherein the preset scheduling strategy is used for indicating scheduling operation to be executed under the condition of the performance index.

Optionally, the calculation unit calculates the performance index by the following formula: the performance index

According to another aspect of the present invention, there is also provided a virtual resource scheduling system, including: a virtualized infrastructure manager VNFM and a virtualized network function VNF, wherein the VNFM is connected with the VNF, and the VNFM includes the above-mentioned virtual resource scheduling apparatus.

Optionally, the system further comprises: a Virtualized Infrastructure Manager (VIM), wherein the VIM is connected to the VNFM, and the VIM is configured to collect and report system performance data of the VNF.

By the invention, performance data related to a Virtualized Network Function (VNF) is acquired; determining a scheduling policy of virtual resources of the VNF according to the performance data; according to the scheduling strategy, the scheduling operation is executed on the virtual resources, so that the problem that the performance of the service system cannot be guaranteed by performing virtual resource scheduling of the VNF according to the performance indexes of the operating system level is solved, and the performance and the stability of the service system are guaranteed.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limiting the invention. In the drawings:

FIG. 1 is a flowchart of a virtual resource scheduling method according to an embodiment of the present invention;

fig. 2 is a block diagram of a virtual resource scheduling apparatus according to an embodiment of the present invention;

fig. 3 is a first schematic structural diagram of a virtual resource scheduling apparatus according to an embodiment of the present invention;

fig. 4 is a schematic diagram of a preferred structure of a virtual resource scheduling apparatus according to an embodiment of the present invention;

fig. 5 is a schematic diagram of a preferred structure of a virtual resource scheduling apparatus according to an embodiment of the present invention;

fig. 6 is a schematic diagram of a preferred structure of a virtual resource scheduling apparatus according to an embodiment of the present invention;

fig. 7 is a schematic diagram of a preferred structure of a virtual resource scheduling apparatus according to an embodiment of the present invention;

fig. 8 is a schematic diagram six of a preferred structure of a virtual resource scheduling apparatus according to an embodiment of the present invention;

FIG. 9 is a schematic structural diagram of an NFV framework in accordance with a preferred embodiment of the present invention;

FIG. 10 is a flowchart of a virtual resource scheduling method in accordance with a preferred embodiment of the present invention;

fig. 11 is a flowchart of a virtual resource scheduling method in a generic virtualized cloud management system according to a preferred embodiment of the present invention.

Detailed Description

The invention will be described in detail hereinafter with reference to the accompanying drawings in conjunction with embodiments. It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.

It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order.

In this embodiment, a virtual resource scheduling method is provided, and fig. 1 is a flowchart of a virtual resource scheduling method according to an embodiment of the present invention, as shown in fig. 1, the flowchart includes the following steps:

step S102, acquiring performance data related to a Virtualized Network Function (VNF);

step S104, determining a scheduling strategy of the virtual resource of the VNF according to the performance data;

and step S106, executing scheduling operation on the virtual resource according to the scheduling strategy.

Through the steps, a scheduling policy is determined according to performance data related to the VNF, and then scheduling operation on the virtual resources of the VNF is executed according to the scheduling policy; for example, the performance data may be system performance data or service index data, the system performance data may be directly obtained from the VNF or obtained from the VIM, and the performance data related to the VNF may reflect the processing capability of the service system, so that the problem that the performance of the service system cannot be guaranteed by performing virtual resource scheduling of the VNF according to a single type of data by the VNF is solved, and the performance and stability of the service system are guaranteed.

Alternatively, the above steps may be performed by VNFM or NFVO, or some and another operations may be performed on VNFM and NFVO, respectively. In addition, it is also conceivable that the above steps are performed by other management entities in the NFV architecture.

Optionally, the performance data includes, but is not limited to, at least one of: system performance data, service index data; wherein the system performance data is quantitative data for measuring performance of virtual resources deploying the VNF; the traffic index data is quantized data for measuring traffic characteristics of the traffic system of the VNF.

Since the system performance data may be collected from the VNF or the VIM, in step S104, the system performance data collected and reported by the VNF and the system performance data collected and reported by the VIM may be summarized; and determining a scheduling strategy of the virtual resources of the VNF according to the service index data and the summarized system performance data.

In addition, considering that the system performance data collected from the VNF and the system performance data collected from the VIM may be the same type of system performance data, at least the following two ways may be adopted when performing the system performance data aggregation:

mode 1, for system performance data that can be collected from both VNF and VIM, system performance data collected from VNF or VIM is the standard;

in the mode 2, the system performance data collected from both the VNF and the VIM are summarized according to a weighted average algorithm.

The system performance data and the service index data are used to represent service processing performance of the service system of the VNF, and there are many parameters that can represent the service processing performance, and the parameters of the concerned service processing performance may be different according to the service type of the VNF. For example, the traffic indicator data includes, but is not limited to, at least one of: traffic throughput per unit time, and/or other performance data related to traffic characteristics; system performance data includes, but is not limited to, at least one of: CPU utilization, memory utilization, disk write/read performance index, remaining disk space, network bandwidth utilization, network load, network card packet loss. These service index data are quantized data.

Optionally, in step S102, the service index data and the system performance data acquired from the VNF are reported periodically or aperiodically by the VNF; although the VIM can also report the system performance data, the reporting period VNF of the VIM cannot be controlled; reporting the service index data and the system performance data through the VNF can achieve simultaneous reporting of the service index data and the system performance data, for example, reporting the service index data and the system performance data simultaneously in the same reporting period of the VNF can achieve simultaneous reporting of the service index data and the system performance data, thereby avoiding a problem of scheduling policy selection deviation caused by asynchronization of the service index data acquired by the VNFM or NFVO through the VNF and the system performance data acquired by the VIM.

Optionally, the reporting period of the system performance data is the same as the reporting period of the service index data, or the reporting period of the service index data is an integral multiple of the reporting period of the system performance data.

In this way, before acquiring the system performance data and the service index data of the VNF, subscription information may be sent to the VNF, where the subscription information is used to subscribe to one or more of the performance data, and a reporting period of the performance data is set. By the method, the personalized subscription can be performed on the performance data needing attention.

Optionally, the subscription information is further used to set a trigger threshold and a duration for reporting the performance data.

Optionally, after setting a trigger threshold and a duration for reporting the performance data by the VNF, optionally, after sending the subscription information to the VNF, the VNF receives the subscription information; and reporting the performance data by the VNF under the condition that the duration of the performance data exceeding the trigger threshold exceeds the duration. By the mode, a large amount of occupation of system resources caused by real-time reporting of the performance data is avoided.

Optionally, when the scheduling operation of the virtual resource is performed in step S106, a pre-scheduling notification may be sent to the VNF, where the pre-scheduling notification is used to indicate that the scheduling operation is to be performed on the virtual resource; and executing scheduling operation on the virtual resource under the condition of receiving a pre-scheduling confirmation response, wherein the pre-scheduling confirmation response is used for indicating that the VNF has completed service processing related to the scheduling operation of the virtual resource and allowing the virtualization system to execute resource scaling operation. By the above method, a notification mode before scheduling is provided, so that normal execution of services on the VNF can be guaranteed, and service interruption or exception caused by virtual resource scheduling is avoided.

Optionally, after the scheduling operation is performed on the virtual resource, a scheduling completion confirmation notification may also be sent to the VNF, where the scheduling completion confirmation notification is used to indicate that the scheduling operation on the virtual resource has been completed. By the above manner, the VNF can sense the completion of virtual resource scheduling, and then perform subsequent processing, for example, the VNF starts to execute a service operation by using the newly allocated virtual resource after receiving the scheduling completion confirmation notification.

Optionally, the performance index of the physical machine may be reported by the VIM, and under the condition that some virtual resources are added (for example, operations such as powering on the virtual machine, adding a new virtual machine, and the like) or virtual resources are changed (for example, virtual machine migration), the selection problem of the physical machine needs to be considered. Generally, the physical machine is selected to provide at least enough virtual resources for powering on the virtual machine, adding a new virtual machine, or migrating a virtual machine. For example, before step S106, the VNFM may receive physical machine system performance data of the physical machine reported by the virtualization infrastructure manager VIM; and the VNFM selects a first physical machine of which the system performance data of the physical machine meets the scheduling operation requirement as a host machine of the newly added virtual resource or the changed virtual resource. In addition, a physical machine with a low resource usage rate may be selected as a host for a new virtual resource or a changed virtual resource in the scheduling operation.

Optionally, in step S104, when determining the scheduling policy, it is a general practice to calculate a weight value of the performance data in a weighting manner, and then select the scheduling policy corresponding to the weight value according to the calculated weight value. The weight of the performance data may be set according to needs, for example, in a case where the service performance of a certain VNF is more concerned, the weight of the service index data may be set to be larger.

Optionally, in step S104, the service performance index of the VNF may be calculated according to the service index data and the system performance data; and inquiring a preset scheduling strategy corresponding to the service performance index, wherein the preset scheduling strategy is used for indicating scheduling operation to be executed under the condition of the service performance index. These scheduling operations include: the virtual machine is powered on and powered off, the CPU is stretched, the memory is stretched, the disk is expanded, the virtual machine is newly built, the virtual machine is deleted, the virtual machine is migrated and the like. The preset scheduling policy may be configured according to the service requirements, respectively.

Alternatively, the service performance indicator may be calculated by the following formula: performance index of service

Wherein, w_iPreset index weight, h, representing the ith index data of the VNF_iValue s representing ith index data of VNF_iA value representing the ith index data at VNF full load operation; the index data includes: service index data and system performance data.

Through the above description of the embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but the former is a better implementation mode in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (such as a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.

In this embodiment, a virtual resource scheduling apparatus is further provided, and the apparatus is used to implement the foregoing embodiments and preferred embodiments, and details of which have been already described are omitted. As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. Although the means described in the embodiments below are preferably implemented in software, an implementation in hardware, or a combination of software and hardware is also possible and contemplated.

Fig. 2 is a block diagram of a virtual resource scheduling apparatus according to an embodiment of the present invention, and as shown in fig. 2, the apparatus includes: an obtaining module 22, a determining module 24 and an executing module 26, where the obtaining module 22 is configured to obtain performance data related to a virtualized network function VNF; a determining module 24, coupled to the obtaining module 22, configured to determine a scheduling policy of the virtual resource of the VNF according to the performance data; an executing module 26, coupled to the determining module 24, is configured to execute a scheduling operation on the virtual resource according to the scheduling policy.

Optionally, the performance data comprises at least one of: system performance data, service index data; wherein the system performance data is quantitative data for measuring performance of virtual resources deploying the VNF; the traffic index data is quantized data for measuring traffic characteristics of the traffic system of the VNF.

Fig. 3 is a block diagram of an optional structure of a virtual resource scheduling apparatus according to an embodiment of the present invention, as shown in fig. 3, optionally, the determining module 24 includes: a summarizing unit 242, configured to summarize the system performance data collected and reported by the VNF and the system performance data collected and reported by the VIM; a determining unit 244, coupled to the summarizing unit 242, configured to determine a scheduling policy of the virtual resource of the VNF according to the service index data and the summarized system performance data.

Optionally, the summarizing unit is configured to: for system performance data that can be collected from both the VNF and the VIM, the system performance data collected from either the VNF or the VIM is subject to; or the system performance data collected from both the VNF and the VIM are summarized according to a weighted average algorithm.

Optionally, the service indicator data comprises at least one of: traffic throughput per unit time, and/or other performance data related to traffic characteristics; the system performance data includes at least one of: CPU utilization, memory utilization, disk write/read performance index, remaining disk space, network bandwidth utilization, network load, network card packet loss.

Fig. 4 is a block diagram of an optional structure of a virtual resource scheduling apparatus according to an embodiment of the present invention, as shown in fig. 4, optionally, the apparatus further includes: a first sending module 42, coupled to the VNF, configured to send subscription information to the VNF, where the subscription information is used to subscribe to one or more of the performance data, and set a reporting period of the performance data.

Optionally, a reporting period of the performance data reported by the VNF is the same as a reporting period of the performance data reported by the virtualized infrastructure manager VIM.

Fig. 5 is a block diagram of an optional structure of a virtual resource scheduling apparatus according to an embodiment of the present invention, and as shown in fig. 5, optionally, the execution module 26 includes: a sending unit 262, coupled to the VNF, configured to send a pre-scheduling notification to the VNF, where the pre-scheduling notification is used to indicate that a scheduling operation is to be performed on a virtual resource; an executing unit 264, coupled to the VNF, configured to execute the scheduling operation on the virtual resource if a pre-scheduling confirmation response is received, where the pre-scheduling confirmation response is used to indicate that the VNF has completed the service processing related to the scheduling operation of the virtual resource, and allow the virtualization system to perform the resource scaling operation.

Fig. 6 is a block diagram of an optional structure of a virtual resource scheduling apparatus according to an embodiment of the present invention, as shown in fig. 6, optionally, the apparatus further includes: a second sending module 62, coupled to the execution module 26 and the VNF, configured to send a scheduling completion acknowledgement notification to the VNF, where the scheduling completion acknowledgement notification is used to indicate that the scheduling operation on the virtual resource is completed.

Fig. 7 is a block diagram of an optional structure of a virtual resource scheduling apparatus according to an embodiment of the present invention, as shown in fig. 7, optionally, the apparatus further includes: a receiving module 72, coupled to the VIM, configured to receive, under a condition that the scheduling operation includes adding a virtual resource or changing a virtual resource, physical machine system performance data of the physical machine, which is reported by the virtualized infrastructure manager VIM; a selecting module 74, coupled to the receiving module 72 and the executing module 26, is configured to select the first physical machine whose physical machine system performance data meets the scheduling operation requirement as a host for the new virtual resource or the changed virtual resource.

Fig. 8 is a block diagram six of an optional structure of a virtual resource scheduling apparatus according to an embodiment of the present invention, as shown in fig. 8, optionally, the determining module 24 includes: a calculating unit 246, configured to calculate a performance index of the VNF according to the performance data; the querying unit 248, coupled to the calculating unit 246, is configured to query a preset scheduling policy corresponding to the performance indicator, where the preset scheduling policy is used to indicate a scheduling operation that should be performed in case of the performance indicator.

Alternatively, the calculation unit calculates the performance index by the following formula: performance index

Wherein, w_iPreset index weight, h, representing ith performance data of VNF_iValue, s, representing the ith capability data of the VNF_iA value representing the ith capability data when the VNF is running at full load.

It should be noted that, the above modules may be implemented by software or hardware, and for the latter, the following may be implemented, but not limited to: the modules are all positioned in the same processor; alternatively, the modules are respectively located in a plurality of processors.

The embodiment also provides a virtual resource scheduling system, which is used for implementing the virtual resource scheduling method. The system comprises: a virtualized infrastructure manager (VNFM) and a Virtualized Network Function (VNF), wherein the VNFM is connected to the VNF, and the VNFM includes the above-mentioned virtual resource scheduling apparatus.

Optionally, the system further comprises: and a Virtualized Infrastructure Manager (VIM), wherein the VIM is connected with the VNFM, and the VIM is used for collecting and reporting system performance data of the VNF.

The embodiment of the present invention also provides software for executing the technical solutions described in the above embodiments and preferred embodiments.

The embodiment of the invention also provides a storage medium. In the present embodiment, the storage medium described above may be configured to store program code for performing the steps of:

Optionally, in this embodiment, the storage medium may include, but is not limited to: various media capable of storing program codes, such as a usb disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic disk, or an optical disk.

Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments and optional implementation manners, and this embodiment is not described herein again.

In order that the description of the embodiments of the invention will be more apparent, reference is now made to the preferred embodiments for illustration.

The preferred embodiment of the invention provides a resource scheduling method for reporting system performance and service index data by a VNF (virtual network function) and performing dynamic expansion and contraction of virtual resources; the method comprises the steps of integrating system performance data and service index data and determining a scheduling strategy of virtual resources of a VNF; according to the scheduling strategy, the method for executing the scheduling operation on the virtual resource solves the problem that the performance of the service system cannot be guaranteed by performing virtual resource scheduling of the VNF solely according to the performance index of the operating system level or solely through the service performance index, and guarantees the performance of the service system.

In the NFV architecture, the VIM may report performance data (CPU, memory, disk I/O) of the VNF bottom operating system level to the VNFM, but some application layer performance data cannot be obtained from the VIM, such as disk remaining space, network throughput, network latency, network usage rate, and VNF service index data itself; the reporting period of the VNF performance index data is dynamically set by the VNF, and the reporting period of the VIM is generally a uniformly determined time period. In the preferred embodiment of the present invention, the VNF may adjust the reporting period of the performance index data as needed, and the performance data for scheduling decision service and the system performance data may be reported through the VNF, so that the data is guaranteed to be uniform in the time dimension, and the configuration of service scheduling is more flexible. In the preferred embodiment, performance data of the physical machine reported by the VIM may also be referred to in the scheduling, which is beneficial to selecting a more reasonable host machine to operate the VNF during scaling.

The preferred embodiment of the present invention is described and illustrated with respect to the NFV framework as an example. Fig. 9 is a schematic structural view of an NFV framework according to a preferred embodiment of the present invention, as shown in fig. 9, the NFV framework includes:

a Network function virtualization infrastructure (NFVI, which includes software and hardware (hardware, HW) environments required for deploying VNF applications), VIM, VNFM, and VNF, and may further include: a Network function virtualization coordinator (NFVO for short); the NFVO may perform scheduling operations instead of VNFM.

In a preferred embodiment of the invention:

the VIM includes the following functions:

1. virtual resources may be managed.

2. The scaling action of the VNF may be performed.

3. Collecting system performance data of a host machine and a client machine;

the VNF includes the following functions:

1. and configuring the statistical period and the reporting period of the corresponding system performance data and the service index data.

2. In a preset statistical period, according to the characteristics of the VNF, writing a corresponding method to generate service index data and system performance data of the cost period, where the system performance data and the service index data include, but are not limited to: the CPU/memory/disk occupancy rate can also consider the disk I/O and the network bandwidth utilization rate and the network throughput as well as the special performance index of the special VNF network element.

3. The VNF may be processed according to dynamic scaling instructions of the VNFM or NVFO. Such as: when one virtual machine is closed, the service of the virtual machine can be ended, and the VNFM transfers the service of the virtual machine to other virtual machines which are not closed for processing or starts a new VNF at other nodes to continue processing the service, thereby ensuring the smoothness of the service.

4. The minimum resource requirement of the service system is guaranteed to be met, and when a certain virtual machine is closed by an instruction transmitted by the service scheduling center, but the number of resources owned by the service system is lower than the minimum resource number after the virtual machine is closed, the virtual machine can be refused to be closed.

5. And processing the service according to the dynamic expansion and contraction result of the virtual resource transmitted by the VNFM or NFVO. Such as: when a new virtual machine is expanded, the new virtual machine is started, and the service system acquires the resource expansion operation result transmitted from the VFNM or NFVO, the service system can initialize the service of the virtual machine, so that the virtual machine can process the service together with other virtual machines.

VNFM or NFVO includes the following functions:

the VNFM can store system performance data and service index data reported by a service system.

And 2, receiving the service index data and the system performance data reported by the VNF by the VNFM.

VNFM or NFVO can create a policy enforcement mechanism that scales dynamically according to system performance data and business metric data.

4. Dynamic scaling instructions may be passed.

5. The dynamic scaling result of the virtual resource can be determined.

6. And system performance data of the host and the client reported by the VIM can be received.

Based on the NFV framework, the scheduling method for performing dynamic scaling according to VNF reported data according to the preferred embodiment of the present invention includes the following steps:

step one, a service data generation module of each service unit of the service system generates service index data in real time according to a service index generation method, and the data reflect the service processing capacity of the service system. Meanwhile, the capability of generating system performance data is provided, and the data is reported to the VNFM.

And step two, a service index data acquisition module of the service system collects the service index data of each service unit to generate reported data, and transmits the reported data to the VNFM.

And step three, the VNFM or the NFVO judges whether a scheduling strategy configured in the system meets a trigger condition, and if the scheduling strategy meets the trigger condition, a notification confirmation flow before operation of the VNF is called according to an interface. The scheduling strategy comprises set service index equivalent and system performance data index, the system performance data index mainly depends on the CPU/memory/disk occupancy rate, and the disk I/O, the network bandwidth utilization rate and the network throughput are considered.

And step four, after the VNF receives the notification before the telescopic operation, preprocessing before the telescopic operation is carried out in the VNF.

And step five, after the VNF finishes the pretreatment before the telescopic operation, the VNF and the VNFM or NFVO are confirmed before the operation.

And step six, after receiving the confirmation, the NFVO or the VNFM informs the VIM of performing resource scaling operation.

And step seven, after the VIM finishes the resource expansion operation, the VNFM or NFVO informs the service system of finishing the expansion.

Step eight, after receiving the notification that the resource expansion and contraction processing is finished, the VNF stores the resource change condition after the expansion and contraction into the service system.

In the preferred embodiment of the present invention, the service unit actively generates the system performance data and service index data which need to be concerned for the condition of resource dynamic scaling trigger. And the performance indexes of the system strategy comprehensive service and the system are configured to comprehensively judge the scaling, and the data such as disk I/O, network throughput, utilization rate and the like are reported on the VNF to perform early warning scaling of the system, so that the stability and reliability of each latitude of the system are ensured.

In addition, in the preferred embodiment of the present invention, a method for notifying and confirming before and after the virtual resource is extended is further provided, so that the service system can more flexibly control the influence of the extension of the virtual resource on the system, and the system can operate more stably after the resource is extended.

By adopting the scheme provided by the preferred embodiment of the invention, the defect that the virtual resource is dynamically adjusted according to indirect indexes such as a CPU (central processing unit), a memory and the like in the prior art can be overcome, and the close relation between the service and the resource can be reflected, so that the processing capacity of the VNF can be truly reflected. Therefore, compared with the prior art, the preferred embodiment perfects that the business system controls the business processing capacity of the business system, achieves the effect of dynamically stretching resources, saves manpower, and improves the capacity of the business system for stably processing the business.

The preferred embodiments of the present invention are described and illustrated below in two examples.

Preferred embodiment 1

Fig. 10 is a flowchart of a virtual resource scheduling method according to a preferred embodiment of the present invention, as shown in fig. 10, the flowchart includes the following steps:

step S1001, sets an acquisition time interval of the VNF service data acquisition module.

Step S1002, set a time interval for reporting VNF service index data.

Here, the scheduling policy of the VNFM may also be set, and the subsequent scaling may be used as a decision setting. The policy setting is to control according to the equivalent percentage calculated by the service index, and also consider the system performance data of the VNF itself, and in order to ensure reliable operation of the VNF, the system disk I/O detection judgment and the network throughput judgment may be set to ensure reliability of the VNF hardware.

Step S1003, deploying a service index data acquisition module in each VNF, and acquiring system performance data and service index raw data of the VNF according to a service requirement of the VNF. These system performance data and metric data may include, but are not limited to: the service processing capacity in unit time, the utilization rate of a service system CPU and a memory, disk I/O, the residual size of disk space, network bandwidth, network load, packet loss rate, special performance indexes of a special VNF and the like.

Step S1004, the VNF acquires the performance index of the service data and the performance data of the operating system level according to the set sampling period, keeps uniform acquisition granularity, and reports the performance data to the VNFM. The VNFM needs to provide a service performance data reporting interface and a VNF system performance data interface, and the VNF system performance data includes, but is not limited to, data related to the service system: CPU, memory, disk I/O, disk remaining space, bandwidth throughput, network utilization, and the like. And finally, acquiring and reporting the data by combining the self characteristic requirements of the service. The VNFM is to provide a common measurement reporting interface. In particular, some application-level data that cannot be obtained from the VIM, such as disk remaining space, network utilization, etc., may be used.

Step S1005, the VNFM summarizes the service index raw data of each VNF node, and converts the data into service index data required for scheduling according to an algorithm. The calculation method of the service index data can be in various manners, and one of the following manners is listed:

during a business data collection cycle, based on: (1) the number n of VNF service data index items, (2) the service raw data h of the VNF node collected in step S1003, (3) the weight w of each service index raw value, (4) the value S of each index item of the VNF during full-load operation, and a calculation is performed to obtain a service index data, where the calculation formula may be the following formula:

step S1006, the VIM calls and the interface period of the VNFM reports the host system capability data to the VNFM as a scheduling resource reference. And performing scaling decision on the subsequent VNFM, for example, selecting a target machine with low utilization rate of host machine resources in the scaling process, and performing decision making according to a strategy of the VNFM.

And step S1007, the VNFM judges the scheduling strategy according to the reported service index data. The strategy for triggering the virtual resource scaling operation configured in the scheduling strategy VNFM is used for scheduling the resource scaling operation according to the values of various service index data.

The scaling operation may be creation, deletion, migration, suspension, wakeup, etc. of the VNF node, and may be configured as needed.

If the system configures two service scheduling strategies:

strategy 1, when the traffic processing capacity is more than 60% of the system processing capacity, the VNF1-2 is increased.

Policy 2, VNF1-2 will be deleted when the traffic throughput is less than 40% of the system processing capacity.

Thus, when the reported service index data is greater than 60%, for example, the reported value is 61%, the resource scheduling module will trigger the policy of the newly added VNF 1-2.

When the reported value is less than 40%, a policy to delete the power of VNF1-2 is triggered.

Meanwhile, the system can also be configured with a service performance scheduling strategy, such as that the CPU, the memory and the disk I/O are migrated according to service characteristics in a certain interval. The policy needs to be set up integrally with the service scheduling policy, because different services may have different requirements on the CPU, the memory, and the disk I/O, some services are, for example, message middleware, have requirements on the use of the memory, some computing services have requirements on the CPU, and some file storage services have high requirements on the memory and the disk I/O. The setting strategy takes the service scenario into consideration.

And (3) carrying out system health judgment on the conditions of the CPU, the disk I/O, the network throughput and the network utilization rate, and if the conditions are not in expectation, migrating the VNF to other target hosts for operation, thereby ensuring normal and reliable service.

Step S1008, the VNFM calls a notification interface before the service scaling operation of the VNF, and sends a notification before the scaling operation.

Step S1009, the VNF performs operations such as policy preprocessing or service switching inside the VNF according to the received notification before scaling, so as to ensure that all functions of the original VNF are not affected during scaling execution.

For example, when the policy is to delete the VNF1-2, the VNFM transfers the traffic on the VNF1-2 to the VNF1-1 for processing, and determines whether the VNF1-2 can be deleted, and if so, configures the relevant traffic information of the VNF1-2 in the system as suspension processing. If the service currently processed by the VNF1-2 is not finished, the scaling operation is finished, the next scaling operation is triggered, and when the step is executed, the processing is judged again.

In step S1010, the VNF calls a pre-scaling operation confirmation interface provided by the VNFM to confirm that the scaling operation can be performed by the VNFM.

In step S1011, after obtaining the pre-scaling confirmation in step S1010, the VNFM calls the VIM to perform the resource scaling operation.

In step S1012, the VIM performs a scaling operation. VNF1-2 needs to be deleted as in step S1009 is performed, at which point VNF1-2 is deleted.

In step S1013, after the resource scaling operation is finished, the VNFM calls the resource scaling operation of the relevant VNF node and notifies the interface, so as to notify that the scaling operation of the VNF has been finished.

In step S1014, the relevant VNF processes the relevant flow as needed after receiving the notification.

For example, if the VNF1-2 is added in step S1009, step S1013 notifies other related VNF nodes, and the related VNF processes the information of the newly added VNF1-2 in this step and subsequently forwards the related data to the VNF1-2, so that data offloading is performed. If step S1009 is to delete VNF1-2, the other VNFs will process the information of the deleted VNF1-2 to avoid sending data to it after receiving the notification.

It should be noted that the above process is performed continuously, the system periodically detects the trigger condition of the scheduling operation, and once the reported system performance data and service index data meet the trigger condition of the scheduling policy, the scaling operation of policy configuration is executed.

Preferred embodiment 2

The following describes preferred embodiments of the present invention with respect to general virtualized cloud management.

Fig. 11 is a flowchart of a virtual resource scheduling method in a virtualized cloud management system according to a preferred embodiment of the present invention, and as shown in fig. 11, the method includes the following steps:

step S1101, setting a service index data acquisition time interval in the service system.

Step S1102 is to set a time interval for reporting the service index data in the service scheduling center.

Step S1103, the service index data is generated by the service data generation module of the service node deployed on each virtual machine according to the service processing amount of the service node, and system performance data and service index raw data of the service node are generated.

These system performance data and metric data may include, but are not limited to: the service processing capacity in unit time, the utilization rate of a service system CPU and a memory, disk IO, the residual size of a disk, network bandwidth and the like.

Step S1104, the service index data collection module of the service system collects the service index raw data generated by each service node in step S1103 according to the time interval set in step S1101, and performs summarization processing on the raw data of each service node, and encapsulates the summarized raw data into service index data to be reported.

The calculation method of the service index data can be in various manners, and one of the following manners is listed:

during a business data collection cycle, based on: (1) the number n of the service nodes which normally operate, (2) the processing capacity s of a single node of the service node when the single node of the service node runs at full load, (3) the original service data h of each service node collected in the step 3, (4) the weight w of each service node (default is 1), and calculation is performed to obtain a service index data, wherein the calculation formula can be the following formula:

step S1105, the service scheduling center calls a service index data reporting interface provided by the virtual resource scheduling system, and reports the service index data generated in step S1104 to the virtual resource scheduling system.

Step S1106, the resource scheduling module of the virtual resource scheduling system determines the scheduling policy according to the reported service index data. The scheduling strategy is a strategy for triggering virtual resource scaling operation configured in the virtualized scheduling system, and performs scheduling of the resource scaling operation according to values of various service index data.

The scaling operation can be powering on and powering off the virtual machine, scaling a CPU, scaling a memory, expanding a disk, newly building the virtual machine, deleting the virtual machine, migrating the virtual machine and the like, and can be configured as required.

If the system is configured with two scheduling strategies:

and strategy 1, when the service processing capacity is more than 60 percent of the system processing capacity, the virtual machine 1 is powered on.

And strategy 2, when the service processing capacity is less than 40% of the system processing capacity, the virtual machine 1 is powered off.

Thus, when the reported service index data is greater than 60%, for example, the reported value is 61%, the resource scheduling module will trigger a policy of turning on the power supply of the virtual machine 1.

And when the reported value is less than 40%, triggering a strategy of turning off the power supply of the virtual machine 1.

It should be noted that, that is, if the service system does not report the service index data, step S1106 is also executed, which is a task executed at regular time, and the task has both data for determining service reporting and virtual machine performance data collected by the virtual resource scheduling system itself, so as to determine the execution of the scheduling policy. That is, only the service index data reported by the service may be determined, only the virtual machine performance data acquired by the virtual resource scheduling system itself may be determined, or both the data may be determined.

In addition, the relevant policy in step S1106 may be created according to the specific requirement of the business system for the virtual resource, for example:

1. a policy can be created that forces the migration of the virtual machine to be performed only under the host with affinity to the virtual machine when the virtual machine migrates. The affinity between the virtual machine and the host machine meets the special processing when the virtual machine has different I/O rates, CPU computing power, network bandwidth, storage and other requirements for different applications.

2. A policy may be created that exclusive virtual machines cannot run under the same host. Exclusivity can meet the requirement of redundant processing units of the same service, and the redundant processing units cannot be deployed on the same host at the same time.

3. And in the non-busy period such as night, the virtual machines which can not be powered off are transferred to the same host, the host is powered off by the calling interface, and in the normal period in the day, the host is powered on by the calling host power-on interface, and the virtual machines originally deployed on the hosts are transferred back, so that the purposes of saving energy and reducing consumption are achieved.

In addition, there may be many strategies, which are not listed here.

Step S1107, the virtual resource scheduling system invokes a service operation notification interface of the service scheduling center to send a notification before the scaling operation.

Step S1108, the service system processing module determines and processes the corresponding service according to the notification before expansion received by the service scheduling center.

For example, when the policy is a power-off operation of the virtual machine 1, the service system processing module transfers the service on the virtual machine 1 to other normal virtual machines for processing, and determines whether the current virtual machine 1 can process the scaling operation, and if so, configures the relevant service information of the virtual machine 1 in the system as a suspension processing. If the service currently processed by the virtual machine 1 is not finished, the scaling operation is finished, the next scaling operation is triggered, and the processing is judged when the step is executed.

Step S1109, after the service system processing module completes the service processing flow change caused by resource scaling, the service scheduling center calls the pre-scaling operation confirmation interface provided by the virtual scheduling system to confirm that the virtual resource scheduling system can execute the scaling operation.

In step S1110, after the virtual resource scheduling system obtains the pre-scaling operation confirmation in step S1109, the virtual resource scheduling system notifies the resource management module to perform the resource scaling operation.

In step S1111, the resource management module performs a scaling operation. If the virtual machine 1 needs to be powered off in step S1108, then the virtual machine 1 will be powered off.

Step S1112, after the virtual resource scheduling system finishes executing the resource scaling operation, the virtual resource scheduling system calls the resource scaling operation of the service scheduling center and then notifies the interface that the scaling operation of the virtual resource has been finished.

And step S1113, the service system processing module carries out subsequent processing on the service system according to the notification after the telescopic operation received by the service scheduling center.

For example, if the step S1108 is the power-on operation of the virtual machine 1, the success may be directly returned in step S1108, no processing is performed, and the service is processed in step S1113. In step S1113, the service system initializes the application deployed on the virtual machine 1, adds the service information of the virtual machine 1 to the service system, and allocates the service operation to the virtual machine 1 for processing after completing other configuration operations.

It should be noted that the above process is performed continuously, the system periodically detects the trigger condition of the scheduling operation, and once the reported service index data meets the trigger condition of the scheduling policy, the scaling operation of policy configuration is executed.

It will be apparent to those skilled in the art that the modules or steps of the present invention described above may be implemented by a general purpose computing device, they may be centralized on a single computing device or distributed across a network of multiple computing devices, and alternatively, they may be implemented by program code executable by a computing device, such that they may be stored in a storage device and executed by a computing device, and in some cases, the steps shown or described may be performed in an order different than that described herein, or they may be separately fabricated into individual integrated circuit modules, or multiple ones of them may be fabricated into a single integrated circuit module. Thus, the present invention is not limited to any specific combination of hardware and software.

The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A virtual resource scheduling method is characterized by comprising the following steps:

obtaining performance data related to a virtualized network function, VNF;

determining a scheduling policy of virtual resources of the VNF according to the performance data;

executing scheduling operation on the virtual resource according to the scheduling strategy;

wherein the performance data comprises at least one of: system performance data and service index data, wherein the system performance data is quantitative data used for measuring the performance of virtual resources deploying the VNF; the service index data is quantitative data for measuring service characteristics of a service system of the VNF;

the determining the scheduling policy for virtual resources of the VNF based on the performance data comprises: summarizing the system performance data collected and reported by the VNF and the system performance data collected and reported by a Virtual Infrastructure Manager (VIM); and determining the scheduling strategy of the virtual resources of the VNF according to the service index data and the summarized system performance data.

2. The method of claim 1, wherein aggregating the system performance data collected and reported by the VNF and the system performance data collected and reported by the VIM comprises:

for system performance data that is collectable from both the VNF and from the VIM, subject to system performance data that is collected from either the VNF or from the VIM; or

And summarizing the system performance data collected from both the VNF and the VIM according to a weighted average algorithm.

3. The method according to claim 1 or 2,

the service indicator data comprises at least one of: traffic throughput in a unit time, other performance data related to traffic characteristics;

the system performance data includes at least one of: CPU utilization, memory utilization, disk write/read performance index, remaining disk space, network bandwidth utilization, network load, network card packet loss.

4. The method of claim 1, wherein prior to obtaining the performance data related to the VNF, the method further comprises:

and sending subscription information to the VNF, wherein the subscription information is used for subscribing one or more of the performance data and setting a reporting period of the performance data.

5. The method of claim 4,

the subscription information is further used for setting a trigger threshold and a duration for reporting the performance data.

6. The method of claim 5, wherein after sending the subscription information to the VNF, the method further comprises:

the VNF receiving the subscription information;

and reporting the performance data by the VNF under the condition that the duration time of the performance data exceeding the trigger threshold exceeds the duration time.

7. The method of claim 4,

the reporting period of the performance data reported by the VNF is the same as the reporting period of the performance data reported by the Virtual Infrastructure Manager (VIM).

8. The method of claim 1, wherein performing the scheduling operation on the virtual resource comprises:

sending a pre-scheduling notification to the VNF, wherein the pre-scheduling notification is used for indicating that a scheduling operation is to be performed on the virtual resource;

and executing scheduling operation on the virtual resource under the condition of receiving a pre-scheduling confirmation response, wherein the pre-scheduling confirmation response is used for indicating that the VNF has completed business processing related to the scheduling operation of the virtual resource and allowing a virtualization system to execute resource scaling operation.

9. The method of claim 1, wherein after performing the scheduling operation on the virtual resource, the method further comprises:

sending a scheduling completion confirmation notification to the VNF, wherein the scheduling completion confirmation notification is used for indicating that the scheduling operation on the virtual resource is completed.

10. The method of claim 1, wherein the scheduling operation comprises: in the case of adding a virtual resource or changing a virtual resource, before performing a scheduling operation on the virtual resource, the method further includes:

receiving physical machine system performance data of a physical machine reported by a Virtual Infrastructure Manager (VIM);

and selecting the first physical machine of which the system performance data of the physical machine meets the scheduling operation requirement as a host machine of the newly added virtual resource or the changed virtual resource.

11. The method according to any of claims 1-2, 4-10, wherein determining the scheduling policy for virtual resources of the VNF based on the performance data comprises:

calculating a performance index of the VNF according to the performance data;

and inquiring a preset scheduling strategy corresponding to the performance index, wherein the preset scheduling strategy is used for indicating scheduling operation to be executed under the condition of the performance index.

12. The method of claim 3, wherein determining the scheduling policy for virtual resources of the VNF based on the performance data comprises:

calculating a performance index of the VNF according to the performance data;

13. The method of claim 11, wherein the performance indicator is calculated by the following formula:

the performance index S =

；

Wherein,

a preset indexing weight representing the ith performance data of the VNF,

A value representing the i-th capability data of the VNF,

A value representing the ith capability data when the VNF is fully operational.

14. The method of claim 12, wherein the performance indicator is calculated by the formula:

the performance index S =

；

Wherein,

a preset indexing weight representing the ith performance data of the VNF,

A value representing the i-th capability data of the VNF,

A value representing the ith capability data when the VNF is fully operational.

15. A virtual resource scheduling apparatus, comprising:

an acquisition module for acquiring performance data related to a virtualized network function, VNF;

a determining module configured to determine a scheduling policy of a virtual resource of the VNF according to the performance data;

the execution module is used for executing scheduling operation on the virtual resource according to the scheduling strategy;

the determining module comprises: a collecting unit, configured to collect and report the system performance data collected and reported by the VNF and the system performance data collected and reported by the virtualized infrastructure manager VIM; a determining unit, configured to determine the scheduling policy of the virtual resource of the VNF according to the service index data and the summarized system performance data.

16. The apparatus of claim 15, wherein the aggregation unit is configured to:

17. The apparatus of claim 15 or 16,

18. The apparatus of claim 15, further comprising:

a first sending module, configured to send subscription information to the VNF, where the subscription information is used to subscribe to one or more of the performance data, and set a reporting period of the performance data.

19. The apparatus of claim 18,

20. The apparatus of claim 18,

21. The apparatus of claim 15, wherein the means for performing comprises:

a sending unit, configured to send a pre-scheduling notification to the VNF, where the pre-scheduling notification is used to indicate that a scheduling operation is to be performed on the virtual resource;

an execution unit, configured to execute a scheduling operation on the virtual resource when a pre-scheduling confirmation response is received, where the pre-scheduling confirmation response is used to indicate that the VNF has completed a service process related to the scheduling operation of the virtual resource, and allow a virtualization system to perform a resource scaling operation.

22. The apparatus of claim 15, further comprising:

a second sending module, configured to send a scheduling completion acknowledgement notification to the VNF, where the scheduling completion acknowledgement notification is used to indicate that scheduling operations on the virtual resource have been completed.

23. The apparatus of claim 15, further comprising:

a receiving module, configured to receive physical machine system performance data of a physical machine, which is reported by a virtualized infrastructure manager VIM, when the scheduling operation includes increasing a virtual resource or changing a virtual resource;

and the selection module is used for selecting the first physical machine of which the system performance data of the physical machine meets the scheduling operation requirement as a host machine of the newly added virtual resource or the changed virtual resource.

24. The apparatus of any one of claims 15-16 and 18-23, wherein the determining means comprises:

a calculating unit, configured to calculate a performance index of the VNF according to the performance data;

and the query unit is used for querying a preset scheduling strategy corresponding to the performance index, wherein the preset scheduling strategy is used for indicating scheduling operation to be executed under the condition of the performance index.

25. The apparatus of claim 17, wherein the determining module comprises:

26. The apparatus according to claim 24, wherein the calculation unit calculates the performance indicator by the following formula:

the performance index S =

；

Wherein,

a preset indexing weight representing the ith performance data of the VNF,

A value representing the i-th capability data of the VNF,

A value representing the ith capability data when the VNF is fully operational.

27. The apparatus according to claim 25, wherein the calculation unit calculates the performance indicator by the following formula:

the performance index S =

；

Wherein,

a preset indexing weight representing the ith performance data of the VNF,

A value representing the i-th capability data of the VNF,

A value representing the ith capability data when the VNF is fully operational.

28. A virtual resource scheduling system, the system comprising: a virtualization infrastructure manager VNFM and a virtualization network function VNF, wherein,

the VNFM is connected with the VNF, the VNFM comprising the virtual resource scheduling apparatus of any one of claims 15 to 27.

29. The virtual resource scheduling system of claim 28 wherein said system further comprises: virtualizing an infrastructure manager, VIM, wherein,

the VIM is connected with the VNFM, and the VIM is used for collecting and reporting system performance data of the VNF.