CN102026105B - Method and device for determining central processing unit load - Google Patents

Abstract

The invention discloses a method and a device for determining a central processing unit (CPU) load. The method comprises the following steps of: selecting a service model for determining the CPU load of a radio network controller (RNC) from various service models supported by the RNC; respectively determining the number of users using each service in the selected service model when the RNC supports the maximum user number, and determining the CPU load occupied by one user who uses the service; and determining the CPU load when the RNC supports the maximum user number according to the user number and the corresponding CPU load which are determined on the basis of each service. By using the technical scheme, the problem that a test instrument cannot test the CPU load of the RNC as the test instrument cannot simulate a communication scene when the RNC supports the maximum user number in the prior art is solved.

Description

Determine method and the device of central processing unit load

Technical field

The present invention relates to communication technical field, particularly a kind of method of definite central processing unit load and device.

Background technology

Radio network controller (RNC, Radio Network Controller) be 3-G (Generation Three mobile communication system) (3G, 3rd Generation Mobile Communications System) the main network element in, it is the part of access network, and RNC is mainly used in providing mobile management, call treatment, link management and handover mechanism.

in large-scale festivals or holidays or when holding large-scale activity, because intensive calling appears in the user, send the communication behaviors such as note, may make the central processing unit (CPU of RNC, Central Processing Unit) load is too high, if the cpu load of RNC surpasses 80%, the delay phenomenon of machine of RNC may appear, therefore must test RNC before the RNC access network, the cpu load of main test RNC when capacity is maximum, the capacity of RNC is the number of users that this RNC supports, the cpu load of RNC when capacity is maximum is the cpu load of this RNC when supporting the maximum number of user amount so, for example, the maximum number of user amount that RNC supports is 100,000, need so to test the cpu load of RNC when supporting 100,000 users, if the test obtain this moment cpu load greater than 80%, this RNC enter may load after network too high, therefore can not be with this RNC access network.

when in prior art, RNC being tested, usually use test instrument analog subscriber carries out the scene of voice-and-data business, for example at first select a business model, comprise two kinds of business of short message service and voice call service in this business model, wherein the business proportion that accounts for of short message service is 1/4, the business proportion that voice call service accounts for is 3/4, the maximum number of user amount that this RNC can support is 150,000, this RNC can support at most 50,000 users to carry out short message service so, support simultaneously 100,000 users to carry out voice call service, when this RNC is tested, if can simulating at most 100,000 users, test instrumentation carries out short message service, while, 200,000 users carried out the scene of voice call service, namely the maximum number of user amount that can support of this test instrumentation is 300,000, can use so 50,000 users of this test instrumentation simulation to carry out short message service, while, 100,000 users carried out the scene of voice call service, then test the cpu load of RNC.But along with the raising of hardware technology and the development of equipment, the maximum number of user amount that RNC can support is increasing, and the maximum number of user amount that test instrumentation can be supported is limited, this just exists test instrumentation can not simulate the communication scenes of RNC when supporting the maximum number of user amount, therefore just can not test out the cpu load of RNC this moment.

Summary of the invention

The embodiment of the present invention provides a kind of method and device of definite central processing unit load, can not simulate the communication scenes of RNC when supporting the maximum number of user amount in order to solve the test instrumentation that exists in prior art, therefore just can not test out the problem of the cpu load of RNC this moment.

Embodiment of the present invention technical scheme is as follows:

A kind of method of definite central processing unit load, the method comprising the steps of: in the miscellaneous service model that radio network controller (RNC) is supported, the business model that the central processor CPU of described RNC uses when loading is determined in selection; For every kind of business in the business model of selecting, carry out respectively: determine to use the user's of this business number of users when RNC supports the maximum number of user amount; Determine the cpu load that a user need take when using this business; According to the number of users of determining for every kind of business and corresponding cpu load, determine the cpu load of described RNC when supporting the maximum number of user amount.

A kind of device of definite central processing unit load comprises: selected cell, and for the miscellaneous service model of supporting at radio network controller (RNC), the business model that the central processor CPU of described RNC uses when loading is determined in selection; The first determining unit for every kind of business of the business model of selecting for selected cell, is determined respectively to use the user's of this business number of users when described RNC supports the maximum number of user amount; The second determining unit for every kind of business of the business model of selecting for selected cell, is determined respectively the cpu load that a user need take when using this business; The 3rd determining unit is used for number of users and the corresponding cpu load determined respectively for every kind of business according to the first determining unit and the second determining unit, determines the cpu load of described RNC when supporting the maximum number of user amount.

in embodiment of the present invention technical scheme, if need to utilize test instrumentation tests RNC, namely determine the cpu load of this RNC, at first select required business model in the miscellaneous service model that RNC supports, then for every kind of business in the business model of selecting, determine respectively when this RNC supports the maximum number of user amount, use the user's of this business number of users, and determine the cpu load that need take when a user uses this business, according to the number of users of determining for every kind of business and corresponding cpu load, determine the cpu load of this RNC when supporting the maximum number of user amount, because test instrumentation in the embodiment of the present invention is no longer simulated RNC all communication scenes when supporting the maximum number of user amount, but simulate respectively when RNC supports the maximum number of user amount, the user uses the communication scenes of every kind of business in business model, and then determine the cpu load of this RNC when supporting the maximum number of user amount, this has just solved the problem that can not simulate the cpu load that can not test out RNC this moment that the communication scenes of RNC when supporting the maximum number of user amount cause due to test instrumentation.

Description of drawings

Fig. 1 is in the embodiment of the present invention, determines the method flow diagram of cpu load;

Fig. 2 is in the embodiment of the present invention, determines the apparatus structure schematic diagram of cpu load.

Embodiment

Embodiment of the present invention technical scheme main realized principle, embodiment and the beneficial effect that should be able to reach is at length set forth below in conjunction with each accompanying drawing.

As shown in Figure 1, for determining the method flow diagram of cpu load in the embodiment of the present invention, its concrete processing procedure is as follows:

Step 11, in the miscellaneous service model that RNC supports, the business model of using during the cpu load of definite this RNC of selection;

If will test RNC, namely determine the cpu load of this RNC when supporting the maximum number of user amount, at first in the miscellaneous service model that this RNC supports, the business model of using when selecting test is if the business model of selecting is { S ₁.S ₂..., S _S, comprise S kind business in this business model, be respectively S ₁Business, S ₂Business ..., S _SBusiness, wherein shared business proportion may be different in the miscellaneous service of this business model for every kind of business in this business model, and for example, every kind of business shared business proportion in the miscellaneous service of this business model can be { P ₁.P ₂..., P _S, P ₁Be S ₁Business proportion, P that business is shared ₂Be S ₂The shared business proportion of business ..., P _SBe S _SThe business proportion that business is shared.

Step 12 for every kind of business in the business model of selecting, is determined to use the user's of this business number of users when RNC supports the maximum number of user amount, wherein this process can but be not limited to following:

For (1≤the i≤S), at first obtain the maximum number of user amount N that this RNC can support of the i kind business in this business model _max, then obtain i kind business shared business proportion P in the miscellaneous service of the business model of selecting _i, according to the N that obtains _maxAnd P _i, pass through N _i=N _maxP _i, calculate when RNC supports the maximum number of user amount, use the user's of i kind business number of users N _i

Step 13 for every kind of business in the business model of selecting, is determined the cpu load that a user need take when using this business, wherein this process can but be not limited to:

For (1≤the i≤S), at first obtain the maximum number of user amount C that test instrumentation can be supported for this business of the i kind business in this business model _i, compare C _iAnd N _iWhich value is less, with value less be defined as this test instrumentation required test subscriber's quantity M when the cpu load that definite this business takies _i, i.e. M _i=min{C _i, N _i, then use this test instrumentation to simulate M _iThe communication scenes of individual user when using i kind business determined under this scene the cpu load R that i kind business is shared _i ¹, test subscriber's quantity of then establishing this test instrumentation is 0.9M _i, use this test instrumentation to simulate 0.9M _iThe communication scenes of individual user when using i kind business determined under this scene the cpu load R that i kind business is shared _i ², according to above-mentioned steps, test subscriber's quantity of determining successively test instrumentation is 0.8M _i, 0.7M _i, 0.6M _iThe time, the cpu load R that i kind business is shared _i ³, R _i ⁴, R _i ⁵, to the { (M that obtains _i, R _i ¹), (0.9M _i, R _i ²), (0.8M _i, R _i ³), (0.7M _i, R _i ⁴), (0.6M _i, R _i ⁵) use typical interpolation arithmetic, and obtain using the relational expression of number of users and the cpu load that i kind business need take of i kind business, with the Monomial coefficient in this relational expression, be defined as the cpu load R that need take when a user uses this business _i

Step 14 according to the number of users of determining for every kind of business and corresponding cpu load, is determined the cpu load of this RNC when supporting the maximum number of user amount.

Wherein can but be not limited to by following manner, determine the cpu load of RNC when supporting the maximum number of user amount:

$R_{\max }=\underset{i=1}{\overset{s}{\mathrm{\Σ}}}{R}_i{N}_i$

Wherein, R _maxBe the cpu load of RNC when supporting the maximum number of user amount.

In addition, the further R to determining _maxRevise, be specially:

At first determine the cpu load R when supporting the maximum number of user amount to this RNC _maxRevise required correction value R _Δ, then according to the correction value R that determines _Δ, by following manner, to determining R _maxRevise:

R _maxΔ＝R _max+R _Δ

Wherein, R _{The max Δ}Be the cpu load through revised RNC.

R _ΔCan but be not limited to determine by following manner:

$R_{\mathrm{\Δ}}=N_{\max }^2\underset{i=1}{\overset{s}{\mathrm{\Σ}}}{P}_i{\mathrm{\Φ}}_i+\underset{i=1}{\overset{s}{\mathrm{\Σ}}}{P}_i{\mathrm{\Δ}}_i+N_{\max }^2{P}_{\max 1}{P}_{\max 2}\mathrm{\Ψ}$

Wherein, Φ _iBe that i kind business is to R _ΔInfluence value, be the quadratic term coefficient in the relational expression of the number of users of the use i kind business that calculates and the cpu load that i kind business need take;

Δ _iFor being 0 o'clock at the number of users that uses i kind business, the cpu load that i kind business need take is the constant in the relational expression of the number of users of the use i kind business that calculates and the cpu load that i kind business need take;

P _Max1And P _Max2Be respectively the business proportion of two kinds of business that in this business model, business proportion is the highest;

Ψ is in this business model, and the correlation between the highest two kinds of business of business proportion is to R _ΔInfluence value.

Wherein definite step of Ψ is as follows:

If the maximum number of user amount that test instrumentation can be supported is C _max, use this test instrumentation to simulate C _maxThe scene that individual user communicates wherein has C _maxP _iIndividual user uses i kind business, and the cpu load of determining RNC this moment is R _max ¹, then use successively this test instrumentation to simulate 0.9C _maxIndividual user, 0.8C _maxIndividual user, 0.7C _maxIndividual user, 0.6C _maxThe scene that individual user communicates is determined the cpu load R of RNC successively _max ², R _max ³, R _max ⁴, R _max ⁵, based on $\min \left\{\underset{3}{\overset{5}{\mathrm{\Σ}}}{(R_{\max }^i-C_{\max }^i\underset{i=1}{\overset{S}{\mathrm{\Σ}}}{R}_i{P}_i-R_{\mathrm{\Δ}})}^2\right\},$ Be the Minimum Mean Square Error principle, namely Ψ calculated estimation, wherein C _max ³Be 0.8C _max, C _max ⁴Be 0.7C _max, C _max ⁵Be 0.6C _max, make $\∂\left(\min \right\{\underset{3}{\overset{5}{\mathrm{\Σ}}}{(R_{\max }^i-C_{\max }^i\underset{i=1}{\overset{S}{\mathrm{\Σ}}}{R}_i{P}_i-R_{\mathrm{\Δ}})}^2\left\}\right)/\∂\mathrm{\Ψ}=0,$ Obtain Ψ ₁And Ψ ₂, make Ψ=(Ψ ₁+ Ψ ₂)/2.

by above-mentioned processing procedure as can be known, in embodiment of the present invention technical scheme, if need to utilize test instrumentation tests RNC, namely determine the cpu load of this RNC, at first select required business model in the miscellaneous service model that RNC supports, then for every kind of business in the business model of selecting, determine respectively when this RNC supports the maximum number of user amount, use the user's of this business number of users, and determine the cpu load that need take when a user uses this business, according to the number of users of determining for every kind of business and corresponding cpu load, determine the cpu load of this RNC when supporting the maximum number of user amount, because test instrumentation in the embodiment of the present invention is no longer simulated RNC all communication scenes when supporting the maximum number of user amount, but simulate respectively when RNC supports the maximum number of user amount, the user uses the communication scenes of every kind of business in business model, and then determine the cpu load of this RNC when supporting the maximum number of user amount, this has just solved the problem that can not simulate the cpu load that can not test out RNC this moment that the communication scenes of RNC when supporting the maximum number of user amount cause due to test instrumentation.

Accordingly, the present invention also provides a kind of device of definite cpu load, as shown in Figure 2, comprises selected cell 21, the first determining unit 22, the second determining unit 23 and the 3rd determining unit 24, wherein:

Selected cell 21, for the miscellaneous service model of supporting at RNC, the business model of using during the cpu load of definite this RNC of selection;

The first determining unit 22 for every kind of business of the business model of selecting for selected cell 21, is determined respectively to use the user's of this business number of users when this RNC supports maximum number of user amount;

The second determining unit 23 for every kind of business of the business model of selecting for selected cell 21, is determined respectively the cpu load that a user need take when using this business;

The 3rd determining unit 24 is used for number of users and the corresponding cpu load determined respectively for every kind of business according to the first determining unit 22 and the second determining unit 23, determines the cpu load of this RNC when supporting the maximum number of user amount.

Preferably, the first determining unit 22 comprises that specifically first obtains subelement, the second acquisition subelement and the first computation subunit, wherein first obtain subelement, every kind of business for the business model of selecting for selected cell 21 obtains the maximum number of user amount that this RNC can support; Second obtains subelement, is used for every kind of business of the business model selected for selected cell 21, obtains this business shared business proportion in the miscellaneous service of the business model of selecting; The first computation subunit, every kind of business for the business model of selecting for selected cell 21, the maximum number of user amount and second that can support according to the first RNC that obtains the subelement acquisition obtains the shared business proportion of this business that subelement obtains, when calculating is supported the maximum number of user amount at RNC, use the user's of this business number of users.

Preferably, the second determining unit 23 comprises that specifically the 3rd obtains subelement, first definite subelement, second definite subelement, the second computation subunit and the 3rd definite subelement, wherein the 3rd obtain subelement, for every kind of business of the business model of selecting for selected cell 21, obtain the maximum number of user amount that test instrumentation can be supported for this business; First determines subelement, every kind of business for the business model of selecting for selected cell 21, in first obtained the number of users of the maximum number of user amount that this test instrumentation that subelement obtains can support for this business and the user who uses this business when RNC supports the maximum number of user amount, the number of users that value is less was defined as this test instrumentation required test subscriber's quantity when the cpu load that definite this business takies; Second determines subelement, every kind of business for the business model of selecting for selected cell 21, utilize this test instrumentation, determine that at the number of users that uses this business be first when determining test subscriber's quantity that subelement determines, the cpu load that this business need take; The second computation subunit, every kind of business for the business model of selecting for selected cell 21, determine according to first the cpu load that this business that test subscriber's quantity and second that subelement is determined determines that subelement is determined need take, by interpolation algorithm, calculate the relational expression of number of users and the cpu load that this business need take of this business of use; The 3rd determines subelement, is used for every kind of business of the business model selected for selected cell 21, and the Monomial coefficient in the relational expression that the second computation subunit is calculated is defined as the cpu load that need take when a user uses this business.

More preferably, the 3rd determining unit 24 is determined the cpu load of this RNC when supporting the maximum number of user amount by following manner:

$R_{\max }=\underset{i=1}{\overset{s}{\mathrm{\Σ}}}{R}_i{N}_i$

Wherein, R _maxBe the cpu load of this RNC when supporting the maximum number of user amount;

S is the species number of business in the business model of selecting;

N _iFor when RNC supports the maximum number of user amount, use the user's of i kind business number of users, 1≤i≤S;

R _iIt is the cpu load that a user need take when using i kind business.

More preferably, the device of above-mentioned definite CPU also comprises the 4th determining unit and amending unit, and wherein the 4th determining unit, be used for definite cpu load that the 3rd determining unit 24 is determined and revise required correction value R _ΔAmending unit is used for the correction value R that determines according to the 4th determining unit _Δ, by following manner, the cpu load that the 3rd determining unit 24 is determined is revised:

R _maxΔ＝R _max+R _Δ

Wherein, R _{The max Δ}Be the cpu load through revised RNC.

More preferably, the 4th determining unit is determined the cpu load of RNC when supporting the maximum number of user amount revised required correction value R by following manner _Δ:

$R_{\mathrm{\Δ}}=N_{\max }^2\underset{i=1}{\overset{s}{\mathrm{\Σ}}}{P}_i{\mathrm{\Φ}}_i+\underset{i=1}{\overset{s}{\mathrm{\Σ}}}{P}_i{\mathrm{\Δ}}_i+N_{\max }^2{P}_{\max 1}{P}_{\max 2}\mathrm{\Ψ}$

Wherein, N _maxThe maximum number of user amount that can support for RNC;

P _iIt is i kind business shared business proportion in the miscellaneous service of the business model of selecting;

Φ _iBe that i kind business is to R _ΔInfluence value;

Δ _iFor the number of users in use i kind business is 0 o'clock, the cpu load that i kind business need take;

P _Max1And P _Max2Be respectively in above-mentioned business model the business proportion of two kinds of business that business proportion is the highest;

Ψ is in above-mentioned business model, and the correlation between the highest two kinds of business of business proportion is to R _ΔInfluence value.

More preferably, Φ _iIt is the quadratic term coefficient in the relational expression of the number of users of the use i kind business that calculates of the second computation subunit and the cpu load that i kind business need take; Δ _iBe the constant in the relational expression of the number of users of the use i kind business that calculates and the cpu load that i kind business need take.

Obviously, those skilled in the art can carry out various changes and modification and not break away from the spirit and scope of the present invention the present invention.Like this, if within of the present invention these are revised and modification belongs to the scope of claim of the present invention and equivalent technologies thereof, the present invention also is intended to comprise these changes and modification interior.

Claims (10)

1. the method for a definite central processing unit load, is characterized in that, comprising:

In the miscellaneous service model that radio network controller (RNC) is supported, the business model that the central processor CPU of described RNC uses when loading is determined in selection;

For every kind of business in the business model of selecting, carry out respectively:

Determine to use the user's of this business number of users when RNC supports the maximum number of user amount;

Obtain the maximum number of user amount that test instrumentation can be supported for this business;

In maximum number of user amount that the described test instrumentation that obtains can be supported for this business with use user's the number of users of this business when RNC supports the maximum number of user amount in, the number of users that value is less is defined as described test instrumentation required test subscriber's quantity when the cpu load that definite this business takies;

Utilize described test instrumentation, determine when the number of users that uses this business is described test subscriber's quantity of determining the cpu load that this business need take;

According to described test subscriber's quantity of determining and this business need take when the number of users that uses this business is described test subscriber's quantity of determining cpu load, by interpolation algorithm, calculate the relational expression of number of users and the cpu load that this business need take of this business of use;

With the Monomial coefficient in the relational expression that calculates, be defined as the cpu load that need take when a user uses this business;

According to the number of users of determining for every kind of business and corresponding cpu load, determine the cpu load of described RNC when supporting the maximum number of user amount.

2. the method for definite central processing unit load as claimed in claim 1, is characterized in that, determines to use the user's of this business number of users when RNC supports the maximum number of user amount, specifically comprises:

Obtain the maximum number of user amount that described RNC can support;

Obtain this business shared business proportion in the miscellaneous service of the business model of selecting;

According to maximum number of user amount and the shared business proportion of this business that the described RNC that obtains can support, calculate when RNC supports the maximum number of user amount, use the user's of this business number of users.

3. the method for definite central processing unit load as claimed in claim 2, is characterized in that, by following manner, determines the cpu load of RNC when supporting the maximum number of user amount:

$R_{\max }=\underset{i=1}{\overset{s}{\mathrm{\Σ}}}{R}_i{N}_i$

Wherein, R _maxBe the cpu load of described RNC when supporting the maximum number of user amount;

S is the species number of business in the described business model of selecting;

N _iFor when RNC supports the maximum number of user amount, use the user's of i kind business number of users, 1≤i≤S;

R _iIt is the cpu load that a user need take when using i kind business.

4. the method for definite central processing unit load as claimed in claim 3, is characterized in that, also comprises:

Determine the cpu load of described RNC when supporting the maximum number of user amount revised required correction value R _Δ

According to the correction value R that determines _Δ, by following manner, the cpu load of described RNC when supporting the maximum number of user amount of determining revised:

R _maxΔ=R _max+R _Δ

Wherein, R _{The max Δ}Be the cpu load through revised described RNC.

5. the method for definite central processing unit load as claimed in claim 4, is characterized in that, determines the cpu load of described RNC when supporting the maximum number of user amount revised required correction value by following manner _RΔ:

$R_{\mathrm{\Δ}}=N_{\max }^2\underset{i=1}{\overset{s}{\mathrm{\Σ}}}{P}_i{\mathrm{\Φ}}_i+\underset{i=1}{\overset{s}{\mathrm{\Σ}}}{P}_i{\mathrm{\Δ}}_i+N_{\max }^2{P}_{\max 1}{P}_{\max 2}\mathrm{\Ψ}$

Wherein, N _maxThe maximum number of user amount that can support for described RNC;

P _iIt is i kind business shared business proportion in the miscellaneous service of described business model;

Φ _iBe that i kind business is to R _ΔInfluence value, be the quadratic term coefficient in the relational expression of the number of users of the use i kind business that calculates and the cpu load that i kind business need take;

Δ _iFor being 0 o'clock at the number of users that uses i kind business, the cpu load that i kind business need take is the constant in the relational expression of the number of users of the use i kind business that calculates and the cpu load that i kind business need take;

P _Max1And P _Max2Be respectively in described business model the business proportion of two kinds of business that business proportion is the highest;

Ψ is in described business model, and the correlation between the highest two kinds of business of business proportion is to R _ΔInfluence value.

6. the device of a definite central processing unit load, is characterized in that, comprising:

Selected cell, for the miscellaneous service model of supporting at radio network controller (RNC), the business model that the central processor CPU of described RNC uses when loading is determined in selection;

The first determining unit for every kind of business of the business model of selecting for selected cell, is determined respectively to use the user's of this business number of users when described RNC supports the maximum number of user amount;

The second determining unit for every kind of business of the business model of selecting for selected cell, is determined respectively the cpu load that a user need take when using this business, specifically comprises:

The 3rd obtains subelement, for every kind of business of the business model of selecting for selected cell, obtains the maximum number of user amount that test instrumentation can be supported for this business;

First determines subelement, every kind of business for the business model of selecting for selected cell, in first obtained the number of users of the maximum number of user amount that described test instrumentation that subelement obtains can support for this business and the user who uses this business when RNC supports the maximum number of user amount, the number of users that value is less was defined as described test instrumentation required test subscriber's quantity when the cpu load that definite this business takies;

Second determines subelement, every kind of business for the business model of selecting for selected cell, utilize described test instrumentation, determine that at the number of users that uses this business be first when determining described test subscriber's quantity that subelement determines, the cpu load that this business need take;

The second computation subunit, every kind of business for the business model of selecting for selected cell, determine according to first the cpu load that this business that described test subscriber's quantity and second that subelement is determined determines that subelement is determined need take, by interpolation algorithm, calculate the relational expression of number of users and the cpu load that this business need take of this business of use;

The 3rd determines subelement, is used for every kind of business of the business model selected for selected cell, and the Monomial coefficient in the relational expression that the second computation subunit is calculated is defined as the cpu load that need take when a user uses this business;

The 3rd determining unit is used for number of users and the corresponding cpu load determined respectively for every kind of business according to the first determining unit and the second determining unit, determines the cpu load of described RNC when supporting the maximum number of user amount.

7. the device of definite central processing unit load as claimed in claim 6, is characterized in that, the first determining unit specifically comprises:

First obtains subelement, and every kind of business for the business model of selecting for selected cell obtains the maximum number of user amount that described RNC can support;

Second obtains subelement, is used for every kind of business of the business model selected for selected cell, obtains this business shared business proportion in the miscellaneous service of the business model of selecting;

The first computation subunit, every kind of business for the business model of selecting for selected cell, the maximum number of user amount and second that can support according to the first described RNC that obtains the subelement acquisition obtains the shared business proportion of this business that subelement obtains, when calculating is supported the maximum number of user amount at RNC, use the user's of this business number of users.

8. the device of definite central processing unit load as claimed in claim 7, is characterized in that, described the 3rd determining unit is determined the cpu load of described RNC when supporting the maximum number of user amount by following manner:

$R_{\max }=\underset{i=1}{\overset{s}{\mathrm{\Σ}}}{R}_i{N}_i$

Wherein, R _maxBe the cpu load of described RNC when supporting the maximum number of user amount;

S is the species number of business in the described business model of selecting;

N _iFor when RNC supports the maximum number of user amount, use the user's of i kind business number of users, 1≤i≤S;

R _iIt is the cpu load that a user need take when using i kind business.

9. the device of definite central processing unit load as claimed in claim 8, is characterized in that, also comprises:

The 4th determining unit is used for the definite cpu load of described RNC when supporting the maximum number of user amount that the 3rd determining unit is determined and revises required correction value R _Δ

Amending unit is used for the correction value R that determines according to the 4th determining unit _Δ, by following manner, the cpu load of described RNC when supporting the maximum number of user amount that the 3rd determining unit is determined revised:

R _maxΔ=R _max+R _Δ

Wherein, R _{The max Δ}Be the cpu load through revised described RNC.

10. the device of definite central processing unit load as claimed in claim 9, is characterized in that, the 4th determining unit is determined the cpu load of described RNC when supporting the maximum number of user amount revised required correction value R by following manner _Δ:

$R_{\mathrm{\Δ}}=N_{\max }^2\underset{i=1}{\overset{s}{\mathrm{\Σ}}}{P}_i{\mathrm{\Φ}}_i+\underset{i=1}{\overset{s}{\mathrm{\Σ}}}{P}_i{\mathrm{\Δ}}_i+N_{\max }^2{P}_{\max 1}{P}_{\max 2}\mathrm{\Ψ}$

Wherein, N _maxThe maximum number of user amount that can support for described RNC;

P _iIt is i kind business shared business proportion in the miscellaneous service of described business model;

Φ _iBe that i kind business is to R _ΔInfluence value, be the quadratic term coefficient in the relational expression of the number of users of the use i kind business that calculates of the second computation subunit and the cpu load that i kind business need take;

Δ _iFor being 0 o'clock at the number of users that uses i kind business, the cpu load that i kind business need take is the constant in the relational expression of the number of users of the use i kind business that calculates and the cpu load that i kind business need take;

P _Max1And P _Max2Be respectively in described business model the business proportion of two kinds of business that business proportion is the highest;

Ψ is in described business model, and the correlation between the highest two kinds of business of business proportion is to R _ΔInfluence value.

Images