CN102158533A - Distributed web service selection method based on QoS (Quality of Service) - Google Patents

Distributed web service selection method based on QoS (Quality of Service) Download PDF

Info

Publication number
CN102158533A
CN102158533A CN2011100352115A CN201110035211A CN102158533A CN 102158533 A CN102158533 A CN 102158533A CN 2011100352115 A CN2011100352115 A CN 2011100352115A CN 201110035211 A CN201110035211 A CN 201110035211A CN 102158533 A CN102158533 A CN 102158533A
Authority
CN
China
Prior art keywords
service
skyline
services
prune
task
Prior art date
Legal status (The legal status 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 status listed.)
Granted
Application number
CN2011100352115A
Other languages
Chinese (zh)
Other versions
CN102158533B (en
Inventor
吴健
潘李敏
陈亮
尹建伟
李莹
邓水光
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang University ZJU
Original Assignee
Zhejiang University ZJU
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
Application filed by Zhejiang University ZJU filed Critical Zhejiang University ZJU
Priority to CN2011100352115A priority Critical patent/CN102158533B/en
Publication of CN102158533A publication Critical patent/CN102158533A/en
Application granted granted Critical
Publication of CN102158533B publication Critical patent/CN102158533B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)

Abstract

本发明涉及计算机领域，公开了一种基于QoS的分布式web服务选择方法，采用分布式的方式进行web服务的选择，通过将web服务选择这个需要非常巨大的计算能力才能解决的大问题分解成为很多小的部分，将这些小的部分分配给许多计算进行并行处理，最后把每个计算结果合并起来得到最终结果。本发明通过pre-prune-refine模型，在不同的机器上进行高效地并行计算，经过pre-prune过程提前删除一些不满足条件的web服务，显著地缩短响应时间，从而能够快速灵活地找到满足用户需求的服务。由于中间产生的输入输出都是临时文件，不会产生额外的输入输出开销，能够显著增强网络的传送效率，大大提高了选择的效率。

The present invention relates to the field of computers, and discloses a QoS-based distributed web service selection method, which uses a distributed method to select web services, and decomposes the large problem of web service selection, which requires a very large computing power to be solved, into A lot of small parts, assign these small parts to many calculations for parallel processing, and finally combine the results of each calculation to get the final result. The present invention uses the pre-prune-refine model to perform efficient parallel computing on different machines, delete some web services that do not meet the conditions in advance through the pre-prune process, and significantly shorten the response time, so that it can quickly and flexibly find satisfactory user required services. Since the input and output generated in the middle are all temporary files, no additional input and output overhead will be generated, which can significantly enhance the transmission efficiency of the network and greatly improve the efficiency of selection.

Description

Distribution Web method for service selection based on QoS
Technical field
The present invention relates to computer realm, related in particular to a kind of distribution Web method for service selection based on QoS.
Background technology
Distributed Calculation is a kind of new account form that proposes in recent years, the PROBLEM DECOMPOSITION how its research could solve a very huge computing capability of needs becomes many little parts, then these little parts are distributed to many computers and carry out parallel processing, each result of calculation is integrated obtain final result at last.Can be placed on program on the computer of optimum bootup window by Distributed Calculation and to move.
Web service selects to come to this a big problem that the very huge computing capability of needs could solve.It is exactly to select the more excellent service of quality to satisfy the demand of service requester from the suitable set of service of a large amount of functions that Web service is selected.Along with the development of service compute and cloud computing, the different service of a large amount of functional similarity not function attribute has appearred.Though Web service is registered in registration center storehouse (UDDI), in fact flatly be distributed on the different server in geographical position, and be connected to each other by Internet.The continual growth of Web service requires accurate, extendible, an efficient and reliable solution to seek and select only service.But when service during, this complexity very from different service provider, therefore how in a large amount of set of services of disperseing fast and flexible ground select the service of meeting consumers' demand, become web and served a important step in the evolution.The research of distribution Web services selection is suggested under this background just.
Along with the continuous development of the Internet, the actual storage of ISP and service becomes and more and more disperses.Service is stored on the different servers usually, and these servers might be in different geographical position.Current web services selection technology based on QoS supposes that mostly all services all are stored on the same platform, there is not to consider the distributed environment of extensive service, along with the increase of quantity of service and the distribution Web service environment of reality, the efficient of traditional services selection technology will reduce greatly, and because the very long experience that might influence the user of response time.The service of how selecting to satisfy different QoS requirements under the distributed environment of extensive set of service has become academia and the common problem of paying close attention to of industrial quarters.
A distributed framework more tallies with the actual situation, because Distributed Architecture allows to carry out better local service management, littler renewal cost, higher fault-tolerance.If but the web services selection technology that direct application has existed can produce very big expense, therefore take all factors into consideration the efficient and the distributed service environment of method for service selection, the Distributed Services choice mechanism of developing a kind of efficient parallel becomes the active demand of industry.
Summary of the invention
The efficient that the present invention is directed to services selection technology in the prior art is low excessively, response time is very long, influence the shortcoming of user experience effect, skyline technology in a kind of binding data storehouse is provided, propose distributed parallel web services selection mechanism, realized the distribution Web method for service selection based on QoS of web services selection efficiently.
In order to solve the problems of the technologies described above, the present invention is solved by following technical proposals:
Distribution Web method for service selection based on QoS comprises the steps:
Step a: master server is being managed all data servers, and allocating task, K map (mapping) task is arranged in this process, S pre-prune (pre-beta pruning) task and 1 refine (refining) task are assigned with (K＞0, S＞0), master server is given a map task, pre-prune task or refine Task Distribution in the machine of a free time;
Step b:Map process: the machine that has been assigned with the map task reads relevant web service list and is the input data, to import data parsing then and become the QoS vector, interim key/value is to (key/value to) in the middle of generating, and be buffered in the internal memory, the input of this process is the web service list, and output is the QoS vector;
Step c: the key/value that is buffered in the internal memory is regional to be divided into S by the subregion function, periodically be written on the local disk afterwards, the key/value of buffer memory is returned to master server to the memory location at local disk, is responsible for the key/value of buffer memory is sent to the pre-prune task again to the memory location on local disk by master server;
Steps d: after the pre-prune task receives the data storage location information that master server sends, the interim key/value in centre that reads buffer memory from the machine at map task place is right, after the machine at pre-prune task place had read all middle ephemeral datas, key (key) sorted outputed on the uniform machinery value (numerical value) with identical interim key in centre;
Step e:Pre-prune process: deletion in advance can not be the service of skyline service, on each machine, according to the skyline algorithm of selecting the QoS vector set of reading is operated, the service that deletion is arranged, obtain local skyline set of service, the middle interim local skyline set of service of generation is buffered in the internal memory;
Step f: the local skyline set of service of buffer memory is returned to master server in the memory location of this locality, by master server these local skyline set of services is sent to the refine task again in the memory location of this locality;
Step g: Refine process: receive the stored position information of the local skyline set of service that master server sends when the refine task after, read the interim local skyline set of service of buffer memory from the machine at pre-prune task place, after the machine at refine task place has read all local skyline set of services, by the local skyline service that the skyline algorithm deletion of selecting is arranged, obtain final overall skyline service;
Step h: after all map, pre-prune and refine task were all finished, master server woke user program up, at this moment, in user program calling just of pre-prune-refine is returned;
After completing successfully all tasks, the output of this model is divided into the M class, is stored in N output file, wherein M=N.
As preferably, the input and output of the Pre-prune task among the described step e are respectively: the interim key/value in the centre of reading to the skyline set of service of this locality.By the process of this pre-beta pruning, can significantly reduce the quantity of candidate skyline service, thereby reduce number of comparisons, therefore the last response time also can significantly reduce.
As preferably, described QoS vector definition: vectorial QVs=(q 1(s) ..., q d(s)) be the QoS vector of web service S, wherein q i(s) value by i the QoS attribute of serving the S issue determines.Adopt distributed mode to carry out the selection of web service, by this big PROBLEM DECOMPOSITION that needs very huge computing capability to solve of web services selection is become a lot of little parts, then that these are little part is distributed to many calculating and is carried out parallel processing, each result of calculation is merged to obtain final result at last.
System based on the distribution Web method for service selection of QoS comprises as the lower part:
Expansion service registration center: expanded traditional service register center, had the function of traditional services registration center, the service describing registration center for searching for registers the ISP that all have been issued, and the position of record web service cluster; Service requester just can be selected from its nearer ISP like this, thereby has reduced cost and response time;
Master server: be the core of web service cluster, be used to write down the position of each data server and the concrete service of each data server storage, provide and manage the directory information of whole system, and managing each data server;
Data server: be the actual cluster under the distributed environment, storing the relevant information of real service;
Skyline set of service storehouse: storing the skyline service that service requester requires of satisfying that each regional web service cluster chooses, only writing down the description of skyline service and the web service cluster at place.
The definition 1. service dominance relations based on the QoS vector: for service A and B, their QoS vector is respectively QV A=(q 1(A) ..., q dAnd QV (A)) B=(q 1(B) ..., q d(B)).Service A domination service B, the value on the QoS vector one dimension in office of and if only if service A is all poor unlike service B, and better than B on a dimension at least, and its formula definition is
Figure BSA00000431530200041
And
Figure BSA00000431530200042
Figure BSA00000431530200043
Definition 2.Skyline service: not by the service of other service dominations.
Services set { A with 17 services with 2 QoS attributes 1(1.1,4.7), A 2(1.8,4), A 3(2,4.7), A 4(1.3,4.2), A 5(1.6,4.34), B 1(3.1,2.1), B 2(3.5,1.5), B 3(3.8,1.8), B 4(3.22,1.6), B 5(3.6,2.2), B 6(4,4.6), C 1(6.1,2.2), C 2(6.3,1.8), C 3(6.8,2.4), C 4(6.5,2.5), C 5(6.9,1.6), C 6(7,5), C 7(7.5,1) } be example, by the Map task, we can be divided into three groups according to first QoS attribute with set of service: { A 1, A 2, A 3, A 4, A 5, { B 1, B 2, B 3, B 4, B 5, B 6, { C 1, C 2, C 3, C 4, C 5, C 6, C 7, each group is put into respectively on the machine and carries out the pre-prune task, after carrying out the pre-prune task, first group can obtain local skyline service { A 1, A 2, A 3, second group can obtain { B 1, B 2, B 3, the 3rd group can obtain { C 1, C 2, C 3, C 7, with the local skyline set { A that obtains 1, A 2, A 3, B 1, B 2, B 3, C 1, C 2, C 3, C 7Output on the machine of carrying out the refine task, after carrying out the refine task, we can obtain the skyline set { A of the overall situation 1, A 2, A 3, B 1, B 2, B 3, C 7.
The specific algorithm such as the table 1 of these three tasks:
Table 1 distribution Web method for service selection is described
Map：
Input：Key：the?name?of?server?Values：web?services?list
Output：Key：group?range?Values：QoS?vectors?of?Web?services
2 WS?is?distributed?to?each?node?by?DistributedCache
3 foreach?s∈WS?do
4 service＝extract?QoS?vector?from?s；
6 output_key＝an?exact?value?as?a?function?of?QoS；
output_value＝service；
7 EmitIntermediate(AsString(output_key)，output_value)；
8 end?for
Pre-prune：
Input?Key：the?intermediate?Key?produced?by?Map?function
Values：QoS?vectors?Vectors
Output?Key：an?identical?key?Values：candidate?skyline?services
Figure BSA00000431530200052
Figure BSA00000431530200053
3 SkylineComputeClass?skylineCompute＝new?SkylineComputeClass()；
4 foreach?v∈Vectors
5 services.add(s)；
6 end?for
7 candidateServices＝skylineCompute.computeSkyine(services)；
8 output_key＝new?IntWritable(1)；
9 output_value＝candidateServices；
10 Emit(output_key，output_value)
Refine：
Input?Key：the?only?one?key?produced?by?Combiner?function
Values：candidate?web?services
Output?Key：group?range?Values：global?skyline?services
Figure BSA00000431530200061
2 globalServices＝computeSkyline(Values)；
3 for?each?global?service?gs?in?globalServices
4 output_key＝an?exact?value?as?a?function?of?QoS
5 EmitIntermediate(output_key，gs)；
6 end?for
The present invention has significant technique effect owing to adopted above technical scheme:
The present invention is by the pre-prune-refine model, on different machines, carry out parallel computation efficiently, owing to can delete some web that does not satisfy condition services in advance through the pre-prune process, thereby shorten the response time significantly, thereby can find the service of meeting consumers' demand fast and flexible.Because the middle input and output that produce all are temporary files, can not produce extra input and output expense, can significantly strengthen the transmission efficiency of network, thereby improved the efficient of selecting greatly.
Description of drawings
Fig. 1 is the architectural framework figure of distribution Web services selection of the present invention.
Fig. 2 is the specific algorithm flow chart of distribution Web services selection of the present invention.
Embodiment
Below in conjunction with accompanying drawing 1 to Fig. 2 and embodiment the present invention is described in further detail:
Embodiment 1
Based on the distribution Web method for service selection of QoS, to shown in Figure 2, comprise the steps: as Fig. 1
Step a: master server is being managed all data servers, and allocating task, K map (mapping) task is arranged in this process, S pre-prune (pre-beta pruning) task and 1 refine (refining) task are assigned with (K＞0, S＞0), master server is given a map task, pre-prune task or refine Task Distribution in the machine of a free time;
Step b:Map process: the machine that has been assigned with the map task reads relevant web service list and is the input data, to import data parsing then and become the QoS vector, interim key/value is to (key/value to) in the middle of generating, and be buffered in the internal memory, the input of this process is the web service list, and output is the QoS vector;
Step c: the key/value that is buffered in the internal memory is regional to be divided into S by the subregion function, periodically be written on the local disk afterwards, the key/value of buffer memory is returned to master server to the memory location at local disk, is responsible for the key/value of buffer memory is sent to the pre-prune task again to the memory location on local disk by master server;
Steps d: after the pre-prune task receives the data storage location information that master server sends, the interim key/value in centre that reads buffer memory from the machine at map task place is right, after the machine at pre-prune task place had read all middle ephemeral datas, key (key) sorted outputed on the uniform machinery value (numerical value) with identical interim key in centre;
Step e:Pre-prune process: deletion in advance can not be the service of skyline service, on each machine, according to the skyline algorithm of selecting the QoS vector set of reading is operated, the service that deletion is arranged, obtain local skyline set of service, the middle interim local skyline set of service of generation is buffered in the internal memory;
Step f: the local skyline set of service of buffer memory is returned to master server in the memory location of this locality, by master server these local skyline set of services is sent to the refine task again in the memory location of this locality;
Step g: Refine process: receive the stored position information of the local skyline set of service that master server sends when the refine task after, read the interim local skyline set of service of buffer memory from the machine at pre-prune task place, after the machine at refine task place has read all local skyline set of services, by the local skyline service that the skyline algorithm deletion of selecting is arranged, obtain final overall skyline service;
Step h: after all map, pre-prune and refine task were all finished, master server woke user program up, at this moment, in user program calling just of pre-prune-refine is returned;
After completing successfully all tasks, the output of this model is divided into the M class, is stored in N output file, wherein M=N.
The input and output of the Pre-prune task among the step e are respectively: the interim key/value in the centre of reading to the skyline set of service of this locality.
QoS vector definition: vectorial QVs=(q 1(s) ..., q d(s)) be the QoS vector of web service S, wherein q i(s) value by i the QoS attribute of serving the S issue determines.
Implement the system of above-mentioned distribution Web method for service selection based on QoS, comprise as the lower part:
Expansion service registration center: expanded traditional service register center, had the function of traditional services registration center, the service describing registration center for searching for registers the ISP that all have been issued, and the position of record web service cluster;
Master server: be the core of web service cluster, be used to write down the position of each data server and the concrete service of each data server storage, provide and manage the directory information of whole system, and managing each data server;
Data server: be the actual cluster under the distributed environment, storing the relevant information of real service;
Skyline set of service storehouse: storing the skyline service that service requester requires of satisfying that each regional web service cluster chooses, only writing down the description of skyline service and the web service cluster at place.
Detailed process based on the distribution Web services selection of QoS is as follows:
1. certain regional web ISP cluster is in the service of expansion service registration center issue self-management and storage, and the request of using self service is responded.Service register center is set up index to it, and is writing down service list and description that this cluster provides.
2. service requester sends the services selection request to expansion service registration center, and the corresponding constrained parameters restriction of some QoS attribute (promptly to) are set.
3. expansion service registration center is according to the request of service requester, at existing service, send the request of selecting to each regional web service cluster, each regional web service cluster utilizes the algorithm in the table 1 to carry out local skyline services selection, and the result is sent in the skyline set of service storehouse, skyline set of service storehouse merges the local skyline set of service in each zone, and delete those local skyline that arranged and serve, obtain final overall skyline set, be stored in the skyline set of service storehouse, select for service requester.
The present invention is by the pre-prune-refine model, on different machines, carry out parallel computation efficiently, owing to can delete some web that does not satisfy condition services in advance through the pre-prune process, thereby shorten the response time significantly, thereby can find the service of meeting consumers' demand fast and flexible.Because the middle input and output that produce all are temporary files, can not produce extra input and output expense, can significantly strengthen the transmission efficiency of network, thereby improved the efficient of selecting greatly.
In a word, the above only is preferred embodiment of the present invention, and all equalizations of being done according to the present patent application claim change and modify, and all should belong to the covering scope of patent of the present invention.

Claims (4)

1.基于QoS的分布式web服务选择方法，其特征在于：包括如下步骤：1. the distributed web service selection method based on QoS, it is characterized in that: comprise the steps: 步骤a：主服务器管理着所有的数据服务器，并分配任务，在这个过程中有K个map(映射)任务，S个pre-prune(预剪枝)任务和1个refine(精炼)任务被分配(K＞0，S＞0)，主服务器将一个map任务、pre-prune任务或refine任务分配给一个空闲的机器；Step a: The master server manages all data servers and assigns tasks. During this process, K map (mapping) tasks, S pre-prune (pre-prune) tasks and 1 refine (refining) tasks are assigned (K>0, S>0), the master server assigns a map task, pre-prune task or refine task to an idle machine; 步骤b：Map过程：被分配了map任务的机器读取相关web服务列表为输入数据，然后将输入数据解析成QoS向量，生成中间临时的key/value对(键/值对)，并缓存在内存中，这一过程的输入是web服务列表，输出是QoS向量；Step b: Map process: The machine assigned the map task reads the list of related web services as input data, then parses the input data into QoS vectors, generates intermediate temporary key/value pairs (key/value pairs), and caches them in In memory, the input to this process is the list of web services, and the output is the QoS vector; 步骤c：缓存在内存中的key/value对被分区函数分成S个区域，之后周期性地写入到本地磁盘上，缓存的key/value对在本地磁盘的存储位置被传回给主服务器，由主服务器负责把缓存的key/value对在本地磁盘上的存储位置再传送给pre-prune任务；Step c: The key/value pairs cached in the memory are divided into S regions by the partition function, and then periodically written to the local disk, and the cached key/value pairs are sent back to the main server at the storage location of the local disk. The master server is responsible for transferring the cached key/value pair storage location on the local disk to the pre-prune task; 步骤d：当pre-prune任务接收到主服务器发来的数据存储位置信息之后，从map任务所在的机器上读取缓存的中间临时key/value对，在pre-prune任务所在的机器读取了所有的中间临时数据之后，对key(键)进行排序使具有相同中间临时key的value(数值)输出到同一机器上；Step d: After the pre-prune task receives the data storage location information sent by the master server, it reads the cached intermediate temporary key/value pair from the machine where the map task is located, and reads it on the machine where the pre-prune task is located. After all the intermediate temporary data, sort the key (key) so that the value (value) with the same intermediate temporary key is output to the same machine; 步骤e：Pre-prune过程：提前删除不可能是skyline服务的服务，在每个机器上，根据选择的skyline算法对读取的QoS向量集合进行操作，删除被支配的服务，得到本地的skyline服务集合，生成的中间临时本地skyline服务集合被缓存在内存中；Step e: Pre-prune process: Delete services that cannot be skyline services in advance. On each machine, operate on the read QoS vector set according to the selected skyline algorithm, delete the dominated services, and obtain local skyline services Collection, the generated intermediate temporary local skyline service collection is cached in memory; 步骤f：缓存的本地skyline服务集合在本地的存储位置被传回给主服务器，由主服务器把这些本地skyline服务集合在本地的存储位置再传送给refine任务；Step f: The cached local skyline service collection is sent back to the master server at the local storage location, and the master server sends these local skyline service collections to the local storage location and then sends them to the refine task; 步骤g：Refine过程：当refine任务接收到主服务器发来的本地skyline服务集合的存储位置信息后，从pre-prune任务所在的机器上读取缓存的临时本地skyline服务集合，当refine任务所在的机器读取了所有的本地skyline服务集合之后，通过选择的skyline算法删除被支配的本地skyline服务，得到最终的全局skyline服务；Step g: Refine process: After the refine task receives the storage location information of the local skyline service collection sent by the master server, it reads the cached temporary local skyline service collection from the machine where the pre-prune task is located. After the machine reads all the local skyline service sets, it deletes the dominated local skyline services through the selected skyline algorithm to obtain the final global skyline service; 步骤h：当所有的map、pre-prune和refine任务都完成之后，主服务器唤醒用户程序，此时，在用户程序里对pre-prune-refine的调用才返回；Step h: After all map, pre-prune and refine tasks are completed, the main server wakes up the user program. At this time, the call to pre-prune-refine in the user program returns; 在成功完成所有任务之后，该模型的输出被分成M类，被存放在N个输出文件中，其中M＝N。After all tasks are successfully completed, the output of the model is divided into M categories and stored in N output files, where M=N. 2.根据权利要求1所述的基于QoS的分布式web服务选择方法，其特征在于：所述的步骤e中的Pre-prune任务的输入和输出分别是：读取的中间临时的key/value对和本地的skyline服务集合。2. The QoS-based distributed web service selection method according to claim 1, characterized in that: the input and output of the Pre-prune task in the step e are respectively: the intermediate temporary key/value read A collection of paired and local skyline services. 3.根据权利要求1所述的基于QoS的分布式web服务选择方法，其特征在于：所述的QoS向量定义：向量QVs＝(q1(s)，…，qd(s))为web服务S的QoS向量，其中qi(s)由服务S发布的第i个QoS属性的值决定。3. the distributed web service selection method based on QoS according to claim 1, characterized in that: said QoS vector definition: vector QVs=(q 1 (s), ..., q d (s)) is web QoS vector of service S, where q i (s) is determined by the value of the ith QoS attribute published by service S. 4.实施如权利要求1-3所述的基于QoS的分布式web服务选择方法的系统，其特征在于：包括如下部分：4. implement the system of the distributed web service selection method based on QoS as claimed in claim 1-3, it is characterized in that: comprise following part: 扩展服务注册中心：扩展了传统的服务注册中心，具有传统服务注册中心的功能，为可搜索的服务描述注册中心，注册所有已经发布的服务提供者，并记录web服务集群的位置；Extended Service Registry: Extends the traditional service registry, has the function of the traditional service registry, describes the registry for searchable services, registers all published service providers, and records the location of the web service cluster; 主服务器：为web服务集群的核心部分，用于记录每个数据服务器的位置以及每个数据服务器存储的具体服务，提供和管理整个系统的目录信息，并且管理着各个数据服务器；Main server: the core part of the web service cluster, used to record the location of each data server and the specific services stored by each data server, provide and manage the directory information of the entire system, and manage each data server; 数据服务器：是分布式环境下的实际集群，存储着真实服务的相关信息；Data server: It is an actual cluster in a distributed environment, storing relevant information of real services; Skyline服务集合库：存储着各个区域的web服务集群选择出来的满足服务请求者要求的skyline服务，只记录skyline服务的描述和所在的web服务集群。Skyline service collection library: stores the skyline services selected by the web service clusters in various regions to meet the requirements of the service requester, and only records the description of the skyline service and the web service cluster where it is located.
CN2011100352115A 2011-01-28 2011-01-28 Distributed web service selection method based on QoS (Quality of Service) Expired - Fee Related CN102158533B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2011100352115A CN102158533B (en) 2011-01-28 2011-01-28 Distributed web service selection method based on QoS (Quality of Service)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2011100352115A CN102158533B (en) 2011-01-28 2011-01-28 Distributed web service selection method based on QoS (Quality of Service)

Publications (2)

Publication Number Publication Date
CN102158533A true CN102158533A (en) 2011-08-17
CN102158533B CN102158533B (en) 2013-11-13

Family

ID=44439710

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2011100352115A Expired - Fee Related CN102158533B (en) 2011-01-28 2011-01-28 Distributed web service selection method based on QoS (Quality of Service)

Country Status (1)

Country Link
CN (1) CN102158533B (en)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103164287A (en) * 2013-03-22 2013-06-19 河海大学 Distributed-type parallel computing platform system based on Web dynamic participation
CN103209102A (en) * 2013-03-11 2013-07-17 北京邮电大学 Web quality of service distributed measurement system and method
CN104717294A (en) * 2015-03-23 2015-06-17 浪潮集团有限公司 Data extracting method, main server and cluster
CN104735166A (en) * 2015-04-13 2015-06-24 李金忠 Skyline service selection method based on MapReduce and multi-target simulated annealing
CN105578212A (en) * 2015-12-15 2016-05-11 南京邮电大学 A peer-to-peer streaming media real-time monitoring method under the big data streaming computing platform
CN106790536A (en) * 2016-12-21 2017-05-31 国网江西省电力公司信息通信分公司 Composite Web services system of selection based on affairs and QoS
CN112787870A (en) * 2021-02-25 2021-05-11 苏州大学 Parallel flexible Skyline service discovery method with service quality perception

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050091344A1 (en) * 2003-10-23 2005-04-28 International Business Machines Corporation Methods and sytems for dynamically reconfigurable load balancing
CN1791117A (en) * 2005-12-26 2006-06-21 北京航空航天大学 Service computing system based on service and underlying resource separation

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050091344A1 (en) * 2003-10-23 2005-04-28 International Business Machines Corporation Methods and sytems for dynamically reconfigurable load balancing
CN1791117A (en) * 2005-12-26 2006-06-21 北京航空航天大学 Service computing system based on service and underlying resource separation

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
吴健等: "基于Skyline的QoS感知的动态服务选择", 《计算机学报》 *

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103209102A (en) * 2013-03-11 2013-07-17 北京邮电大学 Web quality of service distributed measurement system and method
CN103209102B (en) * 2013-03-11 2015-11-04 北京邮电大学 Distributed Measuring System and Method for Web Service Quality
CN103164287B (en) * 2013-03-22 2016-04-06 河海大学 The distributed paralleling calculation platform system that sing on web dynamically participates in
CN103164287A (en) * 2013-03-22 2013-06-19 河海大学 Distributed-type parallel computing platform system based on Web dynamic participation
CN104717294A (en) * 2015-03-23 2015-06-17 浪潮集团有限公司 Data extracting method, main server and cluster
CN104735166B (en) * 2015-04-13 2018-05-01 李金忠 The Skyline method for service selection annealed based on MapReduce and multi-target simulation
CN104735166A (en) * 2015-04-13 2015-06-24 李金忠 Skyline service selection method based on MapReduce and multi-target simulated annealing
CN105578212A (en) * 2015-12-15 2016-05-11 南京邮电大学 A peer-to-peer streaming media real-time monitoring method under the big data streaming computing platform
CN105578212B (en) * 2015-12-15 2019-02-19 南京邮电大学 A real-time monitoring method for point-to-point streaming media under a streaming computing platform in big data
CN106790536A (en) * 2016-12-21 2017-05-31 国网江西省电力公司信息通信分公司 Composite Web services system of selection based on affairs and QoS
CN106790536B (en) * 2016-12-21 2020-06-05 国网江西省电力公司信息通信分公司 A Combination Web Service Selection Method Based on Transaction and QoS
CN112787870A (en) * 2021-02-25 2021-05-11 苏州大学 Parallel flexible Skyline service discovery method with service quality perception
CN112787870B (en) * 2021-02-25 2021-11-02 苏州大学 Service quality-aware parallel flexible Skyline service discovery method
WO2022178917A1 (en) * 2021-02-25 2022-09-01 苏州大学 Parallel flexible skyline service discovery method based on quality-of-service perception
US11777820B2 (en) 2021-02-25 2023-10-03 Soochow University QoS-aware parallel flexible Skyline service discovery method

Also Published As

Publication number Publication date
CN102158533B (en) 2013-11-13

Similar Documents

Publication Publication Date Title
CN102158533B (en) Distributed web service selection method based on QoS (Quality of Service)
CN106528773B (en) Map computing system and method based on Spark platform supporting spatial data management
CN111859027B (en) Graph calculation method and device
CN109643312A (en) Trustship query service
CN111258978B (en) Data storage method
CN101256599B (en) System for gathering data of distributing simulation platform based on grid
CN106034160B (en) Distributed computing system and method
CN109151824B (en) Library data service expansion system and method based on 5G architecture
CN102937918B (en) A kind of HDFS runtime data block balance method
CN103067433A (en) Method, device and system of data migration of distributed type storage system
CN102662639A (en) Mapreduce-based multi-GPU (Graphic Processing Unit) cooperative computing method
CN102243660A (en) Data access method and device
CN110175175A (en) Secondary index and range query algorithm between a kind of distributed space based on SPARK
CN108733803A (en) A kind of Multi-User Dimension keyword query method under road network
CN101299198B (en) Dynamic self-adapting graticule data migration method
CN106055590A (en) Power grid data processing method and system based on big data and graph database
CN105608228B (en) A kind of efficient distributed RDF data storage method
US20110179013A1 (en) Search Log Online Analytic Processing
CN108549909A (en) Object classification method based on crowdsourcing and object classification system
Cambazoglu et al. Quantifying performance and quality gains in distributed web search engines
CN108519856A (en) Data block copy placement method based on heterogeneous Hadoop cluster environment
CN109150964B (en) Migratable data management method and service migration method
CN116992887A (en) Metadata data catalog processing method, device and processing equipment
US12045218B2 (en) Contextual geoanalytics engine in a data analytics system
CN107257356B (en) Social user data optimal placement method based on hypergraph segmentation

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C53 Correction of patent of invention or patent application
CB03 Change of inventor or designer information

Inventor after: Wu Jian

Inventor after: Pan Limin

Inventor after: Chen Liang

Inventor after: Yin Jianwei

Inventor after: Li Ying

Inventor after: Deng Shuiguang

Inventor after: Wu Chaohui

Inventor before: Wu Jian

Inventor before: Pan Limin

Inventor before: Chen Liang

Inventor before: Yin Jianwei

Inventor before: Li Ying

Inventor before: Deng Shuiguang

COR Change of bibliographic data

Free format text: CORRECT: INVENTOR; FROM: WU JIAN PAN LIMIN CHEN LIANG YIN JIANWEI LI YING DENG SHUIGUANG TO: WU JIAN PAN LIMIN CHEN LIANG YIN JIANWEI LI YING DENG SHUIGUANG WU ZHAOHUI

C14 Grant of patent or utility model
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20131113