CN105335208A - Method and device for determining probability of physical computing resource conflict in virtual environment - Google Patents

Abstract

The invention aims to provide a method for determining the probability of a physical computing resource conflict in a virtual environment. Specifically, the method comprises: obtaining physical CPU load information allocated for each virtual machine within a sampling period; obtaining corresponding virtual machine internal CPU load information of each virtual machine within the sampling period; and according to the physical CPU load information and the virtual machine internal CPU load information of each virtual machine, determining the probability of the physical computing resource conflict in the virtual environment. Compared with the prior art, the method has the characteristic that the probability of the physical computing resource conflict in the virtual environment can be easily calculated by measuring the physical CPU load information and the virtual machine internal CPU load information of each virtual machine.

Description

Determine the method and apparatus of physical computing resources collision probability in virtual environment

Technical field

The present invention relates to field of computer technology, particularly relating to a kind of technology of the probability for determining physical computing resources conflict in virtual environment.

Background technology

Virtual machine technique supports the demand that user simulates one or more virtual computing machine on a physical computer, is very easy to the working and learning of user.And under virtual environment, multiple virtual machine shares Same Physical hardware computing resource.Due to sharing of resource, between different virtual machine, there is the competition of physical computing resources, and prior art does not provide the method for the probability calculating physical computing resources conflict under virtual environment.Simultaneously, when unknown physical computing resources collision probability, the resource contention that system and virtual machine all cannot take corresponding recovery measure to cause to the competition eliminating physical computing resources, system finally can be caused to be absorbed in serious resource contention state, especially serious for impact the real-time system run in virtual environment, because the conflict of physical computing resources can cause the remarkable increase of shake and time delay, have a strong impact on the service quality being supplied to user.

Summary of the invention

The object of this invention is to provide a kind of method and apparatus of the probability for determining physical computing resources conflict in virtual environment.

According to an aspect of the present invention, provide a kind of method of the probability for determining physical computing resources conflict in virtual environment, wherein, at least one physical cpu and at least two virtual machines are configured in physical machine corresponding to described virtual environment, virtual machine described in each has at least one virtual cpu, and the method comprises:

A obtains the physical cpu information on load that virtual machine described in each is assigned with within the sampling period;

B obtains the virtual machine internal cpu load information corresponding within the described sampling period of virtual machine described in each;

C, according to the physical cpu information on load of virtual machine described in each and virtual machine internal cpu load information, determines the probability of the physical computing resources conflict in described virtual environment.

According to a further aspect in the invention, additionally provide a kind of collision probability determination equipment of the probability for determining physical computing resources conflict in virtual environment, wherein, at least one physical cpu and at least two virtual machines are configured in physical machine corresponding to described virtual environment, virtual machine described in each has at least one virtual cpu, and this collision probability determination equipment comprises:

First acquisition device, for obtaining the physical cpu information on load that virtual machine described in each is assigned with within the sampling period;

Second acquisition device, for obtaining the virtual machine internal cpu load information corresponding within the described sampling period of virtual machine described in each;

Collision probability determining device, for according to the physical cpu information on load of virtual machine described in each and virtual machine internal cpu load information, determines the probability of the physical computing resources conflict in described virtual environment.

Compared with prior art, the present invention, by measuring physical cpu information on load and the virtual machine internal cpu load information of virtual machine described in each, easily can calculate the probability of the physical computing resources conflict in described virtual environment; In addition, the probability that described physical computing resources conflicts also can be sent to correspondence system by the present invention, the probability conflicted according to described physical computing resources for described system carries out respective handling, make system can improve the reliability of system according to the probability of physical computing resources conflict, especially for real-time system, invention increases the response time of real-time system.

Accompanying drawing explanation

By reading the detailed description done non-limiting example done with reference to the following drawings, other features, objects and advantages of the present invention will become more obvious:

Fig. 1 illustrates the equipment schematic diagram of a kind of probability for determining physical computing resources conflict in virtual environment according to one aspect of the invention;

Fig. 2 illustrates the method flow diagram of a kind of probability for determining physical computing resources conflict in virtual environment according to a further aspect of the present invention.

In accompanying drawing, same or analogous Reference numeral represents same or analogous parts.

Embodiment

Below in conjunction with accompanying drawing, the present invention is described in further detail.

Fig. 1 illustrates the collision probability determination equipment 1 of a kind of probability for determining physical computing resources conflict in virtual environment according to one aspect of the invention, wherein, at least one physical cpu and at least two virtual machines are configured in physical machine corresponding to described virtual environment, virtual machine described in each has at least one virtual cpu, wherein, collision probability determination equipment 1 comprises the first acquisition device 11, second acquisition device 12 and collision probability determining device 13.Particularly, the first acquisition device 11 obtains the physical cpu information on load that virtual machine described in each is assigned with within the sampling period; Second acquisition device 12 obtains the virtual machine internal cpu load information corresponding within the described sampling period of virtual machine described in each; Collision probability determining device 13, according to the physical cpu information on load of virtual machine described in each and virtual machine internal cpu load information, determines the probability of the physical computing resources conflict in described virtual environment.At this, collision probability determination equipment 1 can by network host, single network server, multiple webserver collection or the realization such as set of computers based on cloud computing.At this, cloud is formed by based on a large amount of main frame of cloud computing (CloudComputing) or the webserver, and wherein, cloud computing is the one of Distributed Calculation, the super virtual machine be made up of a group loosely-coupled computing machine collection.Those skilled in the art will be understood that above-mentioned collision probability determination equipment 1 is only citing, and other network equipments that are existing or that may occur from now on, as being applicable to the present invention, within also should being included in scope, and are contained in this at this with way of reference.At this, the network equipment comprise a kind of can according in advance setting or the instruction stored, automatically carry out the electronic equipment of numerical evaluation and information processing, its hardware includes but not limited to microprocessor, special IC (ASIC), programmable gate array (FPGA), digital processing unit (DSP), embedded device etc.

Particularly, the explorer application programming interfaces (API) that the first acquisition device 11 provides by the physical machine self called corresponding to virtual environment, obtain the physical cpu information on load that virtual machine described in each is assigned with within the sampling period; Or the physical cpu information on load be assigned with by each virtual machine configured in the physical machine corresponding to scanning virtual environment, obtains the physical cpu information on load that virtual machine described in each is assigned with within the sampling period.

Such as, suppose in a physical machine as physicalmachine1 installed virtual machine, thus form virtual environment virtualenviroment1, configure m physical cpu in this physical machine (as with P ₁, P ₂..., P _mrepresent), n virtual machine is (as with VM ₁, VM ₂..., VM _nrepresent), virtual machine VM ₁to VM _nthe virtual cpu quantity had is respectively V1, V2, ..., Vn, then the first acquisition device 11 by physical machine as physicalmachine1 self the explorer application programming interfaces (API) that provide, just obtain virtual machine described in each in the sampling period as the physical cpu information on load be assigned with in T, namely obtain virtual machine VM ₁to VM _nthe physical cpu information on load be assigned with in each comfortable sampling period T, available formula (1) is expressed as follows:

$\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_{\mathrm{ji}}\%,j=\mathrm{1,2},...,n---\left(1\right)$

Wherein, a _ji% represents that i-th physical cpu is a jth virtual machine VM _jthe physical cpu information on load distributed, such as:

I) as j=1, above-mentioned formula (1) represents virtual machine VM ₁the physical cpu information on load be assigned with, that is: y _vM1=a1 ₁%+a1 ₂%+...+a1 _m% (wherein, a1 ₁% represents that the 1st physical cpu is VM ₁the physical cpu information on load distributed, a1 ₂% represents that the 2nd physical cpu is VM ₁the physical cpu information on load distributed ..., a1 _m% represents that m physical cpu is VM ₁the physical cpu information on load distributed);

II) as j=2, above-mentioned formula (1) represents virtual machine VM ₂the physical cpu information on load be assigned with, that is: y _vM2=a2 ₁%+a2 ₂%+...+a2 _m% (a2 ₁% represents that the 1st physical cpu is VM ₂the physical cpu information on load distributed, a2 ₂% represents that the 2nd physical cpu is VM ₂the physical cpu information on load distributed ..., a2 _m% represents that m physical cpu is VM ₂the physical cpu information on load distributed).For other virtual machines, can (1) by formula the like.

For another example, if each physical cpu is that the cpu load that same virtual machine distributes is identical within the sampling period; Or, when the first acquisition device 11 gets the average physical cpu load information of each physical cpu for each virtual cpu in same virtual machine within the sampling period, the physical cpu information on load that the first acquisition device 11 is assigned with within the sampling period by calculating virtual machine described in each with under type.

Such as, suppose the first acquisition device 11 by physical machine as physicalmachine1 self the explorer application programming interfaces (API) that provide, m the physical cpu got on it is VM ₁the physical cpu information on load distributed is T ₁%, then the first acquisition device 11 can calculate virtual machine VM ₁physical cpu information on load=the T be assigned with ₁%*m; And this m physical cpu is virtual machine VM ₂the physical cpu information on load distributed is T ₂%, then virtual machine VM ₂physical cpu information on load=the T be assigned with ₂%*m; Similarly, this m physical cpu is virtual machine VM _nthe physical cpu information on load distributed is T _n%, then virtual machine VM _nphysical cpu information on load=the T be assigned with _n%*m, i.e. virtual machine VM _jthe physical cpu information on load be assigned with can represent with following formula (2):

T _j％*m，j＝1,2，...，n(2)

Those skilled in the art will be understood that the mode of the physical cpu information on load that above-mentioned acquisition virtual machine described in each is assigned with within the sampling period is only citing; the mode of the physical cpu information on load that other acquisitions that are existing or that may occur from now on virtual machine described in each is assigned with within the sampling period is as being applicable to the present invention; also within scope should being included in, and this is contained at this with way of reference.

The explorer application programming interfaces (API) that second acquisition device 12 self provides by calling each virtual machine, obtain the virtual machine internal cpu load information that virtual machine described in each is corresponding within the described sampling period.

Such as, for virtual machine VM ₁to VM _nthe then explorer application programming interfaces (API) that self provide by each virtual machine of the second acquisition device 12, obtain the virtual machine internal cpu load information that each virtual machine is corresponding in sampling period T, available formula (3) is expressed as follows:

$\underset{z=1}{\overset{V_j}{\mathrm{\Σ}}}{b}_{\mathrm{jz}}\%,j=\mathrm{1,2},...,n---\left(3\right)$

Wherein, b _jz% represents a jth virtual machine VM _jin the virtual machine internal cpu load information corresponding in sampling period T of z virtual cpu, such as:

I) as j=1, above-mentioned formula (3) represents virtual machine VM ₁corresponding virtual machine internal cpu load information, that is: x _vM1=b1 ₁%+b1 ₂%+...+b1 _v1% (b1 ₁% represents virtual machine VM ₁in the virtual machine internal cpu load information corresponding in sampling period T of the 1st virtual cpu, b1 ₂% represents virtual machine VM ₁in the virtual machine internal cpu load information corresponding in sampling period T of the 2nd virtual cpu ..., b1 _v1% represents virtual machine VM ₁in the virtual machine internal cpu load information corresponding in sampling period T of V1 virtual cpu);

II) as j=2, above-mentioned formula (3) represents virtual machine VM ₂corresponding virtual machine internal cpu load information, that is: x _vM2=b2 ₁%+b2 ₂%+...+b2 _v2% (b2 ₁% represents virtual machine VM ₂in the virtual machine internal cpu load information corresponding in sampling period T of the 1st virtual cpu, b2 ₂% represents virtual machine VM ₂in the virtual machine internal cpu load information corresponding in sampling period T of the 2nd virtual cpu ..., b2 _v2% represents virtual machine VM ₂in the virtual machine internal cpu load information corresponding in sampling period T of V2 virtual cpu).For other virtual machines, can (3) by formula the like.

For another example, if the virtual machine internal cpu load information of each virtual cpu of same virtual machine is identical, or, when the mean virtual machine innernal CPU information on load of each virtual cpu that the second acquisition device 12 gets each virtual machine within the sampling period, the second acquisition device 12 is by calculating each virtual machine virtual machine internal cpu load information corresponding within the described sampling period with under type.Such as, suppose that the second acquisition device 12 is by calling virtual machine VM ₁the explorer application programming interfaces (API) self provided, the mean virtual machine innernal CPU information on load of V1 virtual cpu within the sampling period obtained in this virtual machine is L ₁%, then the second acquisition device 12 can determine virtual machine VM ₁corresponding virtual machine internal cpu load=L ₁%*V1; And for virtual machine VM ₂, the mean virtual machine innernal CPU information on load of V2 virtual cpu within the sampling period that the second acquisition device 12 obtains in this virtual machine is L ₂%, then the second acquisition device 12 can determine virtual machine VM ₂corresponding virtual machine internal cpu load=L ₂%*V2; Similarly, for virtual machine VM _n, the mean virtual machine innernal CPU information on load of Vn virtual cpu within the sampling period that the second acquisition device 12 obtains in this virtual machine is L _n%, then the second acquisition device 12 can determine virtual machine VM _ncorresponding virtual machine internal cpu load=L _n%*Vn, i.e. virtual machine VM _jcorresponding virtual machine internal cpu load can represent with following formula (4):

L _j%*Vj, j=1,2 ..., wherein, Vj represents a jth virtual machine VM to n (4) _jvirtual cpu quantity information.

Those skilled in the art will be understood that the mode of the virtual machine internal cpu load information that above-mentioned acquisition virtual machine described in each is corresponding within the described sampling period is only citing; the mode of the virtual machine internal cpu load information that other acquisitions that are existing or that may occur from now on virtual machine described in each is corresponding within the described sampling period is as being applicable to the present invention; also within scope should being included in, and this is contained at this with way of reference.

At this, it will be appreciated by those skilled in the art that in a particular embodiment, the execution of the first acquisition device 11 and the second acquisition device 12 serializable, the execution that also can walk abreast.When performing in a serial fashion, also first can perform the second acquisition device 12, rear execution first acquisition device 11.

At this, it will be appreciated by those skilled in the art that in a particular embodiment, the first acquisition device 11, together with the second acquisition device 12 accessible site, may also be independently module.

Collision probability determining device 13, according to the physical cpu information on load of virtual machine described in each and virtual machine internal cpu load information, determines the probability of the physical computing resources conflict in described virtual environment.At this, collision probability determining device 13 determines that the mode of described probability includes but not limited to following at least any one:

1) first according to physical cpu information on load and the virtual machine internal cpu load information of virtual machine described in each, the competition cpu load information that corresponding virtual machine is corresponding within the described sampling period is determined; Then according to the physical cpu information on load of virtual machine described in each and competition cpu load information, determine the probability of the physical computing resources conflict in described virtual environment then.

At this, exist when a certain virtual machine has task pending but the situation that taken by other virtual machines of the physical cpu of physical machine within the sampling period, by the competition cpu load information that namely the physical cpu information on load that other virtual machines take is this virtual machine, such as, for being arranged on physical machine as the n on physicalmachine1 virtual machine VM ₁to VM _nif, in sampling period T=1000ms, virtual machine VM ₁the physical cpu information on load be assigned with is 10%*T=100ms, as virtual machine VM ₁when executing the task, its virtual machine internal cpu load information is 80%*T=800ms, at virtual machine VM ₁self it seems, in this 800ms, it all can be executed the task, but in fact its time that can execute the task only has 100ms, and other 700ms in this 800ms are taken by other virtual machines, then namely this 700ms is virtual machine VM ₁competition cpu load information.

Such as, because of each virtual machine internal cpu load information that virtual machine is corresponding within the described sampling period equal its physical cpu information on load within this sampling period and competition cpu load information and, so, can according to the physical cpu information on load of virtual machine described in each and virtual machine internal cpu load information, determine the competition cpu load information that corresponding virtual machine is corresponding within the described sampling period, according to above-mentioned formula (1) and (3), both subtract each other, formula (3) deducts formula (1), the competition cpu load information that virtual machine is corresponding within the described sampling period can be obtained, that is:

${\mathrm{\η}}_j\%=\underset{z=1}{\overset{V_j}{\mathrm{\Σ}}}{b}_{\mathrm{jz}}\%-\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_{\mathrm{ji}}\%,j=\mathrm{1,2},...,n---\left(5\right)$

Wherein, η _j% represents a jth virtual machine VM _jcompetition cpu load information.Such as:

I) as j=1, above-mentioned formula (5) represents the 1st virtual machine VM ₁competition cpu load information, that is:

η ₁％＝x _VM1-y _VM1＝(b1 ₁％+b1 ₂％+...+b1 _V1％)-(a1 ₁％+a1 ₂％+...+a1 _m％)；

II) as j=2, above-mentioned formula (5) represents the 2nd virtual machine VM ₂competition cpu load information, that is:

η ₂％＝x _VM2-y _VM2＝(b2 ₁％+b2 ₂％+...+b2 _V2％)-(a2 ₁％+a2 ₂％+...+a2 _m％)。

Then, collision probability determining device 13 according to physical cpu information on load and the competition cpu load information of virtual machine described in each, determines the probability of the physical computing resources conflict in described virtual environment again.At this, collision probability determining device 13 using the competition cpu load information of all virtual machines and with the physical cpu information on load of all virtual machines and the probability that conflicts as the physical computing resources in virtual environment of ratio, namely determine the probability of the physical computing resources conflict in described virtual environment according to following formula (6):

$\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{\mathrm{\η}}_j\%}{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}(\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_{\mathrm{ji}}\%)}=\frac{{\mathrm{\η}}_1\%+{\mathrm{\η}}_2\%+...+{\mathrm{\η}}_n\%}{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_{1i}\%+\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_{2i}\%+...+\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_{\mathrm{ni}}\%}---\left(6\right)$

When the numerical value of formula (6) equals 0, illustrate now there is no physical computing resources conflict, if the numerical value of formula (6) is greater than 0, illustrate now there is physical computing resources conflict, and numerical value is larger, illustrate that the phenomenon that physical computing resources conflicts is more serious.

For another example, according to above-mentioned formula (2) and (4), the competition cpu load information of each virtual machine can be obtained, as formula (4) is deducted formula (2), can obtain:

C _j％＝L _j％*Vj-T _j％*m，j＝1,2，...，n(7)

Wherein, C _j% represents a jth virtual machine VM _jthe average of competition cpu load information.According to above-mentioned formula (7), then have:

T ₁％*m+C ₁％＝L ₁％*V1(7-1)

T ₂％*m+C ₂％＝L ₂％*V2(7-2)

……

T _n％*m+C _n％＝L _n％*Vn(7-n)

Calculate above formula (7-1) to (7-n) and, then can obtain:

(T ₁%+T ₂%+...+T _n%) * m+ (C ₁%+C ₂%+...C _n%)=L ₁%*V1+L ₂%*V2+...+L _n%*Vn, namely obtains following formula (8):

$(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{T}_j\%)*m+\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{C}_j\%=\underset{j=1}{\overset{n}{\mathrm{\Σ}}}(L_j\%*V_j)---\left(8\right)$

Then, the probability of physical computing resources conflict that collision probability determining device 13 can be determined in described virtual environment is:

$\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{C}_j\%}{(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{T}_j\%)*m}=\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}(L_j\%*V_j)-(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{T}_j\%)*m}{(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{T}_j\%)*m}=\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}(L_j\%*V_j)}{(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{T}_j\%)*m}-1---\left(9\right)$

If when there is not physical computing resources conflict, the numerical value of above-mentioned formula (9) equals 0, if when the numerical value of above-mentioned formula (9) is greater than 0, there is physical computing resources conflict in explanation, and numerical value is larger, illustrate that the phenomenon that physical computing resources conflicts is more serious.

2) according to physical cpu information on load and the virtual machine internal cpu load information of virtual machine described in each, and in conjunction with the quantity information of the physical cpu configured in the physical machine corresponding to described virtual environment, and the virtual cpu quantity information described in each corresponding to virtual machine, determine the probability of the physical computing resources conflict in described virtual environment, as the probability of physical computing resources conflict as described in determining according to above-mentioned formula (9).

Preferably, collision probability determining device 13 also can according to the physical cpu information on load of virtual machine described in each and virtual machine internal cpu load information, and in conjunction with the quantity information of the physical cpu configured in the physical machine corresponding to described virtual environment, the virtual cpu quantity information described in each corresponding to virtual machine and the scheduling overhead information about all virtual machines, determine the probability of the physical computing resources conflict in described virtual environment.

At this, the scheduling overhead information about all virtual machines can remove by collision probability determining device 13 from the virtual machine internal cpu load information sum of all virtual machines, by as follows for formula (9) change, obtains following formula (10):

$\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}(L_j\%*V_j)}{(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{T}_j\%)*m*(1-x\%)}-1---\left(10\right)$

Wherein, x% is the scheduling overhead information about all virtual machines.Then, the probability of the physical computing resources conflict in described virtual environment is determined according to formula (10).If when there is not physical computing resources conflict, the numerical value of above-mentioned formula (10) equals 0, if when the numerical value of above-mentioned formula (10) is not equal to 0, there is physical computing resources conflict in explanation, and numerical value is larger, illustrate that the phenomenon that physical computing resources conflicts is more serious.

Such as, for physical machine as physicalmachine1, in this physical machine, be configured with m=3 physical cpu, and n=6 virtual machine (i.e. VM ₁, VM ₂..., VM ₆), and virtual machine VM ₁to VM ₆on virtual cpu quantity be followed successively by: 1,2,3,4,5,6, virtual machine VM ₁to VM ₆in virtual machine internal cpu load information (the i.e. L of virtual cpu _j%) be followed successively by 80%, 90%, 60%, 70%, 100%, 80%, each physical cpu is virtual machine VM within the sampling period ₁to VM ₆cpu load (the i.e. T distributed _j%) be followed successively by 10%, 20%, 30%, 40%, 50%, 10%, scheduling overhead information about all virtual machines is 5%, i.e. x%=5%, then collision probability determining device 13 can according to above formula (10), the probability obtaining the physical computing resources conflict in virtual environment virtualenviroment1 is 2.728, illustrates to there is physical computing resources conflict.

On virtual machine has been installed, thus form virtual environment virtualenviroment1, configure m physical cpu in this physical machine (as with P ₁, P ₂..., P _mrepresent), n virtual machine is (as with VM ₁, VM ₂..., VM _nrepresent)

3) when the physical cpu information on load that each physical cpu in described virtual environment distributes for same virtual machine is equal and the virtual machine internal cpu load information of each virtual cpu contribution corresponding to each virtual machine is equal, collision probability determining device 13 by with under type, determines the probability of the physical computing resources conflict in described virtual environment:

$y=\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}(L_j\%*V_j)}{(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{T}_j\%)*m*(1-x\%)}$

Wherein, the quantity information of the physical cpu configured in the physical machine of m corresponding to described virtual environment, L _j% is the virtual machine internal cpu load information of each virtual cpu contribution of a jth virtual machine in n virtual machine configuring in described virtual environment within the described sampling period corresponding to it, V _jfor the virtual cpu quantity information that a jth virtual machine has, T _j% be a jth virtual machine within the described sampling period by the physical cpu information on load that each physical cpu distributes, x% is the scheduling overhead information about all virtual machines.

At this, collision probability determining device 13 determines that the probability of the physical computing resources conflict in described virtual environment is identical with the mode of the probability that it conflicts according to the physical computing resources that above-mentioned formula (10) is determined in described virtual environment, for simplicity's sake, therefore do not repeat them here, and comprise by reference therewith.

Constant work between each device of collision probability determination equipment 1.Particularly, the first acquisition device 11 continues to obtain the physical cpu information on load that described in each, virtual machine is assigned with within the sampling period; Second acquisition device 12 continues to obtain the virtual machine internal cpu load information that described in each, virtual machine is corresponding within the described sampling period; The lasting physical cpu information on load according to virtual machine described in each of collision probability determining device 13 and virtual machine internal cpu load information, determine the probability of the physical computing resources conflict in described virtual environment.At this, the probability that those skilled in the art will be understood that " continuing " refers to the physical cpu information on load constantly carrying out obtaining virtual machine respectively between each device of collision probability determination equipment 1, the virtual machine internal cpu load information obtaining virtual machine is conflicted with the physical computing resources determined in virtual environment, until collision probability determination equipment 1 stops the physical cpu information on load obtaining virtual machine in a long time.

Preferably, collision probability determination equipment 1 also comprises dispensing device (not shown).Particularly, the probability that described physical computing resources conflicts is sent to correspondence system by dispensing device, and the probability conflicted according to described physical computing resources for described system carries out respective handling.

At this, the probability of physical computing resources conflict is under many circumstances as the input of premature beats, as at LTERAN (LongTermEvolution, Long Term Evolution, RadioAccessNetwork, wireless access network) in product, utilize the probability of physical computing resources conflict can have premature beats measure below: 1) when the probability of physical computing resources conflict is greater than predefine thresholding and the operable computational resource of application program be less than predefined use maximal value time, watchdog routine is by the more physical treatment resource of cloud management platform request, if resource contention probability does not decline after this request is satisfied, watchdog routine repeats to apply for more computational resource until resource contention probability is less than predefine thresholding, 2) when the probability of physical computing resources conflict is greater than predefine thresholding and the operable computational resource of application program reached predefined use maximal value time, start overload control mechanism, namely system loading is reduced from Signaling Layer by the mode reducing user's access and customer service process, thus reduce resource contention probability, until make it lower than predefine thresholding.

Such as, the probability supposing the physical computing resources conflict that collision probability determining device 13 obtains in virtual environment virtualenviroment1 is 2.728, then dispensing device is by http, https etc. arrange communication mode, or by predefine interface, the probability that described physical computing resources conflicts is sent to correspondence system, the probability conflicted according to described physical computing resources for described system carries out respective handling, as being sent in LTERAN product, suppose that the probability of this physical computing resources conflict is greater than 2.728 predefine thresholdings as 2 and the operable computational resource of application program is less than predefined when using maximal value, then the watchdog routine of LTERAN product is by the more physical treatment resource of cloud management platform request, if resource contention probability does not decline after this request is satisfied, watchdog routine repeats to apply for more computational resource until resource contention probability is less than predefine thresholding.

Fig. 2 illustrates the method flow diagram of a kind of probability for determining physical computing resources conflict in virtual environment according to a further aspect of the present invention.

Wherein, configure at least one physical cpu and at least two virtual machines in the physical machine corresponding to described virtual environment, virtual machine described in each has at least one virtual cpu, wherein, and the method comprising the steps of S1, step S2 and step S3.Particularly, in step sl, collision probability determination equipment 1 obtains the physical cpu information on load that virtual machine described in each is assigned with within the sampling period; In step s 2, collision probability determination equipment 1 obtains the virtual machine internal cpu load information corresponding within the described sampling period of virtual machine described in each; In step s3, collision probability determination equipment 1, according to the physical cpu information on load of virtual machine described in each and virtual machine internal cpu load information, determines the probability of the physical computing resources conflict in described virtual environment.At this, collision probability determination equipment 1 can by network host, single network server, multiple webserver collection or the realization such as set of computers based on cloud computing.At this, cloud is formed by based on a large amount of main frame of cloud computing (CloudComputing) or the webserver, and wherein, cloud computing is the one of Distributed Calculation, the super virtual machine be made up of a group loosely-coupled computing machine collection.Those skilled in the art will be understood that above-mentioned collision probability determination equipment 1 is only citing, and other network equipments that are existing or that may occur from now on, as being applicable to the present invention, within also should being included in scope, and are contained in this at this with way of reference.At this, the network equipment comprise a kind of can according in advance setting or the instruction stored, automatically carry out the electronic equipment of numerical evaluation and information processing, its hardware includes but not limited to microprocessor, special IC (ASIC), programmable gate array (FPGA), digital processing unit (DSP), embedded device etc.

Particularly, in step sl, the explorer application programming interfaces (API) that collision probability determination equipment 1 provides by the physical machine self called corresponding to virtual environment, obtain the physical cpu information on load that virtual machine described in each is assigned with within the sampling period; Or the physical cpu information on load be assigned with by each virtual machine configured in the physical machine corresponding to scanning virtual environment, obtains the physical cpu information on load that virtual machine described in each is assigned with within the sampling period.

Such as, suppose in a physical machine as physicalmachine1 installed virtual machine, thus form virtual environment virtualenviroment1, configure m physical cpu in this physical machine (as with P ₁, P ₂..., P _mrepresent), n virtual machine is (as with VM ₁, VM ₂..., VM _nrepresent), virtual machine VM ₁to VM _nthe virtual cpu quantity had is respectively V1, V2, ..., Vn, then in step sl, collision probability determination equipment 1 by physical machine as physicalmachine1 self the explorer application programming interfaces (API) that provide, just obtain virtual machine described in each in the sampling period as the physical cpu information on load be assigned with in T, namely obtain virtual machine VM ₁to VM _nthe physical cpu information on load be assigned with in each comfortable sampling period T, available formula (11) is expressed as follows:

$\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_{\mathrm{ji}}\%,j=\mathrm{1,2},...,n---\left(11\right)$

Wherein, a _ji% represents that i-th physical cpu is a jth virtual machine VM _jthe physical cpu information on load distributed, such as:

I) as j=1, above-mentioned formula (11) represents virtual machine VM ₁the physical cpu information on load be assigned with, that is: y _vM1=a1 ₁%+a1 ₂%+...+a1 _m% (wherein, a1 ₁% represents that the 1st physical cpu is VM ₁the physical cpu information on load distributed, a1 ₂% represents that the 2nd physical cpu is VM ₁the physical cpu information on load distributed ..., a1 _m% represents that m physical cpu is VM ₁the physical cpu information on load distributed);

II) as j=2, above-mentioned formula (11) represents virtual machine VM ₂the physical cpu information on load be assigned with, that is: y _vM2=a2 ₁%+a2 ₂%+...+a2 _m% (a2 ₁% represents that the 1st physical cpu is VM ₂the physical cpu information on load distributed, a2 ₂% represents that the 2nd physical cpu is VM ₂the physical cpu information on load distributed ..., a2 _m% represents that m physical cpu is VM ₂the physical cpu information on load distributed).For other virtual machines, can (11) by formula the like.

For another example, if each physical cpu is that the cpu load that same virtual machine distributes is identical within the sampling period; Or, when in step sl, when collision probability determination equipment 1 gets the average physical cpu load information of each physical cpu for each virtual cpu in same virtual machine within the sampling period, in step sl, the physical cpu information on load that is assigned with within the sampling period by calculating virtual machine described in each with under type of collision probability determination equipment 1.

Such as, suppose in step sl, collision probability determination equipment 1 by physical machine as physicalmachine1 self the explorer application programming interfaces (API) that provide, m the physical cpu got on it is virtual machine VM ₁the physical cpu information on load distributed is T ₁%, then in step sl, collision probability determination equipment 1 can calculate virtual machine VM ₁physical cpu information on load=the T be assigned with ₁%*m; And this m physical cpu is virtual machine VM ₂the physical cpu information on load distributed is T ₂%, then virtual machine VM ₂physical cpu information on load=the T be assigned with ₂%*m; Similarly, this m physical cpu is virtual machine VM _nthe physical cpu information on load distributed is T _n%, then virtual machine VM _nphysical cpu information on load=the T be assigned with _n%*m, i.e. virtual machine VM _jthe physical cpu information on load be assigned with can represent with following formula (12):

T _j％*m，j＝1,2，...，n(12)

Those skilled in the art will be understood that the mode of the physical cpu information on load that above-mentioned acquisition virtual machine described in each is assigned with within the sampling period is only citing; the mode of the physical cpu information on load that other acquisitions that are existing or that may occur from now on virtual machine described in each is assigned with within the sampling period is as being applicable to the present invention; also within scope should being included in, and this is contained at this with way of reference.

In step s 2, the explorer application programming interfaces (API) that collision probability determination equipment 1 self provides by calling each virtual machine, obtain the virtual machine internal cpu load information that virtual machine described in each is corresponding within the described sampling period.

Such as, for virtual machine VM ₁to VM _nthen in step s 2, the explorer application programming interfaces (API) that collision probability determination equipment 1 self provides by each virtual machine, obtain the virtual machine internal cpu load information that each virtual machine is corresponding in sampling period T, available formula (13) is expressed as follows:

$\underset{z=1}{\overset{V_j}{\mathrm{\Σ}}}{b}_{\mathrm{jz}}\%,j=\mathrm{1,2},...,n---\left(13\right)$

Wherein, b _jz% represents a jth virtual machine VM _jin the virtual machine internal cpu load information corresponding in sampling period T of z virtual cpu, such as:

I) as j=1, above-mentioned formula (13) represents virtual machine VM ₁corresponding virtual machine internal cpu load information, that is: x _vM1=b1 ₁%+b1 ₂%+...+b1 _v1% (b1 ₁% represents virtual machine VM ₁in the virtual machine internal cpu load information corresponding in sampling period T of the 1st virtual cpu, b1 ₂% represents virtual machine VM ₁in the virtual machine internal cpu load information corresponding in sampling period T of the 2nd virtual cpu ..., b1 _v1% represents virtual machine VM ₁in the virtual machine internal cpu load information corresponding in sampling period T of V1 virtual cpu);

II) as j=2, above-mentioned formula (13) represents virtual machine VM ₂corresponding virtual machine internal cpu load information, that is: x _vM2=b2 ₁%+b2 ₂%+...+b2 _v2% (b2 ₁% represents virtual machine VM ₂in the virtual machine internal cpu load information corresponding in sampling period T of the 1st virtual cpu, b2 ₂% represents virtual machine VM ₂in the virtual machine internal cpu load information corresponding in sampling period T of the 2nd virtual cpu ..., b2 _v2% represents virtual machine VM ₂in the virtual machine internal cpu load information corresponding in sampling period T of V2 virtual cpu).For other virtual machines, can (13) by formula the like.

For another example, if the virtual machine internal cpu load information of each virtual cpu of same virtual machine is identical, or, when in step s 2, during the mean virtual machine innernal CPU information on load of each virtual cpu that collision probability determination equipment 1 gets each virtual machine within the sampling period, in step s 2, collision probability determination equipment 1 is by calculating each virtual machine virtual machine internal cpu load information corresponding within the described sampling period with under type.Such as, suppose in step s 2, collision probability determination equipment 1 is by calling virtual machine VM ₁the explorer application programming interfaces (API) self provided, the mean virtual machine innernal CPU information on load of V1 virtual cpu within the sampling period obtained in this virtual machine is L ₁%, then in step s 2, collision probability determination equipment 1 can determine virtual machine VM ₁corresponding virtual machine internal cpu load=L ₁%*V1; And for virtual machine VM ₂, in step s 2, the mean virtual machine innernal CPU information on load of V2 virtual cpu within the sampling period that collision probability determination equipment 1 obtains in this virtual machine is L ₂%, then in step s 2, collision probability determination equipment 1 can determine virtual machine VM ₂corresponding virtual machine internal cpu load=L ₂%*V2; Similarly, for virtual machine VM _n, in step s 2, the mean virtual machine innernal CPU information on load of Vn virtual cpu within the sampling period that collision probability determination equipment 1 obtains in this virtual machine is L _n%, then in step s 2, collision probability determination equipment 1 can determine virtual machine VM _ncorresponding virtual machine internal cpu load=L _n%*Vn, i.e. virtual machine VM _jcorresponding virtual machine internal cpu load can represent with following formula (14):

L _j％*Vj，j＝1,2，...，n(14)

Wherein, Vj represents a jth virtual machine VM _jvirtual cpu quantity information.

Those skilled in the art will be understood that the mode of the virtual machine internal cpu load information that above-mentioned acquisition virtual machine described in each is corresponding within the described sampling period is only citing; the mode of the virtual machine internal cpu load information that other acquisitions that are existing or that may occur from now on virtual machine described in each is corresponding within the described sampling period is as being applicable to the present invention; also within scope should being included in, and this is contained at this with way of reference.

At this, it will be appreciated by those skilled in the art that in a particular embodiment, the execution of step S1 and step S2 serializable, the execution that also can walk abreast.When performing in a serial fashion, also first can perform step S2, rear execution step S1.

In step s3, collision probability determination equipment 1, according to the physical cpu information on load of virtual machine described in each and virtual machine internal cpu load information, determines the probability of the physical computing resources conflict in described virtual environment.At this, in step s3, collision probability determination equipment 1 determines that the mode of described probability includes but not limited to following at least any one:

1) first according to physical cpu information on load and the virtual machine internal cpu load information of virtual machine described in each, the competition cpu load information that corresponding virtual machine is corresponding within the described sampling period is determined; Then according to the physical cpu information on load of virtual machine described in each and competition cpu load information, determine the probability of the physical computing resources conflict in described virtual environment then.

At this, exist when a certain virtual machine has task pending but the situation that taken by other virtual machines of the physical cpu of physical machine within the sampling period, by the competition cpu load information that namely the physical cpu information on load that other virtual machines take is this virtual machine, such as, for being arranged on physical machine as the n on physicalmachine1 virtual machine VM ₁to VM _nif, in sampling period T=1000ms, virtual machine VM ₁the physical cpu information on load be assigned with is 10%*T=100ms, as virtual machine VM ₁when executing the task, its virtual machine internal cpu load information is 80%*T=800ms, at virtual machine VM ₁self it seems, in this 800ms, it all can be executed the task, but in fact its time that can execute the task only has 100ms, and other 700ms in this 800ms are taken by other virtual machines, then namely this 700ms is virtual machine VM ₁competition cpu load information.

Such as, because of each virtual machine internal cpu load information that virtual machine is corresponding within the described sampling period equal its physical cpu information on load within this sampling period and competition cpu load information and, so, can according to the physical cpu information on load of virtual machine described in each and virtual machine internal cpu load information, determine the competition cpu load information that corresponding virtual machine is corresponding within the described sampling period, according to above-mentioned formula (11) and (13), both subtract each other, formula (13) deducts formula (11), the competition cpu load information that virtual machine is corresponding within the described sampling period can be obtained, that is:

${\mathrm{\η}}_j\%=\underset{z=1}{\overset{V_j}{\mathrm{\Σ}}}{b}_{\mathrm{jz}}\%-\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_{\mathrm{ji}}\%,j=\mathrm{1,2},...,n---\left(15\right)$

Wherein, η _j% represents a jth virtual machine VM _jcompetition cpu load information.Such as:

I) as j=1, above-mentioned formula (15) represents the 1st virtual machine VM ₁competition cpu load information, that is:

η ₁％＝x _VM1-y _VM1＝(b1 ₁％+b1 ₂％+...+b1 _V1％)-(a1 ₁％+a1 ₂％+...+a1 _m％)；

II) as j=2, above-mentioned formula (15) represents the 2nd virtual machine VM ₂competition cpu load information, that is:

η ₂％＝x _VM2-y _VM2＝(b2 ₁％+b2 ₂％+...+b2 _V2％)-(a2 ₁％+a2 ₂％+...+a2 _m％)。

Then, in step s3, collision probability determination equipment 1 according to physical cpu information on load and the competition cpu load information of virtual machine described in each, determines the probability of the physical computing resources conflict in described virtual environment again.At this, in step s3, collision probability determination equipment 1 using the competition cpu load information of all virtual machines and with the physical cpu information on load of all virtual machines and the probability that conflicts as the physical computing resources in virtual environment of ratio, namely determine the probability of the physical computing resources conflict in described virtual environment according to following formula (16):

$\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{\mathrm{\η}}_j\%}{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}(\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_{\mathrm{ji}}\%)}=\frac{{\mathrm{\η}}_1\%+{\mathrm{\η}}_2\%+...+{\mathrm{\η}}_n\%}{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_{1i}\%+\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_{2i}\%+...+\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_{\mathrm{ni}}\%}---\left(16\right)$

When the numerical value of formula (16) equals 0, illustrate now there is no physical computing resources conflict, if the numerical value of formula (16) is greater than 0, illustrate now there is physical computing resources conflict, and numerical value is larger, illustrate that the phenomenon that physical computing resources conflicts is more serious.

For another example, according to above-mentioned formula (12) and (14), the competition cpu load information of each virtual machine can be obtained, as formula (14) is deducted formula (12), can obtain:

C _j％＝L _j％*Vj-T _j％*m，j＝1,2，...，n(17)

Wherein, C _j% represents a jth virtual machine VM _jthe average of competition cpu load information.According to above-mentioned formula (17), then have:

T ₁％*m+C ₁％＝L ₁％*V1(17-1)

T ₂％*m+C ₂％＝L ₂％*V2(17-2)

……

T _n％*m+C _n％＝L _n％*Vn(17-n)

Calculate above formula (17-1) to (17-n) and, then can obtain:

(T ₁%+T ₂%+...+T _n%) * m+ (C ₁%+C ₂%+...C _n%)=L ₁%*V1+L ₂%*V2+...+L _n%*V _n, namely obtain following formula (18):

$(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{T}_j\%)*m+\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{C}_j\%=\underset{j=1}{\overset{n}{\mathrm{\Σ}}}(L_j\%*V_j)---\left(18\right)$

Then, in step s3, the probability of physical computing resources conflict that collision probability determination equipment 1 can be determined in described virtual environment is:

$\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{C}_j\%}{(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{T}_j\%)*m}=\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}(L_j\%*V_j)-(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{T}_j\%)*m}{(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{T}_j\%)*m}=\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}(L_j\%*V_j)}{(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{T}_j\%)*m}-1---\left(19\right)$

If when there is not physical computing resources conflict, the numerical value of above-mentioned formula (19) equals 0, if when the numerical value of above-mentioned formula (19) is greater than 0, there is physical computing resources conflict in explanation, and numerical value is larger, illustrate that the phenomenon that physical computing resources conflicts is more serious.

2) according to physical cpu information on load and the virtual machine internal cpu load information of virtual machine described in each, and in conjunction with the quantity information of the physical cpu configured in the physical machine corresponding to described virtual environment, and the virtual cpu quantity information described in each corresponding to virtual machine, determine the probability of the physical computing resources conflict in described virtual environment, as the probability of physical computing resources conflict as described in determining according to above-mentioned formula (19).

Preferably, in step s3, collision probability determination equipment 1 also can according to the physical cpu information on load of virtual machine described in each and virtual machine internal cpu load information, and in conjunction with the quantity information of the physical cpu configured in the physical machine corresponding to described virtual environment, the virtual cpu quantity information described in each corresponding to virtual machine and the scheduling overhead information about all virtual machines, determine the probability of the physical computing resources conflict in described virtual environment.

At this, in step s3, scheduling overhead information about all virtual machines can remove by collision probability determination equipment 1 from the virtual machine internal cpu load information sum of all virtual machines, by as follows for formula (19) change, obtains following formula (20):

$\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}(L_j\%*V_j)}{(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{T}_j\%)*m*(1-x\%)}-1---\left(20\right)$

Wherein, x% is the scheduling overhead information about all virtual machines.Then, the probability of the physical computing resources conflict in described virtual environment is determined according to formula (20).If when there is not physical computing resources conflict, the numerical value of above-mentioned formula (20) equals 0, if when the numerical value of above-mentioned formula (20) is greater than 0, there is physical computing resources conflict in explanation, and numerical value is larger, illustrate that the phenomenon that physical computing resources conflicts is more serious.

Such as, for physical machine as physicalmachine1, in this physical machine, be configured with m=3 physical cpu, and n=6 virtual machine (i.e. VM ₁, VM ₂..., VM ₆), and virtual machine VM ₁to VM ₆on virtual cpu quantity be followed successively by: 1,2,3,4,5,6, virtual machine VM ₁to VM ₆in virtual machine internal cpu load information (the i.e. L of virtual cpu _j%) be followed successively by 80%, 90%, 60%, 70%, 100%, 80%, each physical cpu is virtual machine VM within the sampling period ₁to VM ₆cpu load (the i.e. T distributed _j%) be followed successively by 10%, 20%, 30%, 40%, 50%, 10%, scheduling overhead information about all virtual machines is 5%, i.e. x%=5%, then in step s3, collision probability determination equipment 1 can according to above formula (20), the probability obtaining the physical computing resources conflict in virtual environment virtualenviroment1 is 2.728, illustrates to there is physical computing resources conflict.

On virtual machine has been installed, thus form virtual environment virtualenviroment1, configure m physical cpu in this physical machine (as with P ₁, P ₂..., P _mrepresent), n virtual machine is (as with VM ₁, VM ₂..., VM _nrepresent)

3) when the physical cpu information on load that each physical cpu in described virtual environment distributes for same virtual machine is equal and the virtual machine internal cpu load information of each virtual cpu contribution corresponding to each virtual machine is equal, in step s3, collision probability determination equipment 1 by with under type, determines the probability of the physical computing resources conflict in described virtual environment:

$y=\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}(L_j\%*V_j)}{(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{T}_j\%)*m*(1-x\%)}$

Wherein, the quantity information of the physical cpu configured in the physical machine of m corresponding to described virtual environment, L _j% is the virtual machine internal cpu load information of each virtual cpu contribution of a jth virtual machine in n virtual machine configuring in described virtual environment within the described sampling period corresponding to it, V _jfor the virtual cpu quantity information that a jth virtual machine has, T _j% be a jth virtual machine within the described sampling period by the physical cpu information on load that each physical cpu distributes, x% is the scheduling overhead information about all virtual machines.

At this, in step s3, collision probability determination equipment 1 determines that the probability of the physical computing resources conflict in described virtual environment is identical with the mode of the probability that it conflicts according to the physical computing resources that above-mentioned formula (20) is determined in described virtual environment, for simplicity's sake, therefore do not repeat them here, and comprise by reference therewith.

Constant work between each step of collision probability determination equipment 1.Particularly, in step sl, collision probability determination equipment 1 continues to obtain the physical cpu information on load that described in each, virtual machine is assigned with within the sampling period; In step s 2, collision probability determination equipment 1 continues to obtain the virtual machine internal cpu load information that described in each, virtual machine is corresponding within the described sampling period; In step s3, the lasting physical cpu information on load according to virtual machine described in each of collision probability determination equipment 1 and virtual machine internal cpu load information, determine the probability of the physical computing resources conflict in described virtual environment.At this, the probability that those skilled in the art will be understood that " continuing " refers to the physical cpu information on load constantly carrying out obtaining virtual machine respectively between each step of collision probability determination equipment 1, the virtual machine internal cpu load information obtaining virtual machine is conflicted with the physical computing resources determined in virtual environment, until collision probability determination equipment 1 stops the physical cpu information on load obtaining virtual machine in a long time.

Preferably, the method also comprises step S4 (not shown).Particularly, in step s 4 which, the probability that described physical computing resources conflicts is sent to correspondence system by collision probability determination equipment 1, and the probability conflicted according to described physical computing resources for described system carries out respective handling.

At this, the probability of physical computing resources conflict is under many circumstances as the input of premature beats, as at LTERAN (LongTermEvolution, Long Term Evolution, RadioAccessNetwork, wireless access network) in product, utilize the probability of physical computing resources conflict can have premature beats measure below: 1) when the probability of physical computing resources conflict is greater than predefine thresholding and the operable computational resource of application program be less than predefined use maximal value time, watchdog routine is by the more physical treatment resource of cloud management platform request, if resource contention probability does not decline after this request is satisfied, watchdog routine repeats to apply for more computational resource until resource contention probability is less than predefine thresholding, 2) when the probability of physical computing resources conflict is greater than predefine thresholding and the operable computational resource of application program reached predefined use maximal value time, start overload control mechanism, namely system loading is reduced from Signaling Layer by the mode reducing user's access and customer service process, thus reduce resource contention probability, until make it lower than predefine thresholding.

Such as, suppose in step s3, the probability of the physical computing resources conflict that collision probability determination equipment 1 obtains in virtual environment virtualenviroment1 is 2.728, then in step s 4 which, collision probability determination equipment 1 is by http, https etc. arrange communication mode, or by predefine interface, the probability that described physical computing resources conflicts is sent to correspondence system, the probability conflicted according to described physical computing resources for described system carries out respective handling, as being sent in LTERAN product, suppose that the probability of this physical computing resources conflict is greater than 2.728 predefine thresholdings as 2 and the operable computational resource of application program is less than predefined when using maximal value, then the watchdog routine of LTERAN product is by the more physical treatment resource of cloud management platform request, if resource contention probability does not decline after this request is satisfied, watchdog routine repeats to apply for more computational resource until resource contention probability is less than predefine thresholding.

It should be noted that the present invention can be implemented in the assembly of software and/or software restraint, such as, special IC (ASIC), general object computing machine or any other similar hardware device can be adopted to realize.In one embodiment, software program of the present invention can perform to realize step mentioned above or function by processor.Similarly, software program of the present invention (comprising relevant data structure) can be stored in computer readable recording medium storing program for performing, such as, and RAM storer, magnetic or CD-ROM driver or flexible plastic disc and similar devices.In addition, steps more of the present invention or function can adopt hardware to realize, such as, as coordinating with processor thus performing the circuit of each step or function.

In addition, a part of the present invention can be applied to computer program, such as computer program instructions, when it is performed by computing machine, by the operation of this computing machine, can call or provide according to method of the present invention and/or technical scheme.And call the programmed instruction of method of the present invention, may be stored in fixing or moveable recording medium, and/or be transmitted by the data stream in broadcast or other signal bearing medias, and/or be stored in the working storage of the computer equipment run according to described programmed instruction.At this, comprise a device according to one embodiment of present invention, this device comprises the storer for storing computer program instructions and the processor for execution of program instructions, wherein, when this computer program instructions is performed by this processor, trigger this plant running based on the aforementioned method according to multiple embodiment of the present invention and/or technical scheme.

To those skilled in the art, obviously the invention is not restricted to the details of above-mentioned one exemplary embodiment, and when not deviating from spirit of the present invention or essential characteristic, the present invention can be realized in other specific forms.Therefore, no matter from which point, all should embodiment be regarded as exemplary, and be nonrestrictive, scope of the present invention is limited by claims instead of above-mentioned explanation, and all changes be therefore intended in the implication of the equivalency by dropping on claim and scope are included in the present invention.Any Reference numeral in claim should be considered as the claim involved by limiting.In addition, obviously " comprising " one word do not get rid of other unit or step, odd number does not get rid of plural number.Multiple unit of stating in device claim or device also can be realized by software or hardware by a unit or device.First, second word such as grade is used for representing title, and does not represent any specific order.

Claims (12)

1. one kind for determining the method for the probability of the conflict of physical computing resources in virtual environment, wherein, configure at least one physical cpu and at least two virtual machines in physical machine corresponding to described virtual environment, virtual machine described in each has at least one virtual cpu, and the method comprises:

A obtains the physical cpu information on load that virtual machine described in each is assigned with within the sampling period;

B obtains the virtual machine internal cpu load information corresponding within the described sampling period of virtual machine described in each;

C, according to the physical cpu information on load of virtual machine described in each and virtual machine internal cpu load information, determines the probability of the physical computing resources conflict in described virtual environment.

2. method according to claim 1, wherein, described step c comprises:

-according to the physical cpu information on load of virtual machine described in each and virtual machine internal cpu load information, determine the competition cpu load information that corresponding virtual machine is corresponding within the described sampling period;

-according to the physical cpu information on load of virtual machine described in each and competition cpu load information, determine the probability of the physical computing resources conflict in described virtual environment.

3. method according to claim 1, wherein, described step c comprises:

-according to the physical cpu information on load of virtual machine described in each and virtual machine internal cpu load information, and in conjunction with the quantity information of the physical cpu configured in the physical machine corresponding to described virtual environment, and the virtual cpu quantity information described in each corresponding to virtual machine, determine the probability of the physical computing resources conflict in described virtual environment.

4. method according to claim 3, wherein, described step c comprises:

-according to the physical cpu information on load of virtual machine described in each and virtual machine internal cpu load information, and in conjunction with the quantity information of the physical cpu configured in the physical machine corresponding to described virtual environment, the virtual cpu quantity information described in each corresponding to virtual machine and the scheduling overhead information about all virtual machines, determine the probability of the physical computing resources conflict in described virtual environment.

5. method according to claim 1, wherein, when the virtual machine internal cpu load information of each virtual cpu contribution corresponding to equal and each virtual machine of the physical cpu information on load that each physical cpu in described virtual environment distributes for same virtual machine is equal, described step c comprises:

-in the following manner, determine the probability of the physical computing resources conflict in described virtual environment:

$y=\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}(L_j\%*V_j)}{(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{T}_j\%)*m*(1-x\%)}$

Wherein, the quantity information of the physical cpu configured in the physical machine of m corresponding to described virtual environment, L _j% is the virtual machine internal cpu load information of each virtual cpu contribution of a jth virtual machine in n virtual machine configuring in described virtual environment within the described sampling period corresponding to it, V _jfor the virtual cpu quantity information that a jth virtual machine has, T _j% be a jth virtual machine within the described sampling period by the physical cpu information on load that each physical cpu distributes, x% is the scheduling overhead information about all virtual machines.

6. method according to any one of claim 1 to 5, wherein, the method also comprises:

-probability that described physical computing resources conflicts is sent to correspondence system, the probability conflicted according to described physical computing resources for described system carries out respective handling.

7. one kind for determining the collision probability determination equipment of the probability of the conflict of physical computing resources in virtual environment, wherein, at least one physical cpu and at least two virtual machines are configured in physical machine corresponding to described virtual environment, virtual machine described in each has at least one virtual cpu, and this collision probability determination equipment comprises:

First acquisition device, for obtaining the physical cpu information on load that virtual machine described in each is assigned with within the sampling period;

Second acquisition device, for obtaining the virtual machine internal cpu load information corresponding within the described sampling period of virtual machine described in each;

Collision probability determining device, for according to the physical cpu information on load of virtual machine described in each and virtual machine internal cpu load information, determines the probability of the physical computing resources conflict in described virtual environment.

8. collision probability determination equipment according to claim 7, wherein, described collision probability determining device is used for:

-according to the physical cpu information on load of virtual machine described in each and virtual machine internal cpu load information, determine the competition cpu load information that corresponding virtual machine is corresponding within the described sampling period;

-according to the physical cpu information on load of virtual machine described in each and competition cpu load information, determine the probability of the physical computing resources conflict in described virtual environment.

9. collision probability determination equipment according to claim 7, wherein, described collision probability determining device is used for:

-according to the physical cpu information on load of virtual machine described in each and virtual machine internal cpu load information, and in conjunction with the quantity information of the physical cpu configured in the physical machine corresponding to described virtual environment, and the virtual cpu quantity information described in each corresponding to virtual machine, determine the probability of the physical computing resources conflict in described virtual environment.

10. collision probability determination equipment according to claim 9, wherein, described collision probability determining device is used for:

-according to the physical cpu information on load of virtual machine described in each and virtual machine internal cpu load information, and in conjunction with the quantity information of the physical cpu configured in the physical machine corresponding to described virtual environment, the virtual cpu quantity information described in each corresponding to virtual machine and the scheduling overhead information about all virtual machines, determine the probability of the physical computing resources conflict in described virtual environment.

11. collision probability determination equipment according to claim 7, wherein, when the virtual machine internal cpu load information of each virtual cpu contribution corresponding to equal and each virtual machine of the physical cpu information on load that each physical cpu in described virtual environment distributes for same virtual machine is equal, described collision probability determining device is used for:

-in the following manner, determine the probability of the physical computing resources conflict in described virtual environment:

$y=\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}(L_j\%*V_j)}{(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{T}_j\%)*m*(1-x\%)}$

Wherein, the quantity information of the physical cpu configured in the physical machine of m corresponding to described virtual environment, L _j% is the virtual machine internal cpu load information of each virtual cpu contribution of a jth virtual machine in n virtual machine configuring in described virtual environment within the described sampling period corresponding to it, V _jfor the virtual cpu quantity information that a jth virtual machine has, T _j% be a jth virtual machine within the described sampling period by the physical cpu information on load that each physical cpu distributes, x% is the scheduling overhead information about all virtual machines.

12. collision probability determination equipment according to any one of claim 7 to 11, wherein, this collision probability determination equipment also comprises:

Dispensing device, is sent to correspondence system for the probability conflicted by described physical computing resources, and the probability conflicted according to described physical computing resources for described system carries out respective handling.