CN103873577B - Parallel calculating method of optimizing the combination of data-intensive Web services - Google Patents

Parallel calculating method of optimizing the combination of data-intensive Web services Download PDF

Info

Publication number
CN103873577B
CN103873577B CN 201410107815 CN201410107815A CN103873577B CN 103873577 B CN103873577 B CN 103873577B CN 201410107815 CN201410107815 CN 201410107815 CN 201410107815 A CN201410107815 A CN 201410107815A CN 103873577 B CN103873577 B CN 103873577B
Authority
CN
Grant status
Grant
Patent type
Prior art keywords
service
parallelism
combination
services
degree
Prior art date
Application number
CN 201410107815
Other languages
Chinese (zh)
Other versions
CN103873577A (en )
Inventor
俞东进
李畅
杨威
杨朔
Original Assignee
浙江天正信息科技有限公司
杭州电子科技大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Grant date

Links

Abstract

本发明公开了一种优化数据密集型Web服务组合的并行度计算方法。 The present invention discloses a method for calculating the degree of parallelism to optimize data-intensive Web services composition. 本发明首先获取用户输入指定的总花费和总耗时;获取当前基本服务的服务执行价格、响应时间和吞吐量信息。 The present invention first acquires the user input specifies the total cost and total time; Get current basic service execution price service, response time and throughput information. 其次所获取到的参数信息,通过制定的并行度上限值和下限值的计算公式获取并行度的可取范围。 Secondly, the acquired parameter information acquired by the parallelism possible range calculation formula developed on the degree of parallelism of the upper and lower limits. 最后对并行度的取值范围进行调整。 Finally, the range to adjust the degree of parallelism. 本发明提出的方法在大数据的背景下,组合服务需要处理海量数据的情况下,对Web服务并行化执行,以此来提高组合服务的数据处理能力和服务执行效率。 Case of the method proposed by the present invention in the context of large data, a combination of service handling a huge amount of data, the Web service execution parallelism, in order to improve data processing efficiency and service of the composite service. 并且,基于Web服务并行化执行的模式,对Web服务的并行度进行决策,以平衡组合服务的吞吐量和执行价格之间的关系,使它们的比值最大化,以达到优化组合服务的目的。 And, based on the mode of execution of the parallel Web service, the Web service parallelism make decisions, and the relationship between a certain service execution price balanced combination, so that their ratio is maximized to achieve optimal combination of services.

Description

一种优化数据密集型Web服务组合的并行度计算方法 Parallel calculating method of optimizing the combination of data-intensive Web services

技术领域 FIELD

[0001]本发明属于Web服务优化领域,具体涉及到一种优化数据密集型Web服务组合的并行度计算方法。 [0001] The present invention belongs to the field of optimization of Web services, particularly to a method for optimizing parallelism data-intensive computing Web service composition.

背景技术 Background technique

[0002] 随着社会信息化的快速发展,大数据(Big Data)给传统的数据管理技术带来了巨大挑战,数据量的急剧增加使得传统的数据管理方法不再适用。 [0002] With the rapid development of information society, large data (Big Data) to the traditional data management technology has brought great challenges, a sharp increase in the amount of data makes traditional data management methods are no longer applicable. 由于Web服务技术在功能封装和集成上的巨大成功,学术界和工业界着手借鉴Web服务和面向服务架构来实现数据密集型应用的开发、运行和管理。 Because Web services technology with great success in the functional packaging and integration on, learn from academia and industry working Web services and service-oriented architecture to achieve the development of data-intensive applications, operation and management.

[0003] 组合服务可以由多个基本Web服务通过串行结构、并行结构、条件选择结构和循环结构中的一种甚至多种结构组合而成。 [0003] The combination of a plurality of basic service may select one or even more of the Web service and structure as the cyclic structure formed by a combination of a serial configuration, a parallel configuration, conditions. 对需要处理大规模数据的Web服务组合来说,对组合服务中的基本服务应用并行结构,在减少计算成本、加速数据处理,提高组合服务执行效率方面能够取得一定成果。 The need for large-scale data processing Web service composition, the basic structure of parallel service applications portfolio of services, in terms of reducing the cost of computing and accelerate data processing and improve the efficiency of combined service to achieve certain results. 但是在现实情况下,用户对于整个组合服务有着一定的要求或者限制,比如响应时间能否尽量缩短,用户支付的费用能否尽量减少,用户所获收益能否尽量大等等。 But in reality, the user has a service for the entire portfolio or limit certain requirements, such as response time can minimize the cost paid by a user can minimize, users can try the proceeds and so big. 此处所述的用户收益是指,组合服务在一定时间和一定成本的条件下能够处理的数据量。 User benefits herein refers to the amount of data at a certain time and a certain cost of service that can be processed in combination. 问题的关键在于那些应用了并行结构的基本服务。 The key is that the basic structure of parallel service applications. 由于需要把服务并行化地执行, 这样虽然可以减少响应时间,但用户也不得不为并行的服务副本付出相应的花费,究竟并行度设置为多少才能达到最大限度地优化组合服务的目的成为了基于并行模式的服务组合优化问题。 Due to the need of the services performed in parallel, so although you can reduce the response time, but you also have to pay for a parallel copy of the corresponding service cost, what degree of parallelism to maximize the number of purposes in order to achieve optimal combination of services to be based on parallel mode service portfolio optimization problem.

[0004] 如何平衡组合服务的收益与成本之间的关系,并且通过决定一个合适的并行度来达到优化组合服务的目的是本发明讨论的关键问题。 [0004] The composition object of how to balance the relationship between the benefits and costs of the services, and by an appropriate degree of parallelism determined to achieve optimal combination of services is a key problem of the present invention is discussed.

发明内容 SUMMARY

[0005] 本发明针对现有技术的不足,提供了一种优化数据密集型Web服务组合的并行度计算方法。 [0005] The present invention addresses deficiencies in the prior art, provides parallel calculation method of optimizing the combination of data-intensive Web services.

[0006] 本发明方法的具体步骤是: [0006] In particular steps of the method of the present invention is:

[0007] 步骤(1).输入用户指定的对组合流程总花费C和总耗时T的限制。 Limiting [0007] Step (1) a user-specified combination of input flow and the total cost C T is total time.

[0008] 步骤(2).获取组合服务流程中各个基本服务Si的执行价格Ci,响应时间Ti和吞吐量Pi,i表不第i个基本服务。 [0008] Step (2). Gets composition exercise price of each service flow of basic services Si Ci, Ti response time and throughput Pi, i is not i-th table of basic services.

[0009] 所述的组合服务由多个基本服务通过并行结构或者并行结构与其它结构组合而成,所述的其它结构包括串行结构、条件选择结构和循环结构。 A combination of [0009] the composition by a plurality of service basic service through the parallel structure or a parallel structure and other structures, said structure further comprises a serial structure, the structure and condition selection loop structure.

[0010] 所述的基本服务的执行价格是指服务请求者通过调用服务所需支付的费用; The exercise price of the basic service of [0010] refers to the cost of the service by calling the service requester required to pay;

[0011] 所述的基本服务的响应时间是指从请求服务开始到执行完毕所用时间。 [0011] The basic service response time is measured from the beginning to the service request finished by the time.

[0012] 所述的基本服务的吞吐量是指单位时间能够处理的数据量。 [0012] The basic service is a certain amount of data that can be processed per unit time.

[0013] 步骤(3).计算并行度D0P的下限值p为 [0013] Step (3) calculation of the lower limit of the degree of parallelism p is D0P

Figure CN103873577BD00041

[0015]其中,组合服务由m(m多2)个任务组成,第j个任务由基本服务通过并行结构实现; A,E为大于零的常数,可以由式(1)获得, [0015] wherein the composition is served by m (m more than two) two tasks, the first task is performed by the j-th service through substantially parallel structure; A, E is a constant greater than zero, may be obtained by the formula (1),

Figure CN103873577BD00042

(1) (1)

[0017] CuPuh如上文所述分别表示服务S,的执行价格、吞吐量和响应时间;参数U表示如式(2), [0017] CuPuh service as described above represent S, the execution price, throughput, and response time; parameter U as represented by formula (2),

[0018] U=P3P4...Pm+PlP4...Pm+PlP3P4...Pm+,…,+ΡΐΡ3···Ρϋΐ-1(2) [0018] U = P3P4 ... Pm + PlP4 ... Pm + PlP3P4 ... Pm +, ..., + ΡΐΡ3 ··· Ρϋΐ-1 (2)

[0019] 步骤(4).计算并行度D0P的上限值q为 [0019] Step (4) The upper limit value of q is calculated parallelism D0P

Figure CN103873577BD00043

[0021]步骤(5).基于实际情况,通过再次调整并行度来获得更精确的取值范围。 [0021] Step (5) based on the actual situation, to get a more accurate range by adjusting the degree of parallelism again. 上述步骤中,通过计算得出了并行度的上限值和下限值,此时如果下限值大于上限值,那么需要用户提高组合服务总花费,否则需要重新选择基本服务。 The above-described step, the calculated upper limit value and the lower limit value of the degree of parallelism, at this time if the lower limit is greater than the upper limit, it requires the user to increase the total cost of a combination of services, basic services or need to reselect.

[0022]本发明所提供的优化数据密集型Web服务组合的并行度计算方法由一组功能模块组成,它们包括:参数集获取模块、并行度生成模块和并行度调整模块。 [0022] The present invention provides optimized data parallelism calculation intensive Web service composition by a group of functional modules, which includes: a parameter obtaining module sets, the degree of parallelism generating module and parallelism adjustment module.

[0023] 参数集获取模块需要获取两方面的参数,其一是用户输入指定的总花费和总耗时;其二是获取当前基本服务的服务执行价格、响应时间和吞吐量信息。 [0023] parameter set parameter acquisition module needs to obtain two aspects, one is the user input designated the total cost and total time; the other is to get the service current exercise price of the basic service, response time and throughput information.

[0024] 并行度生成模块根据参数集获取模块所获取到的参数信息,通过制定的并行度上限值和下限值的计算公式获取并行度的可取范围。 [0024] The degree of parallelism generating module obtaining module acquired parameter information according to the parameter set, the possible range of parallelism acquired by upper and lower limits on the computing equation formulated parallelism.

[0025] 并行度调整模块根据参数集获取模块所获取的参数信息,对并行度生成模块生成的并行度的取值范围进行调整。 [0025] The degree of parallelism adjustment parameter obtaining module information acquisition module according to the parameters set in the range generation module generates the parallelism adjust the degree of parallelism.

[0026] 本发明提出的方法在大数据的背景下,组合服务需要处理海量数据的情况下,对Web服务并行化执行,以此来提高组合服务的数据处理能力和服务执行效率。 [0026] The case of the method proposed by the present invention in the context of large data, a combination of service handling a huge amount of data, the Web service execution parallelism, in order to improve data processing efficiency and service of the composite service. 并且,基于Web服务并行化执行的模式,对Web服务的并行度进行决策,以平衡组合服务的吞吐量和执行价格之间的关系,使它们的比值最大化,以达到优化组合服务的目的。 And, based on the mode of execution of the parallel Web service, the Web service parallelism make decisions, and the relationship between a certain service execution price balanced combination, so that their ratio is maximized to achieve optimal combination of services.

附图说明 BRIEF DESCRIPTION

[0027]图1优化数据密集型Web服务组合的并行度计算方法步骤图; Parallelism [0027] FIG 1 optimize data-intensive Web services composition calculation step of FIG;

[0028]图2包含并行结构的服务组合。 [0028] Figure 2 contains a combination of the parallel configuration of the service.

具体实施方式 detailed description

[0029] 本发明所提供的优化数据密集型Web服务组合的并行度计算方法的具体实施方式主要分3步(如图1所示): [0029] Optimization of the data provided by the present invention DETAILED DESCRIPTION intensive Web services composition calculated degree of parallelism mainly three steps (Figure 1):

[0030] (1)输入由用户指定的组合服务的总花费和总耗时的限制,输入组合服务流程中各个基本服务的服务执行价格、响应时间和吞吐量;(2)能够完成用户指定功能需求的既定Web服务部署在服务器上,采用并行结构提高了组合服务的执行效率(如图2所示),缩短了组合服务的响应时间。 [0030] (1) designated by the user input the combination and the total cost of the service Total time constraints, the service execution service prices of basic input composite service flow, response time and throughput; (2) to complete the function designated by the user demand service deployed on a predetermined Web server, using parallelism increases the efficiency of service combination (Figure 2), shortening the response time of the composite service. 由于随着并行度的增大,数据处理能力得到了提升,相应的组合服务执行价格也会增加,因此,基于BR0CS(Benefit Ratio of Composite Service)模型,根据提供的输入参数来计算Web服务的并行度,来平衡组合服务的吞吐量和执行价格,优化组合服务;(3)如果得出的并行度取值范围的下限值大于上限值,说明用户给出的总花费不足或者需要重新选择基本Web服务。 Since with increasing degree of parallelism, data processing capability has been upgraded, the corresponding composite service execution price will increase, and therefore, based on BR0CS (Benefit Ratio of Composite Service) model, calculates parallel Web service according to the input parameters provided degrees, and the strike price to balance the throughput of the composite service, composite service optimization; (3) if the lower limit value of the degree of parallelism obtained is greater than the upper limit of the range, indicating lack of total cost given by the user or may need to reselect The basic Web services. 那么基于实际情况中用户在费用上的限制,通过再次调整以获取并行度的更精确的取值范围。 Then the user based on the actual case limited in cost, by adjusting the degree of parallelism again to obtain more accurate range.

[0031] 为叙述方便,定义相关符号如下: [0031] For convenience, the relevant symbol is defined as follows:

[0032] C:用户能够支付的最高费用。 [0032] C: Users can pay the highest fees.

[0033] P:用户要求的最大吞吐量。 [0033] P: the maximum throughput required by the user.

[0034] T:用户可接受的最长响应时间。 [0034] T: user maximum acceptable response time.

[0035] Qe:组合服务的总成本。 [0035] Qe: the total cost of combined services.

[0036] 0Ρ· ·组合服务的总吞吐量。 [0036] The total throughput 0Ρ · · combined services.

[0037] 0Τ:组合服务的总响应时间。 [0037] 0Τ: total response time of the composite service.

[0038] C1:服务&的服务执行价格。 [0038] C1: service & service execution price.

[0039] Pi:服务Si的吞吐量。 [0039] Pi: the service throughput Si.

[0040] Ti:服务Si的服务响应时间。 [0040] Ti: Si service service response time.

[0041] (1)获取用户要求和数据作为输入参数 [0041] (a) acquiring user requirements and data as input parameters

[0042] 输入由用户指定的对组合服务流程总花费C和总耗时T的限制,获取组合服务流程中各个组合的服务的执行价格匕、响应时间和吞吐量? 1。 [0042] specified by the user input the total cost of C and T total time limit for a combination of service processes, service processes in various combinations to obtain a combination of service execution price dagger, response time and throughput? 1. 其中各个参数的定义如下: Wherein each parameter is defined as follows:

[0043] 组合服务的总花费为各个基本服务的服务执行价格总和,即 The total cost of [0043] the sum of a combination of services for the implementation of basic services prices of services, i.e.,

Figure CN103873577BD00051

[0045]组合服务的总响应时间为各个任务完成的时间总和,即 The total response time [0045] is the sum of the combined service time of each task is completed, i.e.,

Figure CN103873577BD00052

[0047] 服务的响应时间和吞吐量在服务器性能在正常范围内满足关系 [0047] The response time and throughput and services to satisfy the relationship in a normal range in the server performance

[0048] Τ=ΑθΒΡ+ξ(Α>0,Β>0ξ 为响应时间误差)。 [0048] Τ = ΑθΒΡ + ξ (Α> 0, Β> 0ξ response time error).

[0049] (2)计算并行度 [0049] (2) calculating the degree of parallelism

[0050]并行度是影响组合服务性能的一个重要因素,选择一个合适的并行度来平衡组合服务的吞吐量和执行价格,即最大化吞吐量和执行价格的比值。 [0050] The degree of parallelism is an important factor affecting the performance of a combination of services, selecting a proper balance to a certain degree of parallelism and execution price composite service, i.e. the ratio of the price and execution to maximize throughput.

[0051 ] BR0CS模型的定义如下: [0051] BR0CS model defined as follows:

Figure CN103873577BD00061

[0053]基于并行结构的组合服务,研究如何最大化组合服务的收益比,即 [0053] Based on a combination of services parallel structure, how to maximize portfolio returns than research services, namely,

Figure CN103873577BD00062

[0055] 此处,R,W为相关常数,i,j,m如上文表述, [0055] Here, R, W is dependent constant, i, j, m above expression,

Figure CN103873577BD00063

[0057] U=P3P4...Pm+PlP4".Pm+PlP3P4...Pm+,…,+PlP3...Pm-1 [0057] U = P3P4 ... Pm + PlP4 ".Pm + PlP3P4 ... Pm +, ..., + PlP3 ... Pm-1

[0058] 在应用了并行结构的组合服务中,并行度能够影响整个组合服务的响应时间、月艮务执行价格和吞吐量,在BR0CS模型中条件的限制下,结合上述定义推导得出并行度的下限值Ρ为 [0058] Application of a combination of services in parallel structure, the degree of parallelism can affect the response time of the entire composition and services, and monthly price that works to perform certain limits under conditions BR0CS model, as defined above in combination deduced parallelism the lower limit is Ρ

Figure CN103873577BD00064

[0060]其中,Α,Β为大于零的常数,可以由式(1)获得, [0060] wherein, [alpha], Beta is a constant greater than zero, may be obtained by the formula (1),

Figure CN103873577BD00065

(1) (1)

[0062]计算并行度D0P的上限值q为 [0062] The degree of parallelism D0P calculated upper limit value of q

Figure CN103873577BD00066

[0064]为优化组合服务,并行度将在上述范围中取得合适的整数。 [0064] The optimal combination of services, to obtain a suitable degree of parallelism integers within the above range.

[0065] (3)调整并行度取值范围 [0065] (3) adjusting the degree of parallelism in the range

[0066]在实际情况中,用户对组合服务的总花费有所限制,通过上面所得并行度的取值范围,可能出现下限值大于上限值的情况,这说明用户给出的组合服务总花费不足,或者需要重新选取基本服务来调整组合服务收益比,以获得更加精确的取值范围。 [0066] In practice, the total cost of a combination of the user service restrictions, by the degree of parallelism in the range above obtained, the lower limit is greater than the upper limit situation may occur, indicating that the service given by the user on the total composition lack of spending, or need to re-select to adjust the basic service portfolio service benefit ratio, in order to obtain a more precise range of values.

[0067]本发明可用于面向数据密集型应用的Web服务组合,为避免海量数据处理过程中响应时间过长的问题,对基本服务应用并行结构,使Web服务并行执行,并且通过本发明中的方法获得并行度的可取范围,用来平衡吞吐量和成本花费之间的关系,优化了面向数据密集型应用的组合服务,使用户获得最大组合服务收益比。 [0067] The present invention may be used in combination for Web services data-intensive applications, in order to avoid the problem of excessive mass data processing response time, the basic structure of a parallel service applications, Web services are executed in parallel, and the present invention the method of achieving the desired degree of parallelism range, to balance the relationship between throughput and cost spent to optimize the combination of services for data-intensive applications, so that users get the most benefit ratio combinational service.

Claims (1)

  1. 1. 一种优化数据密集型Web服务组合的并行度计算方法,其特征在于该方法的具体步骤是: 步骤(1).输入用户指定的对组合服务流程总花费C和总耗时T的限制; 步骤(2).获取组合服务流程中各个基本服务S1的执行价格C1,响应时间1\和吞吐量Pni 表示第i个基本服务; 所述的组合服务由多个基本服务通过并行结构或者并行结构与其它结构组合而成,所述的其它结构包括串行结构、条件选择结构和循环结构; 所述的基本服务的执行价格是指服务请求者通过调用服务所需支付的费用; 所述的基本服务的响应时间是指从请求服务开始到执行完毕所用时间; 所述的基本服务的吞吐量是指单位时间能够处理的数据量; 步骤(3).计算并行度,即并行执行的服务数量DOP的下限值p为 1. A method of calculating the degree of parallelism to optimize data-intensive Web services composition, characterized in that the specific steps of the method are: Step (1) to limit a user-specified combination of input service flow and total cost C of the total time T ;. step (2) deriving a combined service flow of each basic service S1 execution price C1, the response time of 1 \ and certain Pni denotes an i-th basic services; the combination of a plurality of basic service by a service or in parallel through the parallel structure other structures and combination of structures, other structures including the serial structure, the structure and condition selection loop structure; exercise price of the basic service is the cost of the call to the service required by the service requester to pay; and the response time refers to the basic services from the service start request to the used time is finished; throughput basic service is the amount of data that can be processed per unit of time; step (3) calculating the degree of parallelism, namely the number of services to be performed in parallel DOP is the lower limit of p
    Figure CN103873577BC00021
    其中,组合服务由m个任务组成,第j个任务由基本服务通过并行结构实现;A,B为大于零的常数,可以由式(1)获得, Wherein the composite service tasks by m, the j-th task is performed by a basic service through the parallel structure; A, B is a constant greater than zero, may be obtained by the formula (1),
    Figure CN103873577BC00022
    步骤(4).计筧并行执行的服各数量DOP的h眼倌q为 Step (4) The count of each of the majority of the number of parallel execution service of DOP eye groom q is h
    Figure CN103873577BC00023
    步骤(5).基于实际情况,通过再次调整并行度来获得更精确的取值范围;通过步骤(3) 和(4)计算得出了并行执行的服务数量DOP的上限值和下限值,此时如果下限值大于上限值,那么需要用户提高组合服务总花费,否则需要重新选择基本服务。 Step (5) based on the actual situation, to get a more accurate range by adjusting the degree of parallelism again; in step (3) and (4) the calculated upper limit value of the number of services performed in parallel DOP and lower limit In this case, if the lower limit is greater than the upper limit, it requires the user to increase the total cost of the service portfolio, or need to re-select the basic service.
CN 201410107815 2014-03-21 2014-03-21 Parallel calculating method of optimizing the combination of data-intensive Web services CN103873577B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 201410107815 CN103873577B (en) 2014-03-21 2014-03-21 Parallel calculating method of optimizing the combination of data-intensive Web services

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 201410107815 CN103873577B (en) 2014-03-21 2014-03-21 Parallel calculating method of optimizing the combination of data-intensive Web services

Publications (2)

Publication Number Publication Date
CN103873577A true CN103873577A (en) 2014-06-18
CN103873577B true CN103873577B (en) 2017-02-01

Family

ID=50911691

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 201410107815 CN103873577B (en) 2014-03-21 2014-03-21 Parallel calculating method of optimizing the combination of data-intensive Web services

Country Status (1)

Country Link
CN (1) CN103873577B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6338078B1 (en) * 1998-12-17 2002-01-08 International Business Machines Corporation System and method for sequencing packets for multiprocessor parallelization in a computer network system
CN101217564A (en) * 2008-01-16 2008-07-09 上海理工大学 A parallel communication system and the corresponding realization method of simple object access protocol
CN101567001A (en) * 2009-05-22 2009-10-28 清华大学 Method for managing metadata file layout of parallel file system
CN101977213A (en) * 2010-10-18 2011-02-16 南京大学 Service combination method based on service quality under service oriented architecture (SOA) in network system
CN102420812A (en) * 2011-10-24 2012-04-18 浙江大学 Automatic quality of service (QoS) combination method supporting distributed parallel processing in web service

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6338078B1 (en) * 1998-12-17 2002-01-08 International Business Machines Corporation System and method for sequencing packets for multiprocessor parallelization in a computer network system
CN101217564A (en) * 2008-01-16 2008-07-09 上海理工大学 A parallel communication system and the corresponding realization method of simple object access protocol
CN101567001A (en) * 2009-05-22 2009-10-28 清华大学 Method for managing metadata file layout of parallel file system
CN101977213A (en) * 2010-10-18 2011-02-16 南京大学 Service combination method based on service quality under service oriented architecture (SOA) in network system
CN102420812A (en) * 2011-10-24 2012-04-18 浙江大学 Automatic quality of service (QoS) combination method supporting distributed parallel processing in web service

Also Published As

Publication number Publication date Type
CN103873577A (en) 2014-06-18 application

Similar Documents

Publication Publication Date Title
Wu et al. A revised discrete particle swarm optimization for cloud workflow scheduling
Weng et al. Heuristic scheduling for bag-of-tasks applications in combination with QoS in the computational grid
US20050283754A1 (en) Method and system to generate chart models
Lin et al. Bandwidth‐aware divisible task scheduling for cloud computing
Xu et al. Reducing electricity demand charge for data centers with partial execution
CN103064664A (en) Hadoop parameter automatic optimization method and system based on performance pre-evaluation
Kang et al. A novel discrete particle swarm optimization algorithm for meta-task assignment in heterogeneous computing systems
Liu et al. Greening geographical load balancing
CN102117367A (en) Visual simulation system for airplane assembly site
Chauhan et al. QoS guided heuristic algorithms for grid task scheduling
Li et al. A pricing algorithm for cloud computing resources
CN101763445A (en) High-spectrum image dimensionality reduction chip
Jin et al. Towards optimized fine-grained pricing of IaaS cloud platform
CN104780214A (en) Cloud manufacturing system and method based on cloud computing and three-dimensional printing
US20090281987A1 (en) Integrated distributed query processor for data grids
US20130198758A1 (en) Task distribution method and apparatus for multi-core system
Berriman et al. Montage: The architecture and scientific applications of a national virtual observatory service for computing astronomical image mosaics
Liu et al. Economical and balanced energy usage in the smart home infrastructure: A tutorial and new results
CN102867083A (en) High-rigidity and light-weight design method considering uncertainty of slide block mechanism of press machine
CN103150424A (en) Method for acquiring fine distribution of reactor core three dimensional neutron flux density of reactor
Tseng et al. A minimized makespan scheduler with multiple factors for Grid computing systems
Netto et al. Coordinated rescheduling of Bag‐of‐Tasks for executions on multiple resource providers
CN103020620A (en) Remote-sensing image ortho-rectification method based on central processing unit (CPU) and graphics processing unit (GPU) coprocessing
Jianfang et al. An optimized scheduling algorithm on a cloud workflow using a discrete particle swarm
Zhang et al. An optimization algorithm for service composition based on an improved FOA

Legal Events

Date Code Title Description
C06 Publication
C10 Entry into substantive examination
C14 Grant of patent or utility model