CN101841539A - Method, device and system for grid resource allocation based on trust - Google Patents

Abstract

The invention provides a method, a device and a system for grid resource allocation based on trust. The method comprises the following steps: obtaining resource allocation requests transmitted by nodes in a grid environment; respectively reading the direct trust value of a resource provider and the score information of a resource consumer from prestored trust data according to information released by the resource provider and the information released by the resource consumer, and respectively calculating the global trust value of the resource provider and the global trust value of the resource consumer according to the direct trust value of the resource provider and the score information of the resource consumer; generating the selling price adjustment coefficient of the resource provider and the buying price adjustment coefficient of the resource consumer according to the global trust values; adjusting and matching the selling price in the information released by the resource provider and the buying price in the information released by the resource consumer to generate resource matching information; a customer conducting transaction according to the resource matching information and feeding back a transaction result and scores to a server; and the server updating the trust data containing the direct trust value of the resource provider and the score information of the resource consumer according to the scores. Thereby, the invention has the advantage that the safe grid resource sharing is realized.

Description

A kind of grid resource allocation method, Apparatus and system based on degree of belief

Technical field

The present invention particularly about the safe practice in the gridding resource shared procedure, is a kind of grid resource allocation method based on degree of belief, Apparatus and system about the network security technology field concretely.

Background technology

Grid is to utilize the Internet or dedicated network that the interconnected height of heterogeneous networks node resource of realizing of extensively that distribute, isomery, dynamic gridding resource (as: storage resources, bandwidth resources and CPU handle resource etc.) on the geography is shared and integrated technology.Since the resource in the grid geographically wide area distribute, autonomous, different access cost patterns is arranged and can dynamically add at any time and leave grid environment, so the management and the scheduling of resource is a very complicated problems under the grid environment.

In existing grid resource allocation method, the main foundation of resource allocation is the quotation of dealing node, thereby some safety problems under the grid environment, the network fraud that comprises hitchhike (free-riding) behavior, malicious node, and the collusion behavior etc., make the systematic function that grid resource allocation can't obtain to expect, thereby become the bottleneck of restriction grid resource allocation practical application.

So-called free-riding is meant, in the distributed environment of grid, may have " selfishness " node and participate in certain system the free resource of enjoying and but do not contribute any resource.And the malicious node in the grid may be upset resource allocation and shared procedure by malice supplier or the consumer who disguises as resource.These malicious nodes are during as resource provider: tend to exaggerative oneself have the number of resources of over-evaluating, even might introduce virus; And it is during as Resource consumers: may not want to pay corresponding remuneration, or give the iniquitous scoring of resource provider.

In sum, at the characteristics of grid environment, when realizing efficiently grid resource allocation, can also effectively resist the malicious node in the network and the problem of malicious act thereof and need to be resolved hurrily, to support complicated grid resource allocation demand.

Summary of the invention

The embodiment of the invention provides a kind of grid resource allocation method based on degree of belief, Apparatus and system, when solving efficiently grid resource allocation, can effectively resist the malicious node in the network and the problem of malicious act thereof, realize sharing of gridding resource.

One of purpose of the present invention is, a kind of grid resource allocation method based on degree of belief is provided, this method comprises: obtain the resource allocation request that each node is sent in the grid environment, and extract from resource allocation request that resource provider releases news and Resource consumers releases news; Release news and Resource consumers releases news read the score information of resource provider direct trust value and Resource consumers respectively from the degree of belief data that prestore according to resource provider, and calculate resource provider global trusting value and Resource consumers global trusting value respectively according to the score information of resource provider direct trust value and Resource consumers; Generate the selling price adjustment factor of resource provider according to resource provider global trusting value, and generate the buying price adjustment factor of Resource consumers according to Resource consumers global trusting value; Buying price during selling price during employing selling price adjustment factor and buying price adjustment factor release news to resource provider respectively and Resource consumers release news is regulated and is mated, generation comprises the resource matched information of coupling selling price and coupling buying price, and resource matched information is sent to corresponding node; Obtain the scoring of the Resource consumers that corresponding node sends, and the degree of belief data of the score information that comprises resource provider direct trust value and Resource consumers are upgraded according to scoring.

One of purpose of the present invention is, a kind of grid resource allocation server based on degree of belief is provided, this server comprises: the resource allocation request harvester, be used for obtaining the resource allocation request that each node of grid environment is sent, and extract from resource allocation request that resource provider releases news and Resource consumers releases news; Resource allocation degree of belief calculation element, be used for releasing news and Resource consumers releases news and reads the score information of resource provider direct trust value and Resource consumers from the degree of belief data that prestore respectively, and calculate resource provider global trusting value and Resource consumers global trusting value respectively according to the score information of resource provider direct trust value and Resource consumers according to resource provider; Resource price bidirectional modulation device is used for generating according to resource provider global trusting value the selling price adjustment factor of resource provider, and generates the buying price adjustment factor of Resource consumers according to Resource consumers global trusting value; Resource matched information delivery apparatus, buying price during selling price during employing selling price adjustment factor and buying price adjustment factor release news to resource provider respectively and Resource consumers release news is regulated and is mated, generation comprises the resource matched information of coupling selling price and coupling buying price, and resource matched information is sent to corresponding node; Resource belief updating device is used to obtain the scoring of the Resource consumers that corresponding node sends, and according to scoring the degree of belief data of the score information that comprises resource provider direct trust value and Resource consumers is upgraded.

One of purpose of the present invention is, a kind of grid resource allocation system based on degree of belief is provided, and this system comprises: grid resource allocation server and client side; The grid resource allocation server is connected with client by network; The grid resource allocation server comprises: the resource allocation request harvester is used for obtaining the resource allocation request that each client of grid environment is sent, and extracts from resource allocation request that resource provider releases news and Resource consumers releases news; Resource allocation degree of belief calculation element, be used for releasing news and Resource consumers releases news and reads the score information of resource provider direct trust value and Resource consumers from the degree of belief data that prestore respectively, and calculate resource provider global trusting value and Resource consumers global trusting value respectively according to the score information of resource provider direct trust value and Resource consumers according to resource provider; Resource price bidirectional modulation device is used for generating according to resource provider global trusting value the selling price adjustment factor of resource provider, and generates the buying price adjustment factor of Resource consumers according to described Resource consumers global trusting value; Resource matched information delivery apparatus, buying price during selling price during employing selling price adjustment factor and buying price adjustment factor release news to resource provider respectively and Resource consumers release news is regulated and is mated, generation comprises the resource matched information of coupling selling price and coupling buying price, and resource matched information is sent to corresponding client; Resource belief updating device is used to obtain the scoring of the Resource consumers that corresponding client sends, and according to scoring the degree of belief data of the score information that comprises resource provider direct trust value and Resource consumers is upgraded; Client comprises: the user login information input unit is used to import user's ID authentication information; Resource provides the request input unit, is used to input comprise user identification code, user name, selling price, resource and provide quantity, resource status and resource to provide the resource provider of time to release news; Resources consumption request input unit is used to input and comprises user identification code, user name, buying price, resources consumption quantity, resource status and the Resource consumers of resources consumption time and release news; Resource is joined device displaying result, is used to show resource matched information; Resource consumers scoring input unit is used to import the scoring of Resource consumers.

Beneficial effect of the present invention is, in the grid resource allocation process, introduced a kind ofly based on degree of belief, adopts the Resource Allocation Formula of combined bidirectional resource distribution mode, and realized corresponding grid resource allocation device and system.The present invention adopts the degree of belief of each node to adjust its resource allocation price in the process of exchange that first resource distributes.After each resource allocation transaction was finished, the Resource consumers of participating in the distribution (buyer) was marked to resource provider (seller), and these scorings are used to set up the two-way trust relationship with update system.On the one hand, this method and system is based on the currency pricing mechanism, therefore can make node contribution more resources for node provides excitation, thus the influence that alleviates the network node free-riding behavior.On the other hand, the malicious node of internodal two-way trust relationship in can recognition system, and can limit even stop malicious node and participate in later resource allocation.

Description of drawings

In order to be illustrated more clearly in the embodiment of the invention or technical scheme of the prior art, to do to introduce simply to the accompanying drawing of required use in embodiment or the description of the Prior Art below, apparently, accompanying drawing in describing below only is some embodiments of the present invention, for those skilled in the art, under the prerequisite of not paying creative work, can also obtain other accompanying drawing according to these accompanying drawings.

Fig. 1 is the annexation figure of system of the present invention;

Fig. 2 is the flow chart of the inventive method;

Fig. 3 is the structured flowchart of trust model of the present invention;

Fig. 4 to Fig. 6 is the schematic diagram in belief propagation of the present invention path;

Fig. 7 is the structured flowchart of apparatus of the present invention;

Fig. 8 is the structured flowchart of resource allocation request harvester of the present invention;

Fig. 9 is the structured flowchart of resource allocation degree of belief calculation element of the present invention;

Figure 10 is the structured flowchart of resource price bidirectional modulation device of the present invention;

Figure 11 is the structure chart of resource belief updating device of the present invention;

Figure 12 is the Organization Chart of the algoritic module in the present embodiment server;

Figure 13 is the structured flowchart of client terminal of the present invention;

Figure 14 is the flow chart of allocated bandwidth example of the present invention.

Embodiment

Below in conjunction with the accompanying drawing in the embodiment of the invention, the technical scheme in the embodiment of the invention is clearly and completely described, obviously, described embodiment only is the present invention's part embodiment, rather than whole embodiment.Based on the embodiment among the present invention, those of ordinary skills belong to the scope of protection of the invention not making the every other embodiment that is obtained under the creative work prerequisite.

Embodiment one

As shown in Figure 1, be the grid environment based on the grid resource allocation method of degree of belief of present embodiment, wherein server 100 is connected with client (200a, 200b, 200c and 200d) as network node by network.Carry out present embodiment based on the grid resource allocation method of degree of belief the time, comprise step as shown in Figure 2:

Obtain the resource allocation request that each node is sent in the grid environment, and extract from resource allocation request that resource provider releases news and Resource consumers releases news (step S101); Release news and Resource consumers releases news read the score information of resource provider direct trust value and Resource consumers respectively from the degree of belief data that prestore according to resource provider, and calculate resource provider global trusting value and Resource consumers global trusting value (step S102) respectively according to the score information of resource provider direct trust value and Resource consumers; Generate the selling price adjustment factor of resource provider according to resource provider global trusting value, and generate the buying price adjustment factor (step S103) of Resource consumers according to Resource consumers global trusting value; Buying price during selling price during employing selling price adjustment factor and buying price adjustment factor release news to resource provider respectively and Resource consumers release news is regulated and is mated, generation comprises the resource matched information of coupling selling price and coupling buying price, and resource matched information is sent to corresponding node (step S104); Obtain the scoring of the Resource consumers that corresponding node sends, and the degree of belief data of the score information that comprises resource provider direct trust value and Resource consumers are upgraded (step S105) according to scoring.

In the present embodiment, be provided with a set K, be made up of the resource that the k kind is different, in the communications field, these resources generally include storage resources, CPU computational resource, bandwidth resources etc.; Resource allocation participant number is n (comprising buyer and seller), its auction project set B={ B ₁, B ₂..., B _nRepresent B _jCan be expressed as (a _j, p _j) form, a wherein _j=(a _1j, a _2j..., a _Ij..., a _Kj), a _IjThe quantity of representing the i kind resource that comprised in j the auction project, a _Ij＞0 expression is to the demand of resource i; a _Ij＜0 expression is to the supply of resource i; And a _Ij=0 expression resource i is not in the auction project.p _jExpression bidder j is to the quotation of this combination of resources, p _j＞0 represents to vie for purchase quotation, p _j＜0 represents to vie for selling quotation, and then the combined bidirectional resource allocation problem can be represented with following surface model:

$\max \underset{j=1}{\overset{n}{\mathrm{\Σ}}}{p}_j{x}_j---\left(1\right)$

$s.t.\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}{x}_j\≤0,\∀i\∈K---\left(2\right)$

$x_j\∈\left\{\mathrm{0,1}\right\},\∀j\∈\{1,...,n\}---\left(3\right)$

Wherein, constraints (2) shows that resource owner should guarantee to provide the resource of these quantity at least when the user vies for purchase the resource of some.Above-mentioned specification of a model, although combined bidirectional resource allocation problem more complicated, utilization integer programming method can give good description, in addition, because the restriction of constraints (3), model can be reduced to the one-zero programming problem.

After introducing degree of belief and corresponding trust incentive mechanism, should set up two-way trust relationship according to the scoring after each transaction, then based on the trust value of each node, mapping obtains corresponding bidirectional modulation parameter TP (i) and TC (i), and this parameter is used to regulate the bid of Resource consumers and supplier in the resource allocation.When node during as the supplier, adopt and regulate parameter TP (i), when node during, adopt and regulate parameter TC (i) as the consumer.Therefore, the basic model of introducing in formula (1)～(3) is modified to:

$\max \underset{j=1}{\overset{n}{\mathrm{\Σ}}}{p}_j^{\′}{x}_j---\left(4\right)$

$s.t.\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}{x}_j\≤0,\∀i\∈M---\left(5\right)$

$x_j\∈\left\{\mathrm{0,1}\right\},\∀j\∈\{1,...,n\}$

$p_j^{\′}=\left\{\begin{array}{cc}\mathrm{TP}\left(j\right)\·p_j& \mathrm{if\; peer\; j\; is\; provider}\\ \mathrm{TC}\left(j\right)\·p_j& \mathrm{if\; peer\; j\; is\; customer}\end{array}\right.$

Below detailed description is used to set up the trust model of two-way trust relationship and the mapping method of two bidirectional modulation parameter TP (i) and TC (i).

In Fig. 2, the resource provider global trusting value of step S102 and Resource consumers global trusting value are based on that trust model calculate to obtain, wherein:

As shown in Figure 3, trust model is to be used to set up and to upgrade two-way trust relationship between the grid environment lower node.The global trusting value of resource provider (GTV, Global Trust Value) and the GTV of Resource consumers are the output of this trust model, and calculate separately respectively.

Trust model comprises: direct trust value (DTV:Direct Trust Value) computing module, current direct trust value (CDTV:Current Direct Value) computing module, recommendation trust value (RTV:Recommendation Trust Value) computing module, the total trust value of resource provider (TTV) computing module, resource provider global trusting value GTV computing module and Resource consumers global trusting value GTV computing module.

Resource provider DTV computing module is meant, a certain node is during as resource provider, and other nodes are by the trust value of direct dealing to its generation.With D _IjRepresent node i to the direct trust value of node j (this moment, node i was the resource user, and node j is a resource provider), it is according in the transaction in the past, and node i obtains the score calculation of this node j.Here establish after each transaction finishes, node i is "+1 " or " 1 " by the scoring to node j that grading module provides, respectively satisfied transaction of representative and dissatisfied transaction.The node i that node j obtains all is recorded in parameter Ra ting to its all scorings _{I, j}=(Sat _{I, j}, Unsat _{I, j}) in, Sat wherein _{I, j}The expression node i is to the satisfied scoring total degree of node j, and Unsat _{I, j}The expression node i is to the dissatisfied scoring total degree of node j, and then node i is to the direct trust value D of node j _IjCalculate by following formula:

$D_{\mathrm{ij}}=\frac{{\mathrm{Sat}}_{i,j}-{\mathrm{Unsat}}_{i,j}}{{\mathrm{Sat}}_{i,j}+{\mathrm{Unsat}}_{i,j}}$

As can be seen, node i to the direct trust value scope of node j between (0,1), and normally asymmetric, that is to say, usually D _Ij≠ D _Ji

In resource provider CDTV computing module, because the common right and wrong static state of the residing environment of node, and after transaction is finished at every turn, follow-up additional events can exert an influence to the degree of belief of node, that is to say that along with the prolongation of time, the degree of belief that node had can reduce.Therefore, before two nodes begin transaction, node i to the current direct trust value of node j (with C _IjExpression) usually can be lower than the direct trust value that the last transaction of these two nodes has been upgraded to obtain when just having finished.At this, in order to reflect the characteristic of trust value, introduced time attenuation function " T (t) " along with the time decay, wherein t represents the time interval between twice transaction, and node i is to the current direct trust value C of node j _IjThen be the direct trust value D that obtains after upgrading when transaction is finished by the last time _IjMultiply each other with time attenuation function T (t) and to obtain.According to the characteristic of trusting decay, it is comparatively suitable to choose the time attenuation function " T (t) " with exponential distribution characteristic.We set degree of belief that node has per 7 days (604800 seconds) and just reduce to originally 10% herein, and therefore, before each transaction, node i is to the current direct trust value C of node j _IjAnd time attenuation function T (t) can be calculated by following formula:

C _ij(t ₀+t)＝D _ij(t ₀)×T(t)

$T\left(t\right)=e^{\frac{-2.3}{604800}\×t}$

Wherein, parametric t ₀Timestamp during for node i and node j last transaction, and parametric t is the timestamp between twice transaction.

Resource provider RTV computing module is meant, based on the transferability of trusting, and the scoring that the historical trading of a certain node by third party's node obtains to another node confidence.For example in fact, node i can obtain the recommendation trust value of node i to node j by the historical trading of node i and node k and node k and node j.How that obtain two recommendation trust values between the node according to the historical trading of third party's node, then needs recommendation trust propagation model reasonable in design.

The belief propagation Model Design need solve the problem of three aspects: at first need to determine under the situation of the transaction record between known two nodes and the third party's node, how to obtain the recommendation trust value between these two nodes, be referred to as " recommendation trust propagation algorithm " here; It is less important determines in one network, how effectively to search to offer suggestions and third party's node of reference for the recommendation trust value between two nodes, is referred to as " recommendation trust propagation path (Trust Propagation Path) " here; At last, after having obtained the recommendation trust value that all third party's nodes provide at these two nodes, utilize that the recommendation trust algorithm is synthetic to obtain total recommendation trust value between these two nodes.

As Fig. 4, Fig. 5 and shown in Figure 6, for the algorithm of resource provider RTV computing module is described, existing is example with 10 nodes in the grid environment, describes the computational methods based on the RTV value of recommendation trust propagation path.In this example, the gridding resource kind that needs to distribute can be storage resources, CPU computational resource and bandwidth resources etc.Because 10 nodes are arranged, therefore participating in resource allocation person's number is 10.These participants both can be used as resource provider, also can be used as Resource consumers.If establish each transaction duration is 10800 seconds (3 hours), and then the time attenuation function is:

$T\left(t\right)=e^{\frac{-2.3}{604800}\×10800}=e^{\frac{-2.3}{56}},$

Then internodal current direct trust value of 10 after time attenuation function weighting such as table 1:

Current direct trust value D when showing 1:10 node as resource provider _{I, j}

For current direct trust value D _{I, j}, the span of i and j is: 1,2,3,4,5,6,7,8,9,10.The recommendation trust algorithm comprises: recommendation trust propagation algorithm and recommendation trust composition algorithm, wherein:

(1) resource provider recommendation trust propagation algorithm

Recommendation trust value RTV is based on the trust transmission characteristic, and recommendation that the employing third party provides and suggestion obtain.For example, node i whether when also uncomprehending node j concluded the business in the past, just can depend on the credit value of node j in decision.In brief, although and do not know direct trust value DTV between node i and the node j, if can obtain the direct trust value DTV between node i and node k and node k and the node j, just can obtain the credit value RTV between node i and the node j.

RTV is by the recommendation trust propagation algorithm, is calculated by direct trust value DTV, specifically describes as follows: " if the direct trust value DTV between node i and the node k is D _{I, k}, the direct trust value between node k and the node j is D _{K, j}, then how to infer the credit value RTV between node i and the node j "? from the angle of probability, multiplying each other is to calculate the topmost method of credit value RTV, that is to say, RTV (i, j)=DTV (i, k) * (k j), is abbreviated as RTV to DTV _{I, j}=D _{I, k}* D _{K, j}

The recommendation trust propagation path: the another one problem is exactly " a recommendation trust propagation path " in the recommendation trust propagation model, because propagation model is exactly to propagate between all possible node initial trusting relationship is as much as possible.Then, as shown in Figure 4, have a lot of propagation paths between two nodes, if each node is all contributed to some extent to the recommendation trust value of other nodes, then system complexity can sharply rise, and in fact, the data dissemination model must be traded off between performance and validity usually.Therefore, propagation path should limit the propagation degree of depth (number of intermediate node just) of reference, and as can be known, propagating the degree of depth is 0 o'clock, and credit value RTV is direct trust value DTV just.

Herein, adopt and propagate the degree of depth 2, provide the topological structure of corresponding 10 nodes (peer) among Fig. 4, be calculated as example, the process of demonstration propagation path search with the credit value RTV between node 1 and the node 7.

Fig. 4 has provided node topology structure and the node direct trust value relation between in twos, as can be seen, and with not having direct trust path between dark node 1 that marks out and the node 7, so before node 1 and node 7 transaction, the credit value RTV of necessary calculating between them _1,7

To propagate the degree of depth be 1 o'clock credit value RTV_R1 in order to calculate, and must find all to propagate degree of depth is 1 propagation path, and its intermediate node marks out (node 2 and node 8) with dark color in this example in Fig. 5, and corresponding credit value RTV can calculate respectively.The credit value RTV that each paths is obtained averages at last, promptly obtains propagating the degree of depth and be 1 o'clock credit value RTV_R1, is shown below.

$\mathrm{RTV}\_R1=\frac{\underset{n=1}{\overset{N1}{\mathrm{\Σ}}}{D}_{i,k}\×D_{k,j}}{N1}$

Herein, N1 represents to propagate the number that the degree of depth is 1 o'clock propagation path.

According to the current direct trust value D in the table 1 _{I, j}, can obtain:

${\mathrm{RTV}}_{\mathrm{1,7}}\_R1=\frac{\underset{n=1}{\overset{N1}{\mathrm{\Σ}}}{D}_{i,k}\×D_{k,j}}{N1}=\frac{D_{\mathrm{1,2}}\×D_{\mathrm{2,7}}+D_{\mathrm{1,8}}\×D_{\mathrm{8,7}}}{2}$

$=\frac{0.8326\×0.3512+06749\×0.3578}{2}=0.2669$

For propagating the degree of depth is 2 o'clock credit value RTV_R2, and available similar method obtains, and its intermediate node marks out (node 2, node 3, node 8 and node 9) with dark color in this example in Fig. 6, and credit value RTV_R2 can be obtained by following formula:

$\mathrm{RTV}\_R2=\frac{\underset{n=1}{\overset{N2}{\mathrm{\Σ}}}{D}_{i,p}\×D_{p,q}\×D_{q,j}}{N2}$

Here, N2 represents to propagate the number that the degree of depth is 2 o'clock propagation paths.

According to the current direct trust value D in the table 1 _{I, j}, can obtain:

${\mathrm{RTV}}_{\mathrm{1,7}}\_R2=\frac{\underset{n=1}{\overset{N2}{\mathrm{\Σ}}}{D}_{i,p}\×D_{p,q}\×D_{q,j}}{N2}=\frac{D_{\mathrm{1,2}}\×D_{\mathrm{2,3}}\×D_{\mathrm{3,7}}+D_{\mathrm{1,9}}\×D_{\mathrm{9,8}}\×D_{\mathrm{8,7}}}{2}$

$=\frac{0.8326\×0.5780\×0.4468+0.3568\×0.2568\×0.3578}{2}=0.1239$

(2) resource provider recommendation trust composition algorithm

When all credit values of propagating the degree of depth all obtain, they are weighted calculate whole credit value, be shown below:

RTV＝λ·RTV_R1+γ·RTV_R2+…，λ+γ+…＝1???λ，γ，…≥0

Here, parameter lambda, γ ... Deng being RTV_R1, RTV_R2 respectively ... weight, also just characterized their significance level respectively, in this example, adopt λ=0.8 γ=0.2, therefore have:

RTV＝λ·RTV_R1+γ·RTV_R2，λ+γ＝1??λ，γ≥0

According to the current direct trust value D in the table 1 _{I, j}, can obtain:

RTV _1，7＝λ·RTV _1，7_R1+γ·RTV _1，7_R2

＝0.8·RTV _1，7_R1+0.2·RTV _1，7_R2

＝0.8×0.2669+0.2×0.1239＝0.2383

The TTV computing module obtains by direct trust value DTV and credit value RTV are synthetic, and adopts similar method of weighting to characterize the DTV significance level different with RTV.The weighted value of DTV and RTV is respectively α and β in addition, alpha+beta=1, and α here, and β 〉=0, then total trust value can be calculated by following formula:

TTV＝α·DTV+β·RTV，α+β＝1

If total trust value more depends on direct trust value DTV but not credit value RTV, then α should be greater than β.Here we suppose α=0.8, β=0.2.

According to the current direct trust value D in the table 1 _{I, j}, can obtain:

TTV _1，7＝α·DTV _1，7+β·RTV _1，7

＝0.8×0.6531+0.2×0.2383＝0.5701

So far, trust model builds up, and internodal in twos trusting relationship also can be calculated by following formula accordingly:

TTV＝α·DTV+β·RTV

＝α·DTV+β·(λ·RTV_R1+γ·RTV_R2+…)

And alpha+beta=1 and λ+γ+...=1.

Wherein, parameters such as α, β, λ and γ have reflected the characteristic of using, and depend on that system realizes.

At last, resource provider GTV computing module is meant: whole system is to the scoring of a certain node, this be by in twos between the node averaging of TTV obtain.

Resource consumers GTV computing module is based on consumer's scoring.After each transaction was finished, consumer's last scoring was imported into the scoring filter, and whether adjudicate its scoring just.With Resource consumers i is example, and the total degree of just scoring (Fair Rating) and non-just scoring (Unfair Rating) is recorded in respectively among parameter F air (i) and the Unfair (i), and the GTV of Resource consumers i is calculated as follows:

D _ij＝Fair(i)-3*Unfair(i)

Fair(i)：Fair?Rating?Numbers?of?Customer?i

Unfair(i)：Unfair?Rating?Numbers?of?Customer?i

The judgement principle of scoring filter is the last scoring of Resource consumers i to be marked with the history of other nodes compare, judgement is according to being: if the history consistent with this scoring scoring quantity is bigger by 3 than inconsistent historical scoring quantity, then understanding scoring this time is just; Otherwise,, think that then the right and wrong of this time marking are just if bigger by 3 than the consistent history quantity of marking with the inconsistent historical scoring quantity of this scoring; All the other situations think that then this time scoring is neutral.

In Fig. 2, the adjustment factor of step S103 is based on that the excitation parameters Model Calculation obtains.

The excitation parameters model is used to produce bidirectional modulation coefficient T P (i) and the TC (i) to resource provider and Resource consumers auction price, and wherein TP (i) and TC (i) represent the adjusting parameter of resource provider and Resource consumers respectively.After the GTV of resource provider and Resource consumers calculates, be used to produce TP (i) and TC (i) as the input of excitation parameters model.

The data basis of excitation parameters model is that the resource provider that is stored in the data in server storage device is regulated parameter TP (i) mapping relations and Resource consumers adjusting parameter TC (i) mapping relations.Wherein, resource provider is regulated parameter TP (i) mapping relations and is comprised two fields at least, that is: the value of the span of resource provider GTV and TP (i).Resource consumers is regulated parameter TC (i) mapping relations and is comprised two fields at least, that is: the value of the span of Resource consumers GTV and TC (i).The value of the value of the span of GTV, TP (i) and TC (i) obtains by statistics, and prestores and be stored in the server.

Table 2 is regulated the example of parameter TP (i) mapping relations for resource provider.

Table 2: resource provider is regulated parameter TP (i) mapping table

Resource provider GTV	??TP(i)
		??GTV?of?Providers＜0	??5
??0≤GTV?of?Providers＜0.2	??1.5
		??0.2≤GTV?of?Providers＜0.5	??1
??0.5≤GTV?of?Providers＜0.8	??0.85
		??0.8≤GTV?of?Providers＜1	??0.7

Table 3 is regulated the example of parameter TC (i) mapping relations for Resource consumers.

Table 3: Resource consumers is regulated parameter TC (i) mapping table

Resource consumers GTV	??TC(i)
		??GTV?of?Customers＜0	??0.5
??0≤GTV?of?Customers＜10	??1.1
		??10≤GTV?of?Customers＜20	??1.2
??20≤GTV?of?Customers＜30	??1.3
		??30≤GTV?of?Customers＜40	??1.4
??40≤GTV?of?Customers＜50	??1.5
		??50≤GTV?of?Customers＜60	??1.6
??60≤GTV?of?Customers＜70	??1.7
		??70≤GTV?of?Customers＜80	??1.8
??80≤GTV?of?Customers＜90	??1.9
		??90≤GTV?of?Customers＜100	??2.0
??100≤GTV?of?Customers	??2.5

As shown in table 2, if the GTV=-1 of resource provider first; Then with the input of GTV=-1 as the excitation parameters model, the excitation parameters model is according to the exportable TP of table 2 (i)=5.If the GTV=-1 of this resource provider illustrates that then the degree of belief of this resource provider is very low, if this moment this resource provider the resource selling price be 100 monetary units, utilize the selling price after TP (i) coefficient adjustment to be so:

100 * TP (i)=100 * 5=500 monetary unit.

On the contrary, if the GTV=0.9 of resource provider second; Then with the input of GTV=0.9 as the excitation parameters model, the excitation parameters model is according to the exportable TP of table 2 (i)=0.7.If the GTV=0.9 of this resource provider illustrates that then the degree of belief of this resource provider is very high, if this moment this resource provider the resource selling price be 100 monetary units, utilize the selling price after TP (i) coefficient adjustment to be so:

100 * TP (i)=100 * 0.7=70 monetary unit.

In server, first and second provide under the situation of same asset, and the adjusted price of resource provider second is far below first, and then second has very high transaction priority, and the transaction priority of first is then very low.

As shown in table 3, if the GTV=-1 of Resource consumers third; Then with the input of GTV=-1 as the excitation parameters model, the excitation parameters model is according to the exportable TC of table 3 (i)=0.5.If the GTV=-1 of Resource consumers third illustrates that then the degree of belief of Resource consumers third is very low, if this moment Resource consumers third the resource buying price be 100 monetary units, utilize the buying price after TC (i) coefficient adjustment to be so:

100 * TC (i)=100 * 0.5=50 monetary unit.

On the contrary, if the GTV=100 of Resource consumers fourth; Then with the input of GTV=100 as the excitation parameters model, the excitation parameters model is according to the exportable TC of table 3 (i)=2.5.If the GTV=100 of Resource consumers fourth illustrates that then the degree of belief of Resource consumers fourth is very high, if this moment the Resource consumers fourth the resource buying price be 100 monetary units, utilize the buying price after TC (i) coefficient adjustment to be so:

100 * TC (i)=100 * 2.5=250 monetary unit.

In server, under the situation of third fourth consumption same asset, the adjusted price of fourth is far above the third, and then fourth has very high transaction priority, and third transaction priority is then very low.

At last, to conclude the business priority higher resource provider and Resource consumers of server mated, and matching result sent to client, after the client carries out the transaction of success according to matching result, by Resource consumers is the scoring that resource provider carries out degree of belief, and scoring fed back to server, to upgrade direct trust value and historical data.The final gridding resource that provides by Resource consumers shared resource supplier.

Client modules is installed on each grid node, be mainly used in login, resource provider issue resource, Resource consumers application resource, each transaction finish after Resource consumers resource provider is marked.Server is mainly used in the resource bulletin collecting Resource consumers and be used to apply for resource, it is right to calculate the most favourable transaction coupling of system.This module is based on the combined bidirectional auction system of trust excitation and the nucleus module of system.Server is opened and is monitored (Listener) thread, monitors the message that client (Client) sends over all the time; Start timer, after each cycle that resource issue and the resource request message received in this cycle is comprehensive, extract in the local storage available but coupling is calculated in still untapped resource comprehensive consideration again.Afterwards local data is upgraded the timing next cycle.

Embodiment two

As shown in Figure 7, the grid resource allocation server based on degree of belief of present embodiment comprises: resource allocation request harvester 101, be used for obtaining the resource allocation request that each node of grid environment is sent, and extract from resource allocation request that resource provider releases news and Resource consumers releases news; Resource allocation degree of belief calculation element 102, be used for releasing news and Resource consumers releases news and reads the score information of resource provider direct trust value and Resource consumers from the degree of belief data that prestore respectively, and calculate resource provider global trusting value and Resource consumers global trusting value respectively according to the score information of resource provider direct trust value and Resource consumers according to resource provider; Resource price bidirectional modulation device 103 is used for generating according to resource provider global trusting value the selling price adjustment factor of resource provider, and generates the buying price adjustment factor of Resource consumers according to Resource consumers global trusting value; Resource matched information delivery apparatus 104, buying price during selling price during employing selling price adjustment factor and buying price adjustment factor release news to resource provider respectively and Resource consumers release news is regulated and is mated, generation comprises the resource matched information of coupling selling price and coupling buying price, and resource matched information is sent to corresponding node; Resource belief updating device 105 is used to obtain the scoring of the Resource consumers that corresponding node sends and according to scoring the degree of belief data of the score information that comprises resource provider direct trust value and Resource consumers is upgraded.

As shown in Figure 8, resource allocation request harvester 101 comprises: client-side information receiving element 1011 is used for according to Connector﹠amp; The message that all Client of Listener thread real-time listening send over, and further send to the remainder processing.Authentication information extraction unit 1012 is responsible for user's authentication.The extraction unit 1013 that releases news is used to extract the resource provider that comprises user identification code, user name, selling price, resource and provide quantity, resource status and resource that the time is provided and releases news and comprise user identification code, user name, buying price, resources consumption quantity, resource status and the Resource consumers of resources consumption time and release news.Mark extraction unit 1014 of client is responsible for extracting the grade scoring message that sends over from the Client end, and stores in the associated intermediary.

As shown in Figure 9, resource allocation degree of belief calculation element 102 comprises: direct trust value acquiring unit 1021 is used for reading the resource provider direct trust value from the direct trust value of storage.Current direct trust value computing unit 1022 is used for calculating the current direct trust value of resource provider according to described resource provider direct trust value and default time attenuation function.Recommendation trust value computing unit 1023 is used for going out resource provider recommendation trust value according to current direct trust value of described resource provider and default recommendation trust algorithm computation.Total trust value computing unit 1024 is used for calculating the total trust value of resource provider according to current direct trust value of described resource provider and recommendation trust value.Resource provider global trusting value computing unit 1025 is used for going out described resource provider global trusting value according to total trust value of described resource provider and average algorithm computation.Resource consumers global trusting value computing unit 1026 is used for according to comprising just score information and non-just score information calculates described Resource consumers global trusting value in the score information of interior Resource consumers.

As shown in figure 10, resource price bidirectional modulation device 103 comprises: the mapping relations memory cell, be used for the mapping relations between storage resources supplier degrees of comparison, resource provider global trusting value and the selling price adjustment factor, and the mapping relations between Resource consumers degrees of comparison, Resource consumers global trusting value and the buying price adjustment factor.

Server comprises: degree of belief data storage device 106 is used to store the score information and the resource provider transaction history data of authenticating user identification data, resource provider direct trust value, Resource consumers.

As shown in figure 11, resource belief updating device 105 comprises: Resource consumers scoring acquiring unit 1051 is used to receive the scoring of the Resource consumers that described corresponding node sends; Updating block 1052 is filtered in scoring, be used for the scoring of described Resource consumers is filtered, and the scoring after will filtering is upgraded the score information of the Resource consumers that prestores as the score information of Resource consumers; Scoring feedback updating block 1053 is used for the resource provider transaction history data that prestores being upgraded as the resource provider direct trust value according to the scoring of described Resource consumers.

As shown in figure 12, be the framework of the algoritic module in the present embodiment server.Wherein:

In resource allocation degree of belief calculation element, resource provider global trusting value and Resource consumers global trusting value are based on that trust model calculate to obtain, and wherein: trust model is to be used to set up and to upgrade two-way trust relationship between the grid environment lower node.The global trusting value of resource provider and the GTV of Resource consumers are the output of this trust model, and calculate separately respectively.Trust model comprises: resource provider direct trust value (DTV) module, the current direct trust value of resource provider (CDTV) computing module, resource provider recommendation trust value (RTV) computing module, the total trust value of resource provider (TTV) computing module, resource provider global trusting value GTV computing module and Resource consumers global trusting value GTV computing module.Trust model can comprise resource provider global trusting value GTV calculating section and Resource consumers global trusting value GTV calculating section, the calculating of resource provider global trusting value GTV comprises: the DTV module is obtained the DTV of resource provider, the CDTV computing module calculates CDTV according to DTV and time attenuation function, the RTV computing module goes out RTV according to CDTV and recommendation trust algorithm computation, the TTV computing module calculates TTV according to CDTV and RTV, and resource provider global trusting value GTV computing module calculates resource provider GTV according to TTV.Resource consumers global trusting value GTV calculating section comprises: Resource consumers global trusting value GTV computing module calculates described Resource consumers global trusting value GTV according to the score information that comprises the Resource consumers of just score information and non-just score information.

In resource price bidirectional modulation device, with bidirectional modulation coefficient T P (i) and the TC (i) of excitation parameters model generation to resource provider and Resource consumers auction price, wherein TP (i) and TC (i) represent the adjusting parameter of resource provider and Resource consumers respectively.After the GTV of resource provider and Resource consumers calculates, be used to produce TP (i) and TC (i) as the input of excitation parameters model.

In resource matched information delivery apparatus, as shown in table 2, if the GTV=-1 of resource provider first; Then with the input of GTV=-1 as the excitation parameters model, the excitation parameters model is according to the exportable TP of table 2 (i)=5.If the GTV=-1 of this resource provider illustrates that then the degree of belief of this resource provider is very low, if this moment this resource provider the resource selling price be 100 monetary units, utilize the selling price after TP (i) coefficient adjustment to be so:

100 * TP (i)=100 * 5=500 monetary unit.

On the contrary, if the GTV=0.9 of resource provider second; Then with the input of GTV=0.9 as the excitation parameters model, the excitation parameters model is according to the exportable TP of table 2 (i)=0.7.If the GTV=0.9 of this resource provider illustrates that then the degree of belief of this resource provider is very high, if this moment this resource provider the resource selling price be 100 monetary units, utilize the selling price after TP (i) coefficient adjustment to be so:

100 * TP (i)=100 * 0.7=70 monetary unit.

In server, first and second provide under the situation of same asset, and the adjusted price of resource provider second is far below first, and then second has very high transaction priority, and the transaction priority of first is then very low.

As shown in table 3, if the GTV=-1 of Resource consumers third; Then with the input of GTV=-1 as the excitation parameters model, the excitation parameters model is according to the exportable TC of table 3 (i)=0.5.If the GTV=-1 of Resource consumers third illustrates that then the degree of belief of Resource consumers third is very low, if this moment Resource consumers third the resource buying price be 100 monetary units, utilize the buying price after TC (i) coefficient adjustment to be so:

100 * TC (i)=100 * 0.5=50 monetary unit.

On the contrary, if the GTV=100 of Resource consumers fourth; Then with the input of GTV=100 as the excitation parameters model, the excitation parameters model is according to the exportable TC of table 3 (i)=2.5.If the GTV=100 of Resource consumers fourth illustrates that then the degree of belief of Resource consumers fourth is very high, if this moment the Resource consumers fourth the resource buying price be 100 monetary units, utilize the buying price after TC (i) coefficient adjustment to be so:

100 * TC (i)=100 * 2.5=250 monetary unit.

Under the situation of third fourth consumption same asset, the adjusted price of fourth is far above the third, and then fourth has very high transaction priority, and third transaction priority is then very low.

To conclude the business priority higher resource provider and Resource consumers of resource matched information delivery apparatus mated, and matching result is sent to client.

In resource belief updating device, receive the client and carry out the degree of belief scoring carried out for resource provider after the Successful Transaction according to matching result, and with this scoring renewal direct trust value and historical data.The final gridding resource that provides by Resource consumers shared resource supplier.

As shown in figure 13, in the resource allocation system based on the grid resource allocation server of degree of belief, client 200 comprises: user login information input unit 201 is used to import user's ID authentication information; Resource provides request input unit 202, is used to input comprise user identification code, user name, selling price, resource and provide quantity, resource status and resource to provide the resource provider of time to release news; Resources consumption request input unit 203 is used to input and comprises user identification code, user name, buying price, resources consumption quantity, resource status and the Resource consumers of resources consumption time and release news; Resource is joined device displaying result 204, is used to show resource matched information; Resource consumers scoring input unit 205 is used to import the scoring of Resource consumers.

Mainly comprise following several interface in the client 200:

User's login interface (LoginFrame);

Select interface (SelectFrame), Resource consumers (Customer) and two kinds of selections of resource provider (Provider) are arranged;

The resource request interface comprises: resources consumption request interface (CustomerAnnounceFrame) and resource provide request interface (ProviderAnnounceFrame);

The matching result interface comprises: Resource consumers is interface (CustomerResultFrame) and resource provider interface (ProviderResultFrame) as a result;

The expired interface of resource (ProviderNotransFrame).

Comprise Transaction, Logging and other public classes among the Utils of the program of client 200, wherein Transaction is responsible for writing down the transaction situation of Resource consumers or resource provider, and Logging is responsible for recoding daily log.Be responsible for the data maintenance of subscriber authentication, storage user name and corresponding password.

Resource provides the form of request (ProviderAnnounce) storage file to be:

id＝username；price；quality；state；timestamp。Wherein id is issue application id number of each submission, the username user name, and the price total price, quality is a resource quantity, and state is resource state of living in (being which stage), and timestamp is the time of resource when beginning to apply for.

The form of resources consumption request (CustomerAnnounce) storage file is:

id＝username；price；quality；state；timestamp。Wherein id is resource bid id number of each submission, the username user name, and the price total price, quality is a resource quantity, and state is resource state of living in (being which stage), and timestamp is the time of resource when beginning to apply for.

Adopt the ProviderDTV after concluding the business to safeguard ProviderDTVTable and TimestampTable, be responsible for the data maintenance of resource provider transaction history data (ProviderTransHistory) dividing equally after the transaction (ProviderTransStorage).ProviderTransHistory mainly stores the transaction history data of Provider, concrete data definition such as following table 4:

The attribute column of table 4:ProviderTransHistory table

??Username	?Sequence	??SuccessNum	??FailureNum
				User name	Corresponding sequence number in the DTV table	Number of success	The frequency of failure

ProviderDTV is responsible for the data maintenance of ProviderDTVTable and TimestampTable.The DTV value of ProviderDTVTable record Provider, the corresponding timestamp of TimestampTable record DTV value.These two records all adopt the matrix-style record.

CustomerDTV is responsible for the data maintenance of CustomerDTVTable.CustomerDTVTable writes down the DTV value of each Customer.

As shown in figure 14, the example of present embodiment allocated bandwidth is as follows:

Providing request based on the rapid a1 that strolls, client 1 to server transmission bandwidth;

Appoint in the allocated bandwidth system of degree, the grid resource allocation server is connected with client 1 to client 4 by network.Wherein the flow process of bandwidth resource allocation comprises:

Step a1, client 1 provide request to server transmission bandwidth;

Step a2, client 2 provide request to server transmission bandwidth;

Step a3, client 3 send bandwidth consumption request to server;

Step a4, client 4 send bandwidth consumption request to server;

Step a5, server obtain the resource allocation request that each client is sent in the grid environment, and the extraction bandwidth provides request and bandwidth consumption request from resource allocation request, provides request to read bandwidth from the degree of belief data that prestore according to bandwidth the request direct trust value is provided;

Step a6, server calculate the current direct trust value CDTV of bandwidth supplier according to bandwidth supplier direct trust value DTV and default time attenuation function;

Step a7, server go out bandwidth supplier recommendation trust value RTV according to current direct trust value CDTV of bandwidth supplier and default recommendation trust algorithm computation;

Step a8, server calculate the total trust value TTV of bandwidth supplier according to current direct trust value CDTV of bandwidth supplier and recommendation trust value RTV, go out bandwidth supplier global trusting value GTV according to total trust value TTV of bandwidth supplier and average algorithm computation.

Step a9, server are according to comprising just score information and non-just score information calculates described Resource consumers global trusting value in the score information of interior Resource consumers.

Step a10, server generate the selling price adjustment factor of resource provider according to described resource provider global trusting value GTV.Comprising: set up the mapping relations between resource provider degrees of comparison, resource provider global trusting value and the selling price adjustment factor; And generate the buying price adjustment factor of resource provider according to described Resource consumers global trusting value GTV.Comprising: set up the mapping relations between Resource consumers degrees of comparison, Resource consumers global trusting value and the buying price adjustment factor.

Step a11. is according to bandwidth supplier's GTV and GTV, the TP (i) of bandwidth consumers and the bandwidth match object information that TC (i) generates bandwidth supplier and bandwidth consumers.

Step a11 sends the bandwidth match object informations to client 1 to client 4 to step a15, server;

Step a16 is to step a19, client 1 and client 3 according to the carrying out of the bandwidth match object information success that receives the bandwidth sharing transaction, client 3 has been consumed the bandwidth resources that client 1 provides, and to client's 1 defrayment or virtual expense, and will send to server to the scoring of client 1.Client 2 and client 4 Fail Transactions.

Step a20, server upgrade bandwidth supplier's DTV and transactions history record according to scoring.

Step a21, server filter scoring, and upgrade the score information of bandwidth consumers according to the scoring after filtering.

Present embodiment has been introduced a kind ofly based on degree of belief in the grid resource allocation process, adopts the Resource Allocation Formula of combined bidirectional resource distribution mode, and has realized corresponding grid resource allocation device and system.The present invention adopts the degree of belief of each node to adjust its resource allocation price in the process of exchange that first resource distributes.After each resource allocation transaction was finished, the Resource consumers of participating in the distribution (buyer) was marked to resource provider (seller), and these scorings are used to set up the two-way trust relationship with update system.On the one hand, this method and system is based on the currency pricing mechanism, therefore can make node contribution more resources for node provides excitation, thus the influence that alleviates the network node free-riding behavior.On the other hand, the malicious node of internodal two-way trust relationship in can recognition system, and can limit even stop malicious node and participate in later resource allocation.

Used specific embodiment among the present invention principle of the present invention and execution mode are set forth, the explanation of above embodiment just is used for helping to understand method of the present invention and core concept thereof; Simultaneously, for one of ordinary skill in the art, according to thought of the present invention, the part that all can change in specific embodiments and applications, in sum, this description should not be construed as limitation of the present invention.

Claims (16)

1. the grid resource allocation method based on degree of belief is characterized in that, described method comprises:

Obtain the resource allocation request that each node is sent in the grid environment, and extract from described resource allocation request that resource provider releases news and Resource consumers releases news;

Release news and Resource consumers releases news read the score information of resource provider direct trust value and Resource consumers respectively from the degree of belief data that prestore according to described resource provider, and calculate resource provider global trusting value and Resource consumers global trusting value respectively according to the score information of described resource provider direct trust value and Resource consumers;

Generate the selling price adjustment factor of resource provider according to described resource provider global trusting value, and generate the buying price adjustment factor of Resource consumers according to described Resource consumers global trusting value;

Adopt described selling price adjustment factor and buying price adjustment factor selling price and the buying price of Resource consumers in releasing news in respectively resource provider being released news to regulate and mate, generation comprises the resource matched information of coupling selling price and coupling buying price, and described resource matched information is sent to corresponding node;

Obtain the scoring of the Resource consumers that described corresponding node sends, and the described degree of belief data that comprise the score information of resource provider direct trust value and Resource consumers are upgraded according to described scoring.

2. method according to claim 1 is characterized in that, described resource provider releases news and comprises: user identification code, user name, selling price, resource provide quantity, resource status and resource that the time is provided;

Described Resource consumers releases news and comprises: user identification code, user name, buying price, resources consumption quantity, resource status and resources consumption time.

3. method according to claim 1 is characterized in that, calculates resource provider global trusting value according to described resource provider direct trust value and comprises:

Calculate the current direct trust value of resource provider according to described resource provider direct trust value and default time attenuation function; Go out resource provider recommendation trust value according to current direct trust value of described resource provider and default recommendation trust algorithm computation; Calculate the total trust value of resource provider according to current direct trust value of described resource provider and recommendation trust value; Go out described resource provider global trusting value according to total trust value of described resource provider and average algorithm computation;

Calculating Resource consumers global trusting value according to the score information of described Resource consumers comprises:

Calculate described Resource consumers global trusting value according to comprising just score information and non-just score information score information in interior Resource consumers.

4. method according to claim 1, it is characterized in that the selling price adjustment factor that generates resource provider according to described resource provider global trusting value comprises: set up the mapping relations between resource provider degrees of comparison, resource provider global trusting value and the selling price adjustment factor;

The buying price adjustment factor that generates resource provider according to described Resource consumers global trusting value comprises: set up the mapping relations between Resource consumers degrees of comparison, Resource consumers global trusting value and the buying price adjustment factor.

5. method according to claim 1 is characterized in that, described degree of belief data comprise: the score information of authenticating user identification data, resource provider direct trust value, Resource consumers and resource provider transaction history data.

6. method according to claim 5 is characterized in that, according to described scoring the described degree of belief data that comprise the score information of resource provider direct trust value and Resource consumers is upgraded to comprise:

Scoring to described Resource consumers is filtered, and according to the scoring after filtering, and the score information of the Resource consumers that prestores is upgraded;

According to the scoring of described Resource consumers, the resource provider direct trust value that prestores is upgraded;

According to the scoring of described Resource consumers, the resource provider transaction history data that prestores is upgraded.

7. grid resource allocation server based on degree of belief is characterized in that described server comprises:

The resource allocation request harvester is used for obtaining the resource allocation request that each node of grid environment is sent, and extracts from described resource allocation request that resource provider releases news and Resource consumers releases news;

Resource allocation degree of belief calculation element, be used for releasing news and Resource consumers releases news and reads the score information of resource provider direct trust value and Resource consumers from the degree of belief data that prestore respectively, and calculate resource provider global trusting value and Resource consumers global trusting value respectively according to the score information of described resource provider direct trust value and Resource consumers according to described resource provider;

Resource price bidirectional modulation device is used for generating according to described resource provider global trusting value the selling price adjustment factor of resource provider, and generates the buying price adjustment factor of Resource consumers according to described Resource consumers global trusting value;

Resource matched information delivery apparatus, adopt described selling price adjustment factor and buying price adjustment factor selling price and the buying price of Resource consumers in releasing news in respectively resource provider being released news to regulate and mate, generation comprises the resource matched information of coupling selling price and coupling buying price, and described resource matched information is sent to corresponding node;

Resource belief updating device is used to obtain the scoring of the Resource consumers that described corresponding node sends, and according to described scoring the described degree of belief data that comprise the score information of resource provider direct trust value and Resource consumers is upgraded.

8. server according to claim 7 is characterized in that, described resource allocation request harvester comprises:

Extraction unit releases news, being used to extract the resource provider that comprises user identification code, user name, selling price, resource and provide quantity, resource status and resource that the time is provided releases news and comprises user identification code, user name, buying price, resources consumption quantity, resource status and the Resource consumers of resources consumption time and release news.

9. server according to claim 7 is characterized in that, described resource allocation degree of belief calculation element comprises:

Current direct trust value computing unit is used for calculating the current direct trust value of resource provider according to described resource provider direct trust value and default time attenuation function;

Recommendation trust value computing unit is used for going out resource provider recommendation trust value according to current direct trust value of described resource provider and default recommendation trust algorithm computation;

Total trust value computing unit is used for calculating the total trust value of resource provider according to current direct trust value of described resource provider and recommendation trust value;

Resource provider global trusting value computing unit is used for going out described resource provider global trusting value according to total trust value of described resource provider and average algorithm computation;

Resource consumers global trusting value computing unit calculates described Resource consumers global trusting value according to comprising just score information and the non-just score information score information in interior Resource consumers.

10. server according to claim 7 is characterized in that, described resource price bidirectional modulation device comprises:

The mapping relations memory cell, be used between storage resources supplier degrees of comparison, resource provider global trusting value and the selling price adjustment factor mapping relations and

Mapping relations between Resource consumers degrees of comparison, Resource consumers global trusting value and the buying price adjustment factor.

11. server according to claim 7 is characterized in that, described server comprises:

The degree of belief data storage device is used to store the score information and the resource provider transaction history data of authenticating user identification data, resource provider direct trust value, Resource consumers.

12. server according to claim 11 is characterized in that, described resource belief updating device comprises:

Resource consumers scoring acquiring unit is used to receive the scoring of the Resource consumers that described corresponding node sends;

Updating block is filtered in scoring, is used for the scoring of described Resource consumers is filtered, and according to the scoring after filtering, and the score information of the Resource consumers that prestores is upgraded;

Scoring feedback updating block is used for the scoring according to described Resource consumers, and the resource provider transaction history data that prestores is upgraded.

13. the grid resource allocation system based on degree of belief is characterized in that described system comprises: grid resource allocation server and client side; Described grid resource allocation server is connected with described client by network;

Described grid resource allocation server comprises:

The resource allocation request harvester is used for obtaining the resource allocation request that each client of grid environment is sent, and extracts from described resource allocation request that resource provider releases news and Resource consumers releases news;

Resource allocation degree of belief calculation element, be used for releasing news and Resource consumers releases news and reads the score information of resource provider direct trust value and Resource consumers from the degree of belief data that prestore respectively, and calculate resource provider global trusting value and Resource consumers global trusting value respectively according to the score information of the score information of described resource provider direct trust value and Resource consumers according to described resource provider;

Resource price bidirectional modulation device is used for generating according to described resource provider global trusting value the selling price adjustment factor of resource provider, and generates the buying price adjustment factor of Resource consumers according to described Resource consumers global trusting value;

Resource matched information delivery apparatus, adopt described selling price adjustment factor and buying price adjustment factor selling price and the buying price of Resource consumers in releasing news in respectively resource provider being released news to regulate and mate, generation comprises the resource matched information of coupling selling price and coupling buying price, and described resource matched information is sent to corresponding client;

Resource belief updating device is used to obtain the scoring of the Resource consumers that described corresponding client sends, and according to described scoring the described degree of belief data that comprise the score information of resource provider direct trust value and Resource consumers is upgraded;

Described client comprises:

The user login information input unit is used to import user's ID authentication information;

Resource provides the request input unit, is used to input comprise user identification code, user name, selling price, resource and provide quantity, resource status and resource to provide the resource provider of time to release news;

Resources consumption request input unit is used to input and comprises user identification code, user name, buying price, resources consumption quantity, resource status and the Resource consumers of resources consumption time and release news;

Resource is joined device displaying result, is used to show described resource matched information;

Resource consumers scoring input unit is used to import the scoring of Resource consumers.

14. system according to claim 13 is characterized in that, described resource allocation degree of belief calculation element comprises:

Current direct trust value computing unit is used for calculating the current direct trust value of resource provider according to described resource provider direct trust value and default time attenuation function;

Recommendation trust value computing unit is used for going out resource provider recommendation trust value according to current direct trust value of described resource provider and default recommendation trust algorithm computation;

Total trust value computing unit is used for calculating the total trust value of resource provider according to current direct trust value of described resource provider and recommendation trust value;

Resource provider global trusting value computing unit is used for going out described resource provider global trusting value according to total trust value of described resource provider and average algorithm computation;

Resource consumers global trusting value computing unit calculates described Resource consumers global trusting value according to comprising just score information and the non-just score information score information in interior Resource consumers.

15. system according to claim 13 is characterized in that, described resource price bidirectional modulation device comprises:

The mapping relations memory cell, be used between storage resources supplier degrees of comparison, resource provider global trusting value and the selling price adjustment factor mapping relations and

Mapping relations between Resource consumers degrees of comparison, Resource consumers global trusting value and the buying price adjustment factor.

16. system according to claim 13 is characterized in that, described server comprises:

The degree of belief data storage device is used to store authenticating user identification data, resource provider direct trust value, Resource consumers score information and resource provider transaction history data.

Images