CN105262623A - Resource allocation method and device for software system based on service - Google Patents

Abstract

The present invention discloses a resource allocation method and device for a software system based on service. The method comprises a step of obtaining different resource configuration plans for the component service of the system, forming an initial resource configuration plan set, wherein each of the resource configuration plans comprises the resource configuration strategy of a component service, a step of dividing each resource configuration plan in the initial resource configuration plan set and forming multiple resource configuration plan subsets, a step of obtaining a target resource configuration plan set according to a resource configuration plan subset with a highest domination level, a step of obtaining a target resource configuration strategy for the component service from a target resource configuration plan set, and a step of allocating resource for the component service according to the target resource configuration strategy. Thus, the target resource configuration strategy can be rapidly determined, and the resource consumed by a service provider is small. In addition, the resource allocation based on multiple targets can be realized, and the flexibility of resource allocation of the component service is improved.

Description

For resource allocation methods and the device of the software systems based on service

Technical field

The present invention relates to computer realm, particularly, relate to a kind of resource allocation methods for the software systems based on service and device.

Background technology

In order to the change of business can be adapted to flexibly, making software systems possess further can the feature such as flexible configuration, dynamic restructuring, increasing distributed software system have employed Enterprise SOA (Service-OrientedArchitecture is called for short SOA).This kind of system built based on SOA is also referred to as the software systems (service-basedsoftwaresystem is called for short SBS) based on service.This type systematic is made up of the control flow of some Component service and specified services execution sequence.Such as, Fig. 1 shows the configuration diagram of an example SBS, and as shown in Figure 1, this SBS system can comprise five Component service, is Component service S1, Component service S2, Component service S3, Component service S4 and Component service S5 respectively.Further, the control flow of specified services execution sequence can be as illustrated by the arrows in fig. 1.

Service-level agreement (servicelevelagreement is called for short SLA) defines service quality (QoS) demand of the system of putting on, such as, and throughput, time of delay, required cost etc.In order to meet given SLA, need the workflow for SBS is corresponding to choose suitable Component service, and for Component service distribute suitable computational resource make its reach SLA require service quality.Therefore, how effectively for SBS distributes calculation resources runs the key of SBS under cloud environment.

At present, the resource allocation methods for SBS is that computational resource corresponding for each Component service is divided into different grain size, and the quality of the Component service finding varigrained resource corresponding.Afterwards, by searching for the different Resourse Distribute amounts quality corresponding with Component service, find the Resourse Distribute amount of Component service and the correspondence thereof that can meet SLA to build SBS example.There is two problems in these class methods:

First, along with the granularity of Resourse Distribute amount and QoS partition is thinner, the time needed for Resourse Distribute amount meeting SLA is searched out larger.Suppose that a SBS is made up of m Component service, the cpu resource of each Component service is divided into k grade, and so searching for the required time is O (m ^k), its complexity is exponential.Obviously, be difficult in a lot of situation solve this problem.

Secondly, qos parameter is normalized to a utility function by current method, and then selects the Resourse Distribute amount of Component service and the service correspondence that utility function can be made maximum.It is afoul mutually that this processing mode have ignored qos parameter, and such as, throughput is higher, and Resourse Distribute amount is larger, and corresponding price is also higher.Therefore, for the solution of mutual afoul qos parameter not existence anduniquess.

Summary of the invention

The object of this invention is to provide a kind of resource allocation methods for the software systems based on service and device, to realize based on multiobject Resourse Distribute.

To achieve these goals, the invention provides a kind of resource allocation methods for the software systems based on service, the method comprises: the multiple different resource allocation program obtaining the Component service for this system, form initial resource configuration plan set, wherein, each resource allocation program comprises the resource allocation strategy of described Component service; Divided according to dominance hierarchy by each resource allocation program in described initial resource configuration plan set, form multiple resource allocation program subset, wherein, the resource allocation program in same resource allocation program subset has identical dominance hierarchy; According to the resource allocation program subset with the highest dominance hierarchy, obtain target resource configuration plan set; The target resource collocation strategy for described Component service is obtained from described target resource configuration plan set; And be described Component service Resources allocation according to described target resource collocation strategy.

Alternatively, described each resource allocation program comprises multiple described resource allocation strategy, and wherein, each resource allocation strategy is for a kind of user type.

Alternatively, described method also comprises: the resource allocation program subset in described basis with the highest dominance hierarchy, after obtaining target resource configuration plan set, and before the described target resource collocation strategy obtained from described target resource configuration plan set for described Component service, judge the cycle-index whether actual cycle number of times equals default, wherein, the initial value of described actual cycle number of times is 0, and described default cycle-index is more than or equal to 1; When described actual cycle number of times is not equal to described default cycle-index, described actual cycle number of times is added 1, and using described target resource configuration plan set after new initial resource configuration plan set, re-execute and described each resource allocation program in described initial resource configuration plan set to be divided according to dominance hierarchy, form multiple resource allocation program subset, until described actual cycle number of times equals described default cycle-index.

Alternatively, described each resource allocation program in described initial resource configuration plan set to be divided according to dominance hierarchy, form multiple resource allocation program subset, comprise: determine the dominance relation between each resource allocation program described, wherein, in described dominance relation, the dominance hierarchy of Zhi Peifang is higher than the dominance hierarchy of subject side; According to described dominance relation, the resource allocation program with identical dominance hierarchy is divided into a resource allocation program subset, wherein, do not arrange mutually between resource allocation program in same resource allocation program subset, further, the arbitrary resource allocation program had in the resource allocation program subset of higher dominance hierarchy can arrange the whole resource allocation programs had in the resource allocation program subset of lower dominance hierarchy.

Alternatively, the described dominance relation determined between each resource allocation program described, comprising: from described initial resource configuration plan set, choose first resource configuration plan and Secondary resource configuration plan; Determine the first parameter set and the second parameter set, wherein, described first parameter set comprises QoS parameter under the resource allocation strategy that described Component service comprises in described first resource configuration plan and resource requirement total amount, and described second parameter set comprises QoS parameter under the resource allocation strategy that described Component service comprises in described Secondary resource configuration plan and resource requirement total amount; The 3rd parameter set is determined according to described first parameter set, and determine the 4th parameter set according to described second parameter set, wherein, described 3rd parameter set comprises the QoS parameter of described Component service under described first resource configuration plan and resource requirement total amount, and described 4th parameter set comprises the QoS parameter of described Component service under described Secondary resource configuration plan and resource requirement total amount; Judge whether described first parameter set and described second parameter set meet the parameter threshold condition preset respectively; When the one in described first parameter set and described second parameter set meets described parameter threshold condition, and another one is not when meeting described parameter threshold condition, the domination of the resource allocation program corresponding to parameter set determining to meet described parameter threshold condition does not meet the resource allocation program corresponding to parameter set of described parameter threshold condition; When described first parameter set and described second parameter set all meet described parameter threshold condition, judge that described 3rd parameter set and described 4th parameter set are satisfied first pre-conditioned or second pre-conditioned, wherein, the described first pre-conditioned parameter of the same race be not inferior to for often kind of parameter in described 3rd parameter set in described 4th parameter set, and at least one parameter in described 3rd parameter set is better than the parameter of the same race in described 4th parameter set; The described second pre-conditioned parameter of the same race be not inferior to for often kind of parameter in described 4th parameter set in described 3rd parameter set, and at least one parameter in described 4th parameter set is better than the parameter of the same race in described 3rd parameter set; When described 3rd parameter set and described 4th parameter set meet described first pre-conditioned time, determine that described Secondary resource configuration plan is arranged in described first resource configuration plan; When described 3rd parameter set and described 4th parameter set meet described second pre-conditioned time, determine that described first resource configuration plan is arranged in described Secondary resource configuration plan; When described 3rd parameter set and described 4th parameter set do not meet described first pre-conditioned and do not meet described second pre-conditioned time, or when described first parameter set and described second parameter set all do not meet described parameter threshold condition, determine not arrange mutually between described first resource configuration plan and described Secondary resource configuration plan; From described initial resource configuration plan set, first resource configuration plan and Secondary resource configuration plan is chosen described in repeating, until determine the dominance relation between each resource allocation program described, wherein, the first resource configuration plan at every turn chosen is entirely not identical with Secondary resource configuration plan.

Alternatively, described each resource allocation program comprises multiple described resource allocation strategy, and wherein, each resource allocation strategy is for a kind of user type; And described first parameter set comprises throughput under each resource allocation strategy that described Component service comprises in described first resource configuration plan, time of delay, required cost and resource requirement total amount; Described second parameter set comprises throughput under each resource allocation strategy that described Component service comprises in described Secondary resource configuration plan, time of delay, required cost and resource requirement total amount; Described 3rd parameter set comprises the throughput of described Component service under described first resource configuration plan, time of delay, required cost and resource requirement total amount; Described 4th parameter set comprises the throughput of described Component service under described Secondary resource configuration plan, time of delay, required cost and resource requirement total amount.

Alternatively, described parameter threshold condition comprises:

${\mathrm{\Σ}}_{l=1}^L{F}_{source}^l\left(X^l\right)\≤C_{source}$

$F_{throughput}^l\left(X^l\right)\≥C_{throughput}^l$

$F_{latency}^l\left(X^l\right)\≤C_{latency}^l$

$F_{\cos t}^l\left(X^l\right)\≤C_{\cos t}^l$

${\mathrm{\Σ}}_{l=1}^L{F}_{\cos t}^l\left(X^l\right)\≤C_{\cos t}$

Wherein, X ^lrepresent l the resource allocation strategy that resource allocation program comprises, this l resource allocation strategy is for l kind user type; L represents user type sum; represent l the resource allocation strategy X that described Component service comprises at resource allocation program ^lunder resource requirement total amount; represent l the resource allocation strategy X that described Component service comprises at resource allocation program ^lunder throughput; represent l the resource allocation strategy X that described Component service comprises at resource allocation program ^lunder time of delay; represent l the resource allocation strategy X that described Component service comprises at resource allocation program ^lunder required cost; C _sourcerepresent the resource requirement total amount threshold value for described system preset; represent the throughput threshold for l kind user type preset; represent threshold value time of delay for l kind user type preset; represent preset for cost threshold value needed for l kind user type; C _costrepresent preset for cost threshold value needed for described system.

Alternatively, described basis has the resource allocation program subset of the highest dominance hierarchy, obtain target resource configuration plan set, comprise: from described have choose multiple resource allocation program the resource allocation program subset of the highest dominance hierarchy and carry out cross and variation, generate multiple derivative resource allocation program; Generated multiple derivative resource allocation program is formed as derivative resource allocation program set; The resource allocation program subset having the highest dominance hierarchy by described and described derivative resource allocation program union of sets collection are formed as described target resource configuration plan set.

Alternatively, the described target resource collocation strategy obtained from described target resource configuration plan set for described Component service, comprise: according to the priority order from high to low of each resource allocation program in described target resource configuration plan set, obtain the resource allocation program alternatively resource allocation program of predetermined quantity; Obtain user type information; From each candidate resource configuration in the works, the resource allocation strategy with described user type information match is filtered out; Obtain user to the preference configuration information of resource distribution; From the resource allocation strategy of described and described user type information match, filter out the resource allocation strategy matched with described preference configuration information, as described target resource collocation strategy.

The present invention also provides a kind of resource allocation device for the software systems based on service, this device comprises: the first acquisition module, for obtaining the multiple different resource allocation program of the Component service for this system, form initial resource configuration plan set, wherein, each resource allocation program comprises the resource allocation strategy of described Component service; Subset division module, for each resource allocation program in described initial resource configuration plan set is divided according to dominance hierarchy, form multiple resource allocation program subset, wherein, the resource allocation program in same resource allocation program subset has identical dominance hierarchy; Second acquisition module, for according to the resource allocation program subset with the highest dominance hierarchy, obtains target resource configuration plan set; 3rd acquisition module, for obtaining the target resource collocation strategy for described Component service from described target resource configuration plan set; And resource distribution module, for being described Component service Resources allocation according to described target resource collocation strategy.

Alternatively, described each resource allocation program comprises multiple described resource allocation strategy, and wherein, each resource allocation strategy is for a kind of user type.

Alternatively, described device also comprises: judge module, for having the resource allocation program subset of the highest dominance hierarchy in described basis, after obtaining target resource configuration plan set, and before the described target resource collocation strategy obtained from described target resource configuration plan set for described Component service, judge the cycle-index whether actual cycle number of times equals default, wherein, the initial value of described actual cycle number of times is 0, and described default cycle-index is more than or equal to 1; Circular flow module, for when described actual cycle number of times is not equal to described default cycle-index, described actual cycle number of times is added 1, and using described target resource configuration plan set after new initial resource configuration plan set, rerun described subset division module, until described actual cycle number of times equals described default cycle-index.

Alternatively, described subset division module comprises: dominance relation determination submodule, and for determining the dominance relation between each resource allocation program described, wherein, in described dominance relation, the dominance hierarchy of Zhi Peifang is higher than the dominance hierarchy of subject side; Subset division submodule, for according to described dominance relation, the resource allocation program with identical dominance hierarchy is divided into a resource allocation program subset, wherein, do not arrange mutually between resource allocation program in same resource allocation program subset, further, the arbitrary resource allocation program had in the resource allocation program subset of higher dominance hierarchy can arrange the whole resource allocation programs had in the resource allocation program subset of lower dominance hierarchy.

Alternatively, described dominance relation determination submodule comprises: resource allocation program chooses unit, for choosing first resource configuration plan and Secondary resource configuration plan from described initial resource configuration plan set; First parameter determination unit, for determining the first parameter set and the second parameter set, wherein, described first parameter set comprises QoS parameter under the resource allocation strategy that described Component service comprises in described first resource configuration plan and resource requirement total amount, and described second parameter set comprises QoS parameter under the resource allocation strategy that described Component service comprises in described Secondary resource configuration plan and resource requirement total amount; Second parameter determination unit, for determining the 3rd parameter set according to described first parameter set, and determine the 4th parameter set according to described second parameter set, wherein, described 3rd parameter set comprises the QoS parameter of described Component service under described first resource configuration plan and resource requirement total amount, and described 4th parameter set comprises the QoS parameter of described Component service under described Secondary resource configuration plan and resource requirement total amount; First judging unit, for judging whether described first parameter set and described second parameter set meet the parameter threshold condition preset respectively; First dominance relation determining unit, for meeting described parameter threshold condition when the one in described first parameter set and described second parameter set, and another one is not when meeting described parameter threshold condition, the domination of the resource allocation program corresponding to parameter set determining to meet described parameter threshold condition does not meet the resource allocation program corresponding to parameter set of described parameter threshold condition; Second judging unit, for when described first parameter set and described second parameter set all meet described parameter threshold condition, judge that described 3rd parameter set and described 4th parameter set are satisfied first pre-conditioned or second pre-conditioned, wherein, the described first pre-conditioned parameter of the same race be not inferior to for often kind of parameter in described 3rd parameter set in described 4th parameter set, and at least one parameter in described 3rd parameter set is better than the parameter of the same race in described 4th parameter set; The described second pre-conditioned parameter of the same race be not inferior to for often kind of parameter in described 4th parameter set in described 3rd parameter set, and at least one parameter in described 4th parameter set is better than the parameter of the same race in described 3rd parameter set; Second dominance relation determining unit, for meet when described 3rd parameter set and described 4th parameter set described first pre-conditioned time, determine that described Secondary resource configuration plan is arranged in described first resource configuration plan; 3rd dominance relation determining unit, for meet when described 3rd parameter set and described 4th parameter set described second pre-conditioned time, determine that described first resource configuration plan is arranged in described Secondary resource configuration plan; 4th dominance relation determining unit, for do not meet when described 3rd parameter set and described 4th parameter set described first pre-conditioned and do not meet described second pre-conditioned time, or when described first parameter set and described second parameter set all do not meet described parameter threshold condition, determine not arrange mutually between described first resource configuration plan and described Secondary resource configuration plan; Circular flow unit, for reruning, described resource allocation program chooses unit, until determine the dominance relation between each resource allocation program described, wherein, the first resource configuration plan at every turn chosen is entirely not identical with Secondary resource configuration plan.

Alternatively, described each resource allocation program comprises multiple described resource allocation strategy, and wherein, each resource allocation strategy is for a kind of user type; And described first parameter set comprises throughput under each resource allocation strategy that described Component service comprises in described first resource configuration plan, time of delay, required cost and resource requirement total amount; Described second parameter set comprises throughput under each resource allocation strategy that described Component service comprises in described Secondary resource configuration plan, time of delay, required cost and resource requirement total amount; Described 3rd parameter set comprises the throughput of described Component service under described first resource configuration plan, time of delay, required cost and resource requirement total amount; Described 4th parameter set comprises the throughput of described Component service under described Secondary resource configuration plan, time of delay, required cost and resource requirement total amount.

Alternatively, described parameter threshold condition comprises:

${\mathrm{\Σ}}_{l=1}^L{F}_{source}^l\left(X^l\right)\≤C_{source}$

$F_{throughput}^l\left(X^l\right)\≥C_{throughput}^l$

$F_{latency}^l\left(X^l\right)\≤C_{latency}^l$

$F_{\cos t}^l\left(X^l\right)\≤C_{\cos t}^l$

${\mathrm{\Σ}}_{l=1}^L{F}_{\cos t}^l\left(X^l\right)\≤C_{\cos t}$

Wherein, X ^lrepresent l the resource allocation strategy that resource allocation program comprises, this l resource allocation strategy is for l kind user type; L represents user type sum; represent l the resource allocation strategy X that described Component service comprises at resource allocation program ^lunder resource requirement total amount; represent l the resource allocation strategy X that described Component service comprises at resource allocation program ^lunder throughput; represent l the resource allocation strategy X that described Component service comprises at resource allocation program ^lunder time of delay; represent l the resource allocation strategy X that described Component service comprises at resource allocation program ^lunder required cost; C _sourcerepresent the resource requirement total amount threshold value for described system preset; represent the throughput threshold for l kind user type preset; represent threshold value time of delay for l kind user type preset; represent preset for cost threshold value needed for l kind user type; C _costrepresent preset for cost threshold value needed for described system.

Alternatively, described second acquisition module comprises: cross and variation submodule, for from described have choose multiple resource allocation program in the resource allocation program subset of the highest dominance hierarchy and carry out cross and variation, generate multiple derivative resource allocation program; Derivative resource allocation program set forms submodule, for generated multiple derivative resource allocation program is formed as derivative resource allocation program set; Target resource configuration plan set forms submodule, is formed as described target resource configuration plan set for the resource allocation program subset that has the highest dominance hierarchy by described and described derivative resource allocation program union of sets collection.

Alternatively, described 3rd acquisition module comprises: candidate resource configuration plan obtains submodule, for the priority order from high to low according to each resource allocation program in described target resource configuration plan set, obtain the resource allocation program alternatively resource allocation program of predetermined quantity; User type acquisition of information submodule, for obtaining user type information; First screening submodule, for from each candidate resource configuration in the works, filters out the resource allocation strategy with described user type information match; Preference configuration information obtains submodule, for obtaining the preference configuration information of user to resource distribution; Second screening submodule, in the resource allocation strategy from described and described user type information match, filters out the resource allocation strategy matched with described preference configuration information, as described target resource collocation strategy.

By technique scheme, can, according to multiple different resource allocation program, determine the target resource collocation strategy of the Component service for SBS fast, and the resource that ISP is consumed be minimum.In addition, by resource allocation methods provided by the invention, can consider from multiple different QoS, seek the multiple resource allocation programs do not arranged mutually meeting SLA, realize based on multiobject Resourse Distribute.Such as, some plans have medium throughput, medium response time, are assigned with medium stock number, and another plan has high-throughput, faster response time, is assigned with more stock number.Like this, according to the actual requirements, the target resource collocation strategy of the Component service for SBS can be determined from the resource allocation program that these are not arranged mutually, and be Component service Resources allocation according to this target resource collocation strategy.Thus, the flexibility of the Resourse Distribute of Component service can be improved.

Other features and advantages of the present invention are described in detail in embodiment part subsequently.

Accompanying drawing explanation

Accompanying drawing is used to provide a further understanding of the present invention, and forms a part for specification, is used from explanation the present invention, but is not construed as limiting the invention with embodiment one below.In the accompanying drawings:

Fig. 1 shows the configuration diagram of an example SBS;

Fig. 2 is the flow chart of the resource allocation methods for SBS according to the embodiment of the present invention;

Fig. 3 shows the configuration diagram of the Component service of the SBS that an exemplary embodiment provides;

Fig. 4 A and Fig. 4 B shows the coded sequence schematic diagram of resource allocation program according to the embodiment of the present invention;

Fig. 5 shows the flow chart of the process of the dominance relation between two resource allocation programs how determining in initial resource configuration plan set according to the embodiment of the present invention;

Fig. 6 shows the flow chart of the process of the target resource configuration plan that how to obtain set according to the embodiment of the present invention;

Fig. 7 shows the flow chart how obtaining the process of the target resource collocation strategy for Component service according to the embodiment of the present invention;

Fig. 8 shows the flow chart of the resource allocation methods for SBS according to another embodiment of the present invention;

Fig. 9 is the block diagram of the resource allocation device for SBS according to the embodiment of the present invention;

Figure 10 is the block diagram of the resource allocation device for SBS according to another embodiment of the present invention;

Figure 11 is the block diagram of the resource allocation device for SBS according to another embodiment of the present invention;

Figure 12 is the block diagram of the resource allocation device for SBS according to another embodiment of the present invention;

Figure 13 is the block diagram of the resource allocation device for SBS according to another embodiment of the present invention.

Embodiment

Below in conjunction with accompanying drawing, the specific embodiment of the present invention is described in detail.Should be understood that, embodiment described herein, only for instruction and explanation of the present invention, is not limited to the present invention.

Fig. 2 is the flow chart of the resource allocation methods for SBS according to the embodiment of the present invention.As shown in Figure 2, the method can comprise:

In step s 201, obtain the multiple different resource allocation program of the Component service for SBS, form initial resource configuration plan set, wherein, each resource allocation program can comprise the resource allocation strategy of Component service.

Fig. 3 shows the configuration diagram of the Component service of the SBS that an exemplary embodiment provides.As shown in Figure 3, supposing that SBS system can comprise five Component service, is Component service S1, Component service S2, Component service S3, Component service S4 and Component service S5 respectively.Each Component service has corresponding some computational resources, and wherein, computational resource can such as include but not limited to: CPU, internal memory etc.By distributing different resources to Component service, each Component service obtains corresponding service quality, such as, and throughput, required cost, time of delay etc.In the present invention, the resource distribution corresponding to service quality is called resource configuration unit.

Each Component service can have several resource configuration unit, and wherein, each resource configuration unit is all in the restriction of the assignable max calculation resource of the physical machine self that Component service is corresponding.Further, each resource configuration unit has corresponding service quality.Such as, for Component service S1 distributes the internal memory of 1 CPU and 1GB, then this Component service can reach the throughput of 1000req/s, and required cost is 10.For each Component service, the corresponding relation between its each resource configuration unit and service quality can rule of thumb data be preset.

In the present invention, resource allocation strategy may be used for representing the selection result for the resource configuration unit of each Component service in SBS.Such as, suppose that a Component service corresponds to some resource configuration unit, if for this Component service selects one of them resource configuration unit to be configured, then this resource configuration unit can be encoded as 1, and can be encoded as 0 for other resource configuration unit of this Component service.Selection and the coding of resource configuration unit can be carried out in such a manner to each Component service, thus a resource allocation strategy for all components service can be formed, and this resource allocation strategy can show as a coded sequence (such as, binary code sequence).It should be noted that when selecting resource configuration unit to formulate resource allocation strategy to Component service, each Component service has and only has a corresponding resource configuration unit to be selected.

In an embodiment of the invention, a resource allocation program can comprise a resource allocation strategy.In this case, the coded sequence of the resource allocation strategy formed is the coded sequence of resource allocation program.Such as, Fig. 4 A shows the coded sequence schematic diagram of the example resources configuration plan according to this execution mode.

Alternatively, in actual use, SBS provides the user of service to be often divided into different user types for it, such as, and the user of gold card level and the user of silver card level.The service quality of these dissimilar users to Component service has different demands.Therefore, in yet another embodiment of the present invention, when formulating resource allocation program, can first for each user type, each Component service is carried out to selection and the coding of resource configuration unit, form the coded sequence of the resource allocation strategy of the Component service for each user type.Afterwards, the coded sequence of the resource allocation strategy of the Component service for each user type is spliced into a coded sequence, to form the coded sequence of resource allocation program.That is, a resource allocation program can comprise multiple resource allocation strategy, and wherein, each resource allocation strategy is for a kind of user type.Such as, Fig. 4 B shows the coded sequence schematic diagram that comprises the resource allocation program of the resource allocation strategy of the Component service for two kinds of user types (gold card, silver card).As shown in Figure 4 B, can comprise two resource allocation strategies in this resource allocation program, wherein, first resource allocation strategy is the user for gold card level, and second resource allocation strategy is the user for silver card level.

By this execution mode, the computational resource meeting all types of users demand separately can be distributed simultaneously, the system of disposing can be used jointly by catergories of user.

Can be by the way, set out in advance multiple (such as, 20) different resource allocation program (can be arranged at random by computing equipment, or be arranged by artificial), these resource allocation programs can form initial resource configuration plan set.

Turn back to Fig. 2 now, next, in step S202, each resource allocation program in initial resource configuration plan set is divided according to dominance hierarchy, form multiple resource allocation program subset, wherein, the resource allocation program in same resource allocation program subset has identical dominance hierarchy.

Particularly, when dividing resource allocation program subset, the dominance relation between each resource allocation program in initial resource configuration plan set can first be determined.Afterwards, according to dominance relation, the resource allocation program with identical dominance hierarchy is divided into a resource allocation program subset, wherein, do not arrange mutually between resource allocation program in same resource allocation program subset, further, the arbitrary resource allocation program had in the resource allocation program subset of higher dominance hierarchy can arrange the whole resource allocation programs had in the resource allocation program subset of lower dominance hierarchy.

In the present invention, the dominance relation between resource allocation program can comprise a resource allocation program and arrange another resource allocation program.The resource allocation program that this dominance relation can show to be in Zhi Peifang is better than being in the resource allocation program of subject side in service quality, thus, the dominance hierarchy being in the resource allocation program of Zhi Peifang is higher than the dominance hierarchy of resource allocation program being in subject side.In addition, the dominance relation between resource allocation program can also comprise between two resource allocation programs does not arrange mutually.This dominance relation can show that two resource allocation programs do not have dividing of quality in service quality, and thus, these two resource allocation programs have identical dominance hierarchy.

Suppose that initial resource configuration plan set expression is P, so according to the dominance hierarchy of each resource allocation program, it can be divided, form multiple resource allocation program subset P ₁, P ₂..., P _a, wherein, a represents the sum of the resource allocation program subset of formation.Wherein, the resource allocation program in same resource allocation program subset can have identical dominance hierarchy, that is, does not arrange mutually between the resource allocation program in same resource allocation program subset.In addition, there is each resource allocation program in the resource allocation program subset of higher dominance hierarchy, all resource allocation programs had in the resource allocation program subset of lower dominance hierarchy can both be arranged.

Fig. 5 shows the flow chart of the process of the dominance relation between two resource allocation programs how determining in initial resource configuration plan set according to the embodiment of the present invention.As shown in Figure 5, determine that the process of dominance relation can comprise the following steps:

In step S501, from initial resource configuration plan set, choose first resource configuration plan and Secondary resource configuration plan.Wherein, this first resource configuration plan and Secondary resource configuration plan are two resource allocation programs not identical arbitrarily in initial resource configuration plan set.

In step S502, determine the first parameter set and the second parameter set, wherein, this first parameter set comprises QoS parameter under the resource allocation strategy that Component service comprises in first resource configuration plan and resource requirement total amount, and this second parameter set comprises QoS parameter under the resource allocation strategy that Component service comprises in Secondary resource configuration plan and resource requirement total amount.

In step S503, the 3rd parameter set is determined according to the first parameter set, and determine the 4th parameter set according to the second parameter set, wherein, 3rd parameter set comprises the QoS parameter of Component service under first resource configuration plan and resource requirement total amount, and the 4th parameter set comprises the QoS parameter of Component service under Secondary resource configuration plan and resource requirement total amount.

As previously mentioned, each resource allocation program in initial resource configuration plan set can comprise multiple resource allocation strategy, and wherein, each resource allocation strategy is for a kind of user type.In this case, the first parameter set can comprise throughput under each resource allocation strategy that Component service comprises in first resource configuration plan, time of delay, required cost and resource requirement total amount; Second parameter set can comprise throughput under each resource allocation strategy that Component service comprises in Secondary resource configuration plan, time of delay, required cost and resource requirement total amount; 3rd parameter set can comprise the throughput of Component service under first resource configuration plan, time of delay, required cost and resource requirement total amount; And the 4th parameter set can comprise the throughput of Component service under Secondary resource configuration plan, time of delay, required cost and resource requirement total amount.

In step S504, judge whether the first parameter set and the second parameter set meet the parameter threshold condition preset respectively.

In an example embodiment, shown parameter threshold condition can comprise:

$\begin{array}{c}{\mathrm{\Σ}}_{l=1}^L{F}_{source}^l\left(X^l\right)\≤C_{source}\\ F_{throughput}^l\left(X^l\right)\≥C_{throughput}^l\\ F_{latency}^l\left(X^l\right)\≤C_{latency}^l\\ F_{\cos t}^l\left(X^l\right)\≤C_{\cos t}^l\\ {\mathrm{\Σ}}_{l=1}^L{F}_{\cos t}^l\left(X^l\right)\≤C_{\cos t}\end{array}---\left(1\right)$

Wherein, X ^lbe a vector, it represents l the resource allocation strategy that resource allocation program comprises, and this l resource allocation strategy is for l kind user type; L represents user type sum; represent l the resource allocation strategy X that Component service comprises at resource allocation program ^lunder resource requirement total amount; represent l the resource allocation strategy X that Component service comprises at resource allocation program ^lunder throughput; represent l the resource allocation strategy X that Component service comprises at resource allocation program ^lunder time of delay; represent l the resource allocation strategy X that Component service comprises at resource allocation program ^lunder required cost; C _sourcerepresent the resource requirement total amount threshold value for described system preset; represent the throughput threshold for l kind user type preset; represent threshold value time of delay for l kind user type preset; represent preset for cost threshold value needed for l kind user type; C _costrepresent preset for cost threshold value needed for described system.

Wherein, can be determined by following equation (2) ~ (5) with

$F_{source}^l\left(X^l\right)={\mathrm{\Σ}}_{j=1}^{\left|N\right|}{\mathrm{\Σ}}_{k=1}^{\left|M\right|}{x}_{jk}^l*{\mathrm{cs}}_{jk}---\left(2\right)$

$F_{\cos t}^l\left(X^l\right)={\mathrm{\Σ}}_{j=1}^{\left|N\right|}{\mathrm{\Σ}}_{k=1}^{\left|M\right|}{x}_{jk}^l*{\mathrm{cost}}_{jk}---\left(3\right)$

$F_{latency}^l\left(X^l\right)={\mathrm{\Σ}}_{j=1}^{\left|N^{\′}\right|}{\mathrm{\Σ}}_{k=1}^{\left|M\right|}{x}_{jk}^l*{\mathrm{lat}}_{jk}---\left(4\right)$

$F_{throughput}^l\left(X^l\right)={\mathrm{Min}}_{j\∈N,k\∈M}(x_{jk}^l*{\mathrm{thr}}_{jk})---\left(5\right)$

Wherein, cs _jk, cost _jk, lat _jkand thr _jkrepresent the total resources of a kth resource configuration unit of a jth Component service, required cost, time of delay and throughput respectively, these four parameters can be obtained for the resource configuration unit preset of a jth Component service and the corresponding relation of service quality by inquiry.N and M represents the quantity of Component service in SBS respectively, and the quantity of the resource configuration unit of Component service; N ' represents the quantity of the Component service that the critical path (path that time of delay is the longest) of SBS comprises; Vector X ^lin element representation be wherein, j represents a jth Component service in SBS, and k represents the kth resource configuration unit that this jth Component service is corresponding, and wherein, if having selected this kth resource configuration unit, then value be 1, otherwise be 0.

Can be found out by equation (1), parameter threshold term restriction is for often kind of user type, and the service quality (that is, throughput, time of delay, required cost) of Component service will meet quality of service requirement corresponding to this user type, such as, throughput can not lower than the throughput threshold for this user type l preset should be understood that, for user type not of the same race, with can be different.

In addition, this parameter threshold condition also defines in a resource allocation program, the resource requirement total amount of system (that is, Component service this resource allocation program comprise for the resource requirement total amount sum under the resource allocation strategy of various user type) the default resource requirement total amount threshold value C for this system can not be exceeded _source, and the required cost of system (that is, Component service this resource allocation program comprise for the resource allocation strategy of various user type under needed for cost sum) can not exceed default for cost threshold value C needed for this system _cost, thus enable this resource allocation program meet SLA requirement.

When a certain parameter set meets parameter threshold condition, resource allocation program corresponding to this parameter set can be called as " feasible solution ", and when a certain parameter set does not meet parameter threshold condition, the resource allocation program corresponding to this parameter set can be called as " infeasible solution ".

In step S505, when the one in the first parameter set and the second parameter set meets parameter threshold condition, and another one is not when meeting parameter threshold condition, determine that the resource allocation program domination met corresponding to the parameter set of parameter threshold condition does not meet the resource allocation program corresponding to parameter set of parameter threshold condition.That is, in the present invention, feasible solution domination infeasible solution.

Such as, suppose that the first parameter set meets parameter threshold condition, and the second parameter set does not meet parameter threshold condition, then determine first resource configuration plan domination Secondary resource configuration plan.

In step S506, when the first parameter set and the second parameter set all do not meet parameter threshold condition, determine not arrange mutually between first resource configuration plan and Secondary resource configuration plan.

In step s 507, when the first parameter set and the second parameter set all meet parameter threshold condition, judge that the 3rd parameter set and the 4th parameter set are satisfied first pre-conditioned or second pre-conditioned, wherein, this first pre-conditioned parameter of the same race being often kind of parameter in the 3rd parameter set and not being inferior in the 4th parameter set, and at least one parameter in the 3rd parameter set is better than the parameter of the same race in the 4th parameter set; This second pre-conditioned parameter of the same race being often kind of parameter in the 4th parameter set and not being inferior in the 3rd parameter set, and at least one parameter in the 4th parameter set is better than the parameter of the same race in the 3rd parameter set.

Such as, pre-conditionedly can be expressed as following equation (6) by first, and pre-conditionedly be expressed as following equation (7) by second:

$\left[\begin{array}{c}Thr\left(p_i\right)\&GreaterEqual;Thr\left(p_j\right)and\\ Cost\left(p_i\right)\&le;Cost\left(p_j\right)and\\ Lat\left(p_i\right)\&le;Lat\left(p_j\right)and\\ Source\left(p_i\right)\&le;Source\left(p_j\right)\end{array}\right]and\left[\begin{array}{c}Thr\left(p_i\right)>Thr\left(p_j\right)or\\ Cost\left(p_i\right)<Cost\left(p_j\right)or\\ Lat\left(p_i\right)<Lat\left(p_j\right)or\\ Source\left(p_i\right)<Source\left(p_j\right)\end{array}\right]=1---\left(6\right)$

$\left[\begin{array}{c}Thr\left(p_j\right)\&GreaterEqual;Thr\left(p_i\right)and\\ Cost\left(p_j\right)\&le;Cost\left(p_i\right)and\\ Lat\left(p_j\right)\&le;Lat\left(p_i\right)and\\ Source\left(p_j\right)\&le;Source\left(p_i\right)\end{array}\right]and\left[\begin{array}{c}Thr\left(p_j\right)>Thr\left(p_i\right)or\\ Cost\left(p_j\right)<Cost\left(p_i\right)or\\ Lat\left(p_j\right)<Lat\left(p_i\right)or\\ Source\left(p_j\right)<Source\left(p_i\right)\end{array}\right]=1---\left(7\right)$

Wherein, p _irepresent first resource configuration plan; p _jrepresent Secondary resource configuration plan; Thr (p _i) representing the throughput of Component service under first resource configuration plan, this throughput can be that Component service is at first resource configuration plan p _icomprise for the throughput sum under the resource allocation strategy of various user type; Thr (p _j) representing the throughput of Component service under Secondary resource configuration plan, this throughput can be that Component service is at Secondary resource configuration plan p _jcomprise for the throughput sum under the resource allocation strategy of various user type; Cost (p _i) representing Component service cost needed under first resource configuration plan, this required cost can be that Component service is at first resource configuration plan p _icomprise for cost sum needed under the resource allocation strategy of various user type; Cost (p _j) representing Component service cost needed under Secondary resource configuration plan, this required cost can be that Component service is at Secondary resource configuration plan p _jcomprise for cost sum needed under the resource allocation strategy of various user type; Lat (p _i) representing that Component service configures the time of delay under planning at first resource, this time of delay can be that Component service is at first resource configuration plan p _icomprise for sum time of delay under the resource allocation strategy of various user type; Lat (p _j) representing that Component service configures the time of delay under planning at Secondary resource, this time of delay can be that Component service is at Secondary resource configuration plan p _jcomprise for sum time of delay under the resource allocation strategy of various user type; Source (p _i) representing the resource requirement total amount of Component service under first resource configuration plan, this resource requirement total amount can be that Component service is at first resource configuration plan p _icomprise for the resource requirement total amount sum under the resource allocation strategy of various user type; Source (p _j) representing the resource requirement total amount of Component service under Secondary resource configuration plan, this resource requirement total amount can be that Component service is at Secondary resource configuration plan p _jcomprise for the resource requirement total amount sum under the resource allocation strategy of various user type.

In step S508, when the 3rd parameter set and the 4th parameter set meet first pre-conditioned time, determine first resource configuration plan domination Secondary resource configuration plan.

In step S509, when the 3rd parameter set and the 4th parameter set meet second pre-conditioned time, determine Secondary resource configuration plan domination first resource configuration plan.

That is, in the present invention, when two resource allocation programs are feasible solution, the parameter which parameter set comprises is more excellent, and corresponding resource allocation program arranges another resource allocation program.

In step S510, when the 3rd parameter set and the 4th parameter set meet first pre-conditioned and meet second pre-conditioned time, determine not arrange mutually between first resource configuration plan and Secondary resource configuration plan.

Said process (that is, step S501 is to step S510) can repeat, until determine the dominance relation between each resource allocation program, wherein, the first resource configuration plan at every turn chosen is entirely not identical with Secondary resource configuration plan.

Turn back to Fig. 2 now, after marking off resource allocation program subset, in step S203, according to the resource allocation program subset with the highest dominance hierarchy, obtain target resource configuration plan set.

Fig. 6 shows the flow chart of the process of the target resource configuration plan that how the to obtain set according to an example embodiment of the present invention.As shown in Figure 6, this step S203 can comprise:

In step s 601, from the resource allocation program subset with the highest dominance hierarchy, choose multiple resource allocation program carry out cross and variation, generate multiple derivative resource allocation program.

Such as, suppose that the resource allocation program subset with the highest dominance hierarchy is expressed as P ₁, so can first from P ₁in choose (such as, random selecting) two resource allocation programs and carry out cross and variation, generate two derivative resource allocation programs.Afterwards, can judge whether the quantity of generated derivative resource allocation program reaches the quantity of default derivative resource allocation program.If do not arrived, then repeat from P ₁in choose the operation that two resource allocation programs carry out cross and variation, until the quantity of derivative resource allocation program generated reaches the quantity of default derivative resource allocation program, wherein, at every turn from P ₁in two resource allocation programs choosing entirely not identical.

In step S602, added to by generated multiple derivative resource allocation program in derivative resource allocation program set, wherein, this derivative resource allocation program set is initially empty set.

In step S603, the resource allocation program subset and derivative resource allocation program union of sets collection with the highest dominance hierarchy are formed as target resource configuration plan set.

This process is based on the selection in genetic algorithm, intersection, mutation operation, new individuality is generated (namely from some resource allocation programs (can be described as " individuality ") that the resource allocation program subset with the highest dominance hierarchy comprises, derivative resource allocation program), thus target resource configuration plan set can be expanded, promote the chance of seeking more excellent solution.

Turn back to Fig. 2 now, after getting target resource configuration plan set, in step S204, from target resource configuration plan set, obtain the target resource collocation strategy for Component service; And in step S205, be Component service Resources allocation according to target resource collocation strategy.

As previously mentioned, each resource allocation program in initial resource configuration plan set can comprise multiple resource allocation strategy, and wherein, each resource allocation strategy is for a kind of user type.In this case, such as can arrange the preference of resource distribution based on user type and user, obtain the target resource collocation strategy of the Component service for this user.

Particularly, Fig. 7 shows the flow chart how obtaining the process of the target resource collocation strategy for Component service according to this execution mode.As shown in Figure 7, this step S204 can comprise:

In step s 701, according to the priority order from high to low of each resource allocation program in target resource configuration plan set, the resource allocation program alternatively resource allocation program of predetermined quantity is obtained.

In the present invention, the priority of resource allocation program can be determined in the following manner:

Suppose that two resource allocation programs are represented as p respectively _xand p _y, so:

(1) if p _xfor feasible solution, p _yfor infeasible solution, so p _x> _rp _y;

(2) if p _xand p _ybe feasible solution, and p _xdominance hierarchy higher than p _ydominance hierarchy, so p _x> _rp _y;

(3) if p _xand p _ybe feasible solution, p _xand p _ydominance hierarchy identical, and crowding_distance [p _x] >crowding_distance [p _y], so p _x> _rp _y.

Wherein, p _x> _rp _yrepresent p _xpriority be greater than p _ypriority; Crowding_distance [p _x] represent p _xcrowding distance; Crowding_distance [p _y] represent p _ycrowding distance.

In step S702, obtain user type information.Such as, user is when signing in system, and system can according to the user ID of user, comes, from the mapping relations of the user ID preset and user type, to inquire the user type information that this user ID is corresponding.

In step S703, from each candidate resource configuration in the works, filter out the resource allocation strategy with user type information match.Suppose that candidate resource configuration plan comprises p ₁, p ₂, p ₃, p ₄, wherein, every bar candidate resource configuration plan can comprise for the resource allocation strategy of gold card user and the resource allocation strategy for silver card user.Suppose that getting user type information is gold card user, then can travel through candidate resource configuration plan p ₁, p ₂, p ₃, p ₄, filter out each candidate resource and configure the resource allocation strategy for gold card user comprised in the works.

In step S704, obtain user to the preference configuration information of resource distribution.

In one embodiment, can provide the preference of resource distribution that interface is set to user, user by carrying out on the surface editing, the operation such as click, its preference configuration information to resource distribution can be inputted.Afterwards, system can get the preference configuration information of user's input.

Or, in another embodiment, by directly the resource allocation strategy with user type information match being shown to user, in the mode selected by user, the preference configuration information of user to resource distribution can be obtained.

In step S705, from the resource allocation strategy of user type information match, filter out the resource allocation strategy that matches with preference configuration information as target resource collocation strategy.

Such as, suppose user preference configuration information instruction require that throughput is maximum, so, in step S705, just can from the resource allocation strategy of user type information match choose the maximum resource allocation strategy of throughput as target resource collocation strategy.Again such as, suppose that the preference configuration information instruction of user requires that required cost is minimum, so, in step S705, just can from the resource allocation strategy of user type information match choose the minimum resource allocation strategy of required cost as target resource collocation strategy, etc.

By this execution mode, can to different user types, the user with different hobby provides the service of customization, thus can meet the user demand of different user, greatly improves Consumer's Experience.

Thus, by resource allocation methods provided by the invention, can, according to multiple different resource allocation program, determine the target resource collocation strategy of the Component service for SBS fast, and the resource that ISP is consumed be minimum.In addition, by resource allocation methods provided by the invention, can consider from multiple different QoS, seek the multiple resource allocation programs do not arranged mutually meeting SLA, realize based on multiobject Resourse Distribute.Such as, some plans have medium throughput, medium response time, are assigned with medium stock number, and another plan has high-throughput, faster response time, is assigned with more stock number.Like this, according to the actual requirements, the target resource collocation strategy of the Component service for SBS can be determined from the resource allocation program that these are not arranged mutually, and be Component service Resources allocation according to this target resource collocation strategy.Thus, the flexibility of the Resourse Distribute of Component service can be improved.

Fig. 8 shows the flow chart of the resource allocation methods for SBS according to another embodiment of the present invention.As shown in Figure 8, the method can also comprise:

In step S206 after step S203 and before step S204, judge the cycle-index whether actual cycle number of times equals default, wherein, the initial value of actual cycle number of times is 0, and the cycle-index preset is more than or equal to 1.

In step S207, when actual cycle number of times is not equal to default cycle-index, actual cycle number of times is added 1, and using the target resource configuration plan set that gets after step S203 after new initial resource configuration plan set, return step S202, described each resource allocation program in initial resource configuration plan set to be divided according to dominance hierarchy to re-execute, form multiple resource allocation program subset, and follow-up step, until actual cycle number of times equals default cycle-index.

The loop iteration process of this execution mode is based on genetic algorithm, and wherein, each resource allocation program can be regarded as body one by one.In this genetic algorithm, the population (namely target resource configuration plan set) that every generation can be obtained is as follow-on initial population (namely initial resource configuration plan set), and again carry out dominance hierarchy division, to determine the resource allocation program with the highest dominance hierarchy from new initial population, and generate new target resource configuration plan set with this.

In addition, have in the resource allocation program subset of the highest dominance hierarchy and save the solution in last iterative process with the highest dominance hierarchy, the solution produced with next iteration (namely, derivative resource allocation program set) to compare, the priority that the resource allocation program subset with the highest dominance hierarchy comprises solution might not be lower than the solution of derivative resource allocation program set.Therefore, in a preferred embodiment of the invention, in current iteration, remain the solution of the resource allocation program subset with the highest dominance hierarchy, and will there is the input as next iteration of the resource allocation program subset of the highest dominance hierarchy and derivative resource allocation program union of sets collection.

Thus, according to the principle of the survival of the fittest and the survival of the fittest, can develop by generation and produce the approximate solution of becoming better and better, thus the optimal resource allocation of Component service can be realized.In addition, the time searched required for optimal solution is O (NP × η), and wherein NP is the number of times (that is, the cycle-index preset) of the overall iteration of algorithm, and η is the number of times that algorithm performs cross and variation.Therefore, for existing method, can significantly reduce the resource searching time, promote the efficiency of resource searching.

Fig. 9 is the block diagram of the resource allocation device for the software systems based on service according to the embodiment of the present invention.As shown in Figure 9, this device can comprise: the first acquisition module 901, for obtaining the multiple different resource allocation program of the Component service for this system, forms initial resource configuration plan set, wherein, each resource allocation program comprises the resource allocation strategy of Component service; Subset division module 902, for each resource allocation program in initial resource configuration plan set is divided according to dominance hierarchy, form multiple resource allocation program subset, wherein, the resource allocation program in same resource allocation program subset has identical dominance hierarchy; Second acquisition module 903, for according to the resource allocation program subset with the highest dominance hierarchy, obtains target resource configuration plan set; 3rd acquisition module 904, for obtaining the target resource collocation strategy for Component service from target resource configuration plan set; And resource distribution module 905, for being Component service Resources allocation according to target resource collocation strategy.

Alternatively, each resource allocation program comprises multiple resource allocation strategy, and wherein, each resource allocation strategy is for a kind of user type.

Figure 10 is the block diagram of the resource allocation device for the software systems based on service according to another embodiment of the present invention.As shown in Figure 10, this device can also comprise: judge module 1001, for having the resource allocation program subset of the highest dominance hierarchy in basis, after obtaining target resource configuration plan set, and before obtain the target resource collocation strategy for Component service from target resource configuration plan set, judge the cycle-index whether actual cycle number of times equals default, wherein, the initial value of this actual cycle number of times is 0, and this cycle-index preset is more than or equal to 1; Circular flow module 1002, for when actual cycle number of times is not equal to described default cycle-index, actual cycle number of times is added 1, and using target resource configuration plan set after new initial resource configuration plan set, rerun subset division module 902, until actual cycle number of times equals default cycle-index.

Figure 11 is the block diagram of the resource allocation device for the software systems based on service according to another embodiment of the present invention.As shown in figure 11, subset division module 902 can comprise: dominance relation determination submodule 1101, for determining the dominance relation between each resource allocation program described, wherein, in dominance relation, the dominance hierarchy of Zhi Peifang is higher than the dominance hierarchy of subject side; Subset division submodule 1102, for according to described dominance relation, the resource allocation program with identical dominance hierarchy is divided into a resource allocation program subset, wherein, do not arrange mutually between resource allocation program in same resource allocation program subset, further, the arbitrary resource allocation program had in the resource allocation program subset of higher dominance hierarchy can arrange the whole resource allocation programs had in the resource allocation program subset of lower dominance hierarchy.

Alternatively, this dominance relation determination submodule 1101 can comprise:

Resource allocation program chooses unit, for choosing first resource configuration plan and Secondary resource configuration plan from initial resource configuration plan set;

First parameter determination unit, for determining the first parameter set and the second parameter set, wherein, this first parameter set comprises QoS parameter under the resource allocation strategy that Component service comprises in first resource configuration plan and resource requirement total amount, and this second parameter set comprises QoS parameter under the resource allocation strategy that Component service comprises in Secondary resource configuration plan and resource requirement total amount;

Second parameter determination unit, for determining the 3rd parameter set according to the first parameter set, and determine the 4th parameter set according to the second parameter set, wherein, 3rd parameter set comprises the QoS parameter of Component service under first resource configuration plan and resource requirement total amount, and the 4th parameter set comprises the QoS parameter of Component service under Secondary resource configuration plan and resource requirement total amount;

First judging unit, for judging whether the first parameter set and the second parameter set meet the parameter threshold condition preset respectively;

First dominance relation determining unit, for meeting parameter threshold condition when the one in the first parameter set and the second parameter set, and another one is not when meeting parameter threshold condition, determine that the resource allocation program domination met corresponding to the parameter set of parameter threshold condition does not meet the resource allocation program corresponding to parameter set of parameter threshold condition;

Second judging unit, for when the first parameter set and the second parameter set all meet parameter threshold condition, judge that the 3rd parameter set and the 4th parameter set are satisfied first pre-conditioned or second pre-conditioned, wherein, this first pre-conditioned parameter of the same race being often kind of parameter in the 3rd parameter set and not being inferior in the 4th parameter set, and at least one parameter in the 3rd parameter set is better than the parameter of the same race in the 4th parameter set; This second pre-conditioned parameter of the same race being often kind of parameter in the 4th parameter set and not being inferior in the 3rd parameter set, and at least one parameter in the 4th parameter set is better than the parameter of the same race in the 3rd parameter set;

Second dominance relation determining unit, for meet when the 3rd parameter set and the 4th parameter set first pre-conditioned time, determine first resource configuration plan domination Secondary resource configuration plan;

3rd dominance relation determining unit, for meet when the 3rd parameter set and the 4th parameter set second pre-conditioned time, determine Secondary resource configuration plan domination first resource configuration plan;

4th dominance relation determining unit, for when the 3rd parameter set and the 4th parameter set meet first pre-conditioned and meet second pre-conditioned time, or when the first parameter set and the second parameter set all do not meet parameter threshold condition, determine not arrange mutually between first resource configuration plan and Secondary resource configuration plan;

Circular flow unit, for reruning, resource allocation program chooses unit, until determine the dominance relation between each resource allocation program described, wherein, the first resource configuration plan at every turn chosen is entirely not identical with Secondary resource configuration plan.

Figure 12 is the block diagram of the resource allocation device for the software systems based on service according to another embodiment of the present invention.As shown in figure 12, the second acquisition module 903 can comprise: cross and variation submodule 1201, carrying out cross and variation, generating multiple derivative resource allocation program for choosing multiple resource allocation program from the resource allocation program subset with the highest dominance hierarchy; Derivative resource allocation program adds submodule 1202, and for adding in derivative resource allocation program set by generated multiple derivative resource allocation program, wherein, this derivative resource allocation program set is initially empty set; Target resource configuration plan set forms submodule 1203, for the resource allocation program subset and derivative resource allocation program union of sets collection with the highest dominance hierarchy are formed as target resource configuration plan set.

Figure 13 is the block diagram of the resource allocation device for the software systems based on service according to another embodiment of the present invention.As shown in figure 13,3rd acquisition module 904 can comprise: candidate resource configuration plan obtains submodule 1301, for the priority order from high to low according to each resource allocation program in target resource configuration plan set, obtain the resource allocation program alternatively resource allocation program of predetermined quantity; User type acquisition of information submodule 1302, for obtaining user type information; First screening submodule 1303, for from each candidate resource configuration in the works, filters out the resource allocation strategy with user type information match; Preference configuration information obtains submodule 1304, for obtaining the preference configuration information of user to resource distribution; Second screening submodule 1305, for from the resource allocation strategy of user type information match, filters out the resource allocation strategy matched with preference configuration information, as target resource collocation strategy.

About the device in the various embodiments described above, wherein the concrete mode of modules executable operations has been described in detail in about the embodiment of the method, will not elaborate explanation herein.

Below the preferred embodiment of the present invention is described in detail by reference to the accompanying drawings; but; the present invention is not limited to the detail in above-mentioned execution mode; within the scope of technical conceive of the present invention; can carry out multiple simple variant to technical scheme of the present invention, these simple variant all belong to protection scope of the present invention.

In addition, although describe the operation of the inventive method in the accompanying drawings with particular order, this is not that requirement or hint must perform these operations according to this particular order, or must perform the result that all shown operation could realize expectation.Additionally or alternatively, some step can be omitted, multiple step be merged into a step and perform, and/or a step is decomposed into multiple step and perform.

One skilled in the art will appreciate that embodiments of the present invention can be implemented as a kind of system, device, equipment, method or computer program.Therefore, embodiments of the present invention can be implemented as following form, that is: hardware, completely software (comprising firmware, resident software, microcode etc.) completely, or the form that hardware and software combines.

It should be noted that in addition, each the concrete technical characteristic described in above-mentioned embodiment, in reconcilable situation, can be combined by any suitable mode.In order to avoid unnecessary repetition, the present invention illustrates no longer separately to various possible compound mode.

In addition, also can carry out combination in any between various different execution mode of the present invention, as long as it is without prejudice to thought of the present invention, it should be considered as content disclosed in this invention equally.

Claims (18)

1., for a resource allocation methods for the software systems based on service, it is characterized in that, the method comprises:

Obtain the multiple different resource allocation program for the Component service of this system, form initial resource configuration plan set, wherein, each resource allocation program comprises the resource allocation strategy of described Component service;

Divided according to dominance hierarchy by each resource allocation program in described initial resource configuration plan set, form multiple resource allocation program subset, wherein, the resource allocation program in same resource allocation program subset has identical dominance hierarchy;

According to the resource allocation program subset with the highest dominance hierarchy, obtain target resource configuration plan set;

The target resource collocation strategy for described Component service is obtained from described target resource configuration plan set; And

Be described Component service Resources allocation according to described target resource collocation strategy.

2. method according to claim 1, is characterized in that, described each resource allocation program comprises multiple described resource allocation strategy, and wherein, each resource allocation strategy is for a kind of user type.

3. method according to claim 1, is characterized in that, described method also comprises:

There is in described basis the resource allocation program subset of the highest dominance hierarchy, after obtaining target resource configuration plan set, and before the described target resource collocation strategy obtained from described target resource configuration plan set for described Component service, judge the cycle-index whether actual cycle number of times equals default, wherein, the initial value of described actual cycle number of times is 0, and described default cycle-index is more than or equal to 1;

When described actual cycle number of times is not equal to described default cycle-index, described actual cycle number of times is added 1, and using described target resource configuration plan set after new initial resource configuration plan set, re-execute and described each resource allocation program in described initial resource configuration plan set to be divided according to dominance hierarchy, form multiple resource allocation program subset, until described actual cycle number of times equals described default cycle-index.

4. the method according to any one of claim 1-3, is characterized in that, is describedly divided according to dominance hierarchy by each resource allocation program in described initial resource configuration plan set, forms multiple resource allocation program subset, comprising:

Determine the dominance relation between each resource allocation program described, wherein, in described dominance relation, the dominance hierarchy of Zhi Peifang is higher than the dominance hierarchy of subject side;

According to described dominance relation, the resource allocation program with identical dominance hierarchy is divided into a resource allocation program subset, wherein, do not arrange mutually between resource allocation program in same resource allocation program subset, further, the arbitrary resource allocation program had in the resource allocation program subset of higher dominance hierarchy can arrange the whole resource allocation programs had in the resource allocation program subset of lower dominance hierarchy.

5. method according to claim 4, is characterized in that, the described dominance relation determined between each resource allocation program described, comprising:

First resource configuration plan and Secondary resource configuration plan is chosen from described initial resource configuration plan set;

Determine the first parameter set and the second parameter set, wherein, described first parameter set comprises QoS parameter under the resource allocation strategy that described Component service comprises in described first resource configuration plan and resource requirement total amount, and described second parameter set comprises QoS parameter under the resource allocation strategy that described Component service comprises in described Secondary resource configuration plan and resource requirement total amount;

The 3rd parameter set is determined according to described first parameter set, and determine the 4th parameter set according to described second parameter set, wherein, described 3rd parameter set comprises the QoS parameter of described Component service under described first resource configuration plan and resource requirement total amount, and described 4th parameter set comprises the QoS parameter of described Component service under described Secondary resource configuration plan and resource requirement total amount;

Judge whether described first parameter set and described second parameter set meet the parameter threshold condition preset respectively;

When the one in described first parameter set and described second parameter set meets described parameter threshold condition, and another one is not when meeting described parameter threshold condition, the domination of the resource allocation program corresponding to parameter set determining to meet described parameter threshold condition does not meet the resource allocation program corresponding to parameter set of described parameter threshold condition;

When described first parameter set and described second parameter set all meet described parameter threshold condition, judge that described 3rd parameter set and described 4th parameter set are satisfied first pre-conditioned or second pre-conditioned, wherein, the described first pre-conditioned parameter of the same race be not inferior to for often kind of parameter in described 3rd parameter set in described 4th parameter set, and at least one parameter in described 3rd parameter set is better than the parameter of the same race in described 4th parameter set; The described second pre-conditioned parameter of the same race be not inferior to for often kind of parameter in described 4th parameter set in described 3rd parameter set, and at least one parameter in described 4th parameter set is better than the parameter of the same race in described 3rd parameter set;

When described 3rd parameter set and described 4th parameter set meet described first pre-conditioned time, determine that described Secondary resource configuration plan is arranged in described first resource configuration plan;

When described 3rd parameter set and described 4th parameter set meet described second pre-conditioned time, determine that described first resource configuration plan is arranged in described Secondary resource configuration plan;

When described 3rd parameter set and described 4th parameter set do not meet described first pre-conditioned and do not meet described second pre-conditioned time, or when described first parameter set and described second parameter set all do not meet described parameter threshold condition, determine not arrange mutually between described first resource configuration plan and described Secondary resource configuration plan;

From described initial resource configuration plan set, first resource configuration plan and Secondary resource configuration plan is chosen described in repeating, until determine the dominance relation between each resource allocation program described, wherein, the first resource configuration plan at every turn chosen is entirely not identical with Secondary resource configuration plan.

6. method according to claim 5, is characterized in that, described each resource allocation program comprises multiple described resource allocation strategy, and wherein, each resource allocation strategy is for a kind of user type; And

Described first parameter set comprises throughput under each resource allocation strategy that described Component service comprises in described first resource configuration plan, time of delay, required cost and resource requirement total amount;

Described second parameter set comprises throughput under each resource allocation strategy that described Component service comprises in described Secondary resource configuration plan, time of delay, required cost and resource requirement total amount;

Described 3rd parameter set comprises the throughput of described Component service under described first resource configuration plan, time of delay, required cost and resource requirement total amount;

Described 4th parameter set comprises the throughput of described Component service under described Secondary resource configuration plan, time of delay, required cost and resource requirement total amount.

7. method according to claim 6, is characterized in that, described parameter threshold condition comprises:

${\mathrm{\&Sigma;}}_{l=1}^L{F}_{source}^l\left(X^l\right)\&le;C_{source}$

$F_{throughput}^l\left(X^l\right)\&GreaterEqual;C_{throughput}^l$

$F_{latency}^l\left(X^l\right)\&le;C_{latency}^l$

$F_{\cos t}^l\left(X^l\right)\&le;C_{\cos t}^l$

${\mathrm{\&Sigma;}}_{l=1}^L{F}_{\cos t}^l\left(X^l\right)\&le;C_{\cos t}$

Wherein, X ^lrepresent l the resource allocation strategy that resource allocation program comprises, this l resource allocation strategy is for l kind user type; L represents user type sum; represent l the resource allocation strategy X that described Component service comprises at resource allocation program ^lunder resource requirement total amount; represent l the resource allocation strategy X that described Component service comprises at resource allocation program ^lunder throughput; represent l the resource allocation strategy X that described Component service comprises at resource allocation program ^lunder time of delay; represent l the resource allocation strategy X that described Component service comprises at resource allocation program ^lunder required cost; C _sourcerepresent the resource requirement total amount threshold value for described system preset; represent the throughput threshold for l kind user type preset; represent threshold value time of delay for l kind user type preset; represent preset for cost threshold value needed for l kind user type; C _costrepresent preset for cost threshold value needed for described system.

8. the method according to any one of claim 1-3, is characterized in that, described basis has the resource allocation program subset of the highest dominance hierarchy, obtains target resource configuration plan set, comprising:

From described have choose multiple resource allocation program the resource allocation program subset of the highest dominance hierarchy and carry out cross and variation, generate multiple derivative resource allocation program;

Added to by generated multiple derivative resource allocation program in derivative resource allocation program set, wherein, this derivative resource allocation program set is initially empty set;

The resource allocation program subset having the highest dominance hierarchy by described and described derivative resource allocation program union of sets collection are formed as described target resource configuration plan set.

9. method according to claim 2, is characterized in that, the described target resource collocation strategy obtained from described target resource configuration plan set for described Component service, comprising:

According to the priority order from high to low of each resource allocation program in described target resource configuration plan set, obtain the resource allocation program alternatively resource allocation program of predetermined quantity;

Obtain user type information;

From each candidate resource configuration in the works, the resource allocation strategy with described user type information match is filtered out;

Obtain user to the preference configuration information of resource distribution;

From the resource allocation strategy of described and described user type information match, filter out the resource allocation strategy matched with described preference configuration information, as described target resource collocation strategy.

10., for a resource allocation device for the software systems based on service, it is characterized in that, this device comprises:

First acquisition module, for obtaining the multiple different resource allocation program of the Component service for this system, form initial resource configuration plan set, wherein, each resource allocation program comprises the resource allocation strategy of described Component service;

Subset division module, for each resource allocation program in described initial resource configuration plan set is divided according to dominance hierarchy, form multiple resource allocation program subset, wherein, the resource allocation program in same resource allocation program subset has identical dominance hierarchy;

Second acquisition module, for according to the resource allocation program subset with the highest dominance hierarchy, obtains target resource configuration plan set;

3rd acquisition module, for obtaining the target resource collocation strategy for described Component service from described target resource configuration plan set; And

Resource distribution module, for being described Component service Resources allocation according to described target resource collocation strategy.

11. devices according to claim 10, is characterized in that, described each resource allocation program comprises multiple described resource allocation strategy, and wherein, each resource allocation strategy is for a kind of user type.

12. devices according to claim 10, is characterized in that, described device also comprises:

Judge module, for having the resource allocation program subset of the highest dominance hierarchy in described basis, after obtaining target resource configuration plan set, and before the described target resource collocation strategy obtained from described target resource configuration plan set for described Component service, judge the cycle-index whether actual cycle number of times equals default, wherein, the initial value of described actual cycle number of times is 0, and described default cycle-index is more than or equal to 1;

Circular flow module, for when described actual cycle number of times is not equal to described default cycle-index, described actual cycle number of times is added 1, and using described target resource configuration plan set after new initial resource configuration plan set, rerun described subset division module, until described actual cycle number of times equals described default cycle-index.

13. devices according to any one of claim 10-12, it is characterized in that, described subset division module comprises:

Dominance relation determination submodule, for determining the dominance relation between each resource allocation program described, wherein, in described dominance relation, the dominance hierarchy of Zhi Peifang is higher than the dominance hierarchy of subject side;

Subset division submodule, for according to described dominance relation, the resource allocation program with identical dominance hierarchy is divided into a resource allocation program subset, wherein, do not arrange mutually between resource allocation program in same resource allocation program subset, further, the arbitrary resource allocation program had in the resource allocation program subset of higher dominance hierarchy can arrange the whole resource allocation programs had in the resource allocation program subset of lower dominance hierarchy.

14. devices according to claim 13, is characterized in that, described dominance relation determination submodule comprises:

Resource allocation program chooses unit, for choosing first resource configuration plan and Secondary resource configuration plan from described initial resource configuration plan set;

First parameter determination unit, for determining the first parameter set and the second parameter set, wherein, described first parameter set comprises QoS parameter under the resource allocation strategy that described Component service comprises in described first resource configuration plan and resource requirement total amount, and described second parameter set comprises QoS parameter under the resource allocation strategy that described Component service comprises in described Secondary resource configuration plan and resource requirement total amount;

Second parameter determination unit, for determining the 3rd parameter set according to described first parameter set, and determine the 4th parameter set according to described second parameter set, wherein, described 3rd parameter set comprises the QoS parameter of described Component service under described first resource configuration plan and resource requirement total amount, and described 4th parameter set comprises the QoS parameter of described Component service under described Secondary resource configuration plan and resource requirement total amount;

First judging unit, for judging whether described first parameter set and described second parameter set meet the parameter threshold condition preset respectively;

First dominance relation determining unit, for meeting described parameter threshold condition when the one in described first parameter set and described second parameter set, and another one is not when meeting described parameter threshold condition, the domination of the resource allocation program corresponding to parameter set determining to meet described parameter threshold condition does not meet the resource allocation program corresponding to parameter set of described parameter threshold condition;

Second judging unit, for when described first parameter set and described second parameter set all meet described parameter threshold condition, judge that described 3rd parameter set and described 4th parameter set are satisfied first pre-conditioned or second pre-conditioned, wherein, the described first pre-conditioned parameter of the same race be not inferior to for often kind of parameter in described 3rd parameter set in described 4th parameter set, and at least one parameter in described 3rd parameter set is better than the parameter of the same race in described 4th parameter set; The described second pre-conditioned parameter of the same race be not inferior to for often kind of parameter in described 4th parameter set in described 3rd parameter set, and at least one parameter in described 4th parameter set is better than the parameter of the same race in described 3rd parameter set;

Second dominance relation determining unit, for meet when described 3rd parameter set and described 4th parameter set described first pre-conditioned time, determine that described Secondary resource configuration plan is arranged in described first resource configuration plan;

3rd dominance relation determining unit, for meet when described 3rd parameter set and described 4th parameter set described second pre-conditioned time, determine that described first resource configuration plan is arranged in described Secondary resource configuration plan;

4th dominance relation determining unit, for do not meet when described 3rd parameter set and described 4th parameter set described first pre-conditioned and do not meet described second pre-conditioned time, or when described first parameter set and described second parameter set all do not meet described parameter threshold condition, determine not arrange mutually between described first resource configuration plan and described Secondary resource configuration plan;

Circular flow unit, for reruning, described resource allocation program chooses unit, until determine the dominance relation between each resource allocation program described, wherein, the first resource configuration plan at every turn chosen is entirely not identical with Secondary resource configuration plan.

15. devices according to claim 14, is characterized in that, described each resource allocation program comprises multiple described resource allocation strategy, and wherein, each resource allocation strategy is for a kind of user type; And

Described first parameter set comprises throughput under each resource allocation strategy that described Component service comprises in described first resource configuration plan, time of delay, required cost and resource requirement total amount;

Described second parameter set comprises throughput under each resource allocation strategy that described Component service comprises in described Secondary resource configuration plan, time of delay, required cost and resource requirement total amount;

Described 3rd parameter set comprises the throughput of described Component service under described first resource configuration plan, time of delay, required cost and resource requirement total amount;

Described 4th parameter set comprises the throughput of described Component service under described Secondary resource configuration plan, time of delay, required cost and resource requirement total amount.

16. devices according to claim 15, is characterized in that, described parameter threshold condition comprises:

${\mathrm{\&Sigma;}}_{l=1}^L{F}_{source}^l\left(X^l\right)\&le;C_{source}$

$F_{throughput}^l\left(X^l\right)\&GreaterEqual;C_{throughput}^l$

$F_{latency}^l\left(X^l\right)\&le;C_{latency}^l$

$F_{\cos t}^l\left(X^l\right)\&le;C_{\cos t}^l$

${\mathrm{\&Sigma;}}_{l=1}^L{F}_{\cos t}^l\left(X^l\right)\&le;C_{\cos t}$

Wherein, X ^lrepresent l the resource allocation strategy that resource allocation program comprises, this l resource allocation strategy is for l kind user type; L represents user type sum; represent l the resource allocation strategy X that described Component service comprises at resource allocation program ^lunder resource requirement total amount; represent l the resource allocation strategy X that described Component service comprises at resource allocation program ^lunder throughput; represent l the resource allocation strategy X that described Component service comprises at resource allocation program ^lunder time of delay; represent l the resource allocation strategy X that described Component service comprises at resource allocation program ^lunder required cost; C _sourcerepresent the resource requirement total amount threshold value for described system preset; represent the throughput threshold for l kind user type preset; represent threshold value time of delay for l kind user type preset; represent preset for cost threshold value needed for l kind user type; C _costrepresent preset for cost threshold value needed for described system.

17. devices according to any one of claim 10-12, it is characterized in that, described second acquisition module comprises:

Cross and variation submodule, for from described have choose multiple resource allocation program in the resource allocation program subset of the highest dominance hierarchy and carry out cross and variation, generate multiple derivative resource allocation program;

Derivative resource allocation program adds submodule, and for adding in derivative resource allocation program set by generated multiple derivative resource allocation program, wherein, this derivative resource allocation program set is initially empty set;

Target resource configuration plan set forms submodule, is formed as described target resource configuration plan set for the resource allocation program subset that has the highest dominance hierarchy by described and described derivative resource allocation program union of sets collection.

18. devices according to claim 11, is characterized in that, described 3rd acquisition module comprises:

Candidate resource configuration plan obtains submodule, for the priority order from high to low according to each resource allocation program in described target resource configuration plan set, obtains the resource allocation program alternatively resource allocation program of predetermined quantity;

User type acquisition of information submodule, for obtaining user type information;

First screening submodule, for from each candidate resource configuration in the works, filters out the resource allocation strategy with described user type information match;

Preference configuration information obtains submodule, for obtaining the preference configuration information of user to resource distribution;

Second screening submodule, in the resource allocation strategy from described and described user type information match, filters out the resource allocation strategy matched with described preference configuration information, as described target resource collocation strategy.