User Manual for GeoJModelBuilder
Contents
1. 2 4 Check QoS information Finally we click Finish button to get the final evaluation result as shown in Figure 3 2 4 The final QoS value normalizes between 0 and 1 QoS Evaluation Result List Name Version GeoBufferProcess http geopw whu edu co s080 wps MWebProcessingS3er 41 GeoBufferProcess http geos whu edu cno 8080 wps MWebProcessingServ Figure 3 2 4 Evaluation Result If the Service Proxy does not have attached QoS data we can do a QoS Test to 7 get partial information which includes execution time availability reliability about the service Click Do Test button input a number as the request times in the popup dialog as shown in Figure 3 2 5 QoS Evaluation Wizard QoS Evaluation Choose QoS factors Step 3 Check QoS information and discover services Determine Weights Check 005 Information QoS Information of GeoBufferProcess Throughput Performance Availabilit You will get the execution time availability and reliability of the service in the test Please set number of request times for doing online test Figure 3 2 5 QoS Test ClickOK button to do this quality test the progress bar on right bottom corner display the progress of the test When the test is done we can get the test result as shown in Figure 3 2 6 Name GeoBufferProcess Uri hitp igeopw wWhu edu cn 6080 iwpsiWebProcessin Qo5 Information Throughput Performance Figure 3 2 6 Test Result 3 4 Service Chain Op2. evaluation methods direct and indirect The direct evaluation methods determine data quality through the comparison of the data with reference information Indirect evaluation methods infer or estimate data quality using provenance The quality of geospatial data is generally measured by their completeness logical consistency positional accuracy thematic accuracy and temporal accuracy If the geoprocessing service record data provenance we can get the information to help us evaluate the data quality In the Check Qos Panel expand Data Quality node click Provenance node click Run button to execute the service and then we will get the provenance information of the data in the processing As shown in Figure 3 2 1 QoS factors Performance Availability Reliability Reputation Data Quality Provenance Information List Run service to get its proveance information Provenance Visualization Statistics Evaluation OutputFormat service URL Service Name GeoBufferProcess Value of Data Quality Indicator First Level Positional Accuracy Second Levet Figure 3 2 1 Data Proveance 3 3 QoS Evaluation In order to get the QoS evaluation value there are two things we must to do Choose the QoS factors you want to use in the evaluation process 2 Determine the weights for each QoS factors Select service model you want to get its evaluation QoS value in the working panel Click menu QoS Service Evaluation open e
3. Run Run to run the service and you will get your result you wanted 4 Water Extract Case Here we offer you a case to optimize service chain to get an optimal services composition In this case we use past MODIS images as our data Red band and NIR band of each image go through the Normalized Difference Vegetation Index NDVI calculation binaryzation and rendering processes to derive the water body from images 4 1 Builder Service Chain Click the Add Model button OF toolbar to add an abstract service model set J ti parameters of model as shown in Figure 4 1 1 Then click Add Flow button on the toolbar to chain the abstract models together we can get an abstract service chain as shown in Figure 4 1 2 15 Name NDE Abstract Po Parameter List Complex Parameter Complex Parameter Figure 4 1 1 New a service model Figure 4 1 2 Build service chain 4 2 Service Chain Optimization Having built your service chain it s time to do the optimization namely to select a best candidate service set to run the service chain Click menu QoS Chain Evaluation open optimization wizard as shown in Figure 3 3 3 16 service Chain Optimization Local Selection Global Optimization Step 1 Set the keyword for each abstract service Abstract Services RasterMapcalcProcess Keywords RasterMapcalcProcess Step 2 Set the QoS filters for each abstract service optional Qos Factor The av
4. Services combo box adjust the keywords to get candidate services of this abstract service model The candidate set will display in the Candidate Service List of bottom table We can click on the table cell to select a candidate then we can click Detail button to check its detail information click Delete Button to delete the candidate form services set Step 2 is optional we can set the filters for each abstract service and sometimes maybe we want the availability of all the candidate services is greater than 0 8 so the service which has availability less than 0 8 is passed by the filter 2 After get the candidate set of each abstract service model click the Next button we go into second part Global Optimization as shown in Figure 3 3 4 Step 3 1s to choose the QoS factors which you will use in the optimization process Next setp 11 is to determine the weight of each QoS factor and then set the constraints of whole service chain this part metioned before not tired in words here Then we will do the global optimization using Genetic Algorithm you can click Settings button to change the paramters of this optimization alorithm or you can keep its default value Service Chain Optimization Local Selection Global Optimization Step 3 Choose the QoS factor using in the global optimization process Performance Availability Reliability C Reputation C Data Quality Step 4 Determine the weight of each QoS factor optional e S
5. User Manual for GeoQos State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing LIESMARS Wuhan University June 2014 Catalog ls ABDOU COCO ina aaaa ns 1 ki AMO ducati 1 12 COpPYrrOht mito Maa ss 1 Lo Tcen ACTES MEN 1 2a AAA O O l ZA Greoprocessin Service Main cies setae a 1 2 2 Qualy Ol GEOproCeSsin Servicess aT N 1 2 3 Geoprocessing Service Chain OptIMIZAt1OD ooocccnnncnnnnnnnonnnnnnnnnnnnnanacnnnnanannos 2 Se BAS Opera Maia 2 i fl CRC ds 2 32 DUAF CN INC set ate E EN EE 4 39 QOS Aoc 4 SH Service ain O pum Zain id 9 A Water EXtacuUC ISEE das 15 4l Builder Service AAA n eieneaanaie 15 42 SNC Ena OPIO ira 16 43 Check Qosim the World Winds iia 17 AA AA hid haheausooiad duiiace teakadintdablagetenieeeedd 18 1 About GeoQoS 1 1 Introduction GeoQoS is tool for QoS aware geoprocessing It can evaluate the quality of geoprocessing services with users feedback and other quality factors including availability and reliability QoS for each geoprocessing services can be visualized in a Virtual Globe environment to facilitate service selection Users can set QoS constraints for processing nodes in geoprocessing workflows and then optimize workflows with constraints satisfaction The tool has been implemented as an extension to an existing geoprocessing modeling tool GeoJModel Builder 1 2 Copyright information Copyright C 2013 State Key Laboratory of Information Engineerin
6. ailability of the service The reliability of the service The reputation of the service Candidate Service List Version RasterMapcalcProcess http geos wht e RasterMapcalcProcess http geopw wnh RasterMapcalcProcess hitpigeopw whu Figure 3 3 3 Service Chain Optimization According to the wizard we can finish optimization to select the optimal candidate service for each abstract service model 4 3 Check QoS in the WorldWind Click Finish button the QoS information will load to WorldWind you can get an easy way to check some of the information including each single service QoS and the total service chain QoS value As shown in Figue 4 3 1 17 File Project Model Run QoS Data SensorWeb Help walo el x cam 2 8 gt i el al gt gt 10 WorldWind Service List WorldWind Default Tab Layers lv Stars y Atmosphere e NASA Blue Marble Image y Blue Marble WMS 2004 y i cubed Landsat USDA NAIP USDA NAIP USGS x Name RasterColorsProcess _ MS Virtual Earth Aerial ey E 7 EOR Urthttp geopw whu edu cn 8080 wps WebProcessingService Layers Version 0 4 0 Added Data QoS Information Added Service Throughput Time 1123 53 C Default Tabo Availability 1 00 Reliability 0 79 Reputation 3 Star Data Quality Good Performance http geos whu edu cn 8080 wps Webl 100 Km Altitude 1 221 km La
7. ce Method i Analytic Hierarchy Process AHP Tip Please input weight number ranged from 0 to 9 Performance Availability Figure 3 2 2 Determine weights QoS Evaluation Choose Qos factors Step 2 Determine the weights of each QoS factor Determine Weights Check QoS Information O Subjective Experience Method a Analytic Hierarchy Process AHP Performance O Extremely Important Fu Pairwise Comparison 0 Very Important a O Quite Important Performance Availability Reliability 0 Weakly Important Performance 3 115 Equally Important Availability o a Weakly Important Retiabiiy o O O i Quite Important O Very Important O Extremely Important Availability Figure 3 2 3 AHP Next panel shows the QoS information of the service and other recommended service list As shown in Figure 3 2 4 We can adjust the keywords to get candidate service which is similar in functionality Click Detail button we can check some QoS information of selected service or we can click Delete button to delete the service 6 from candidate set QoS Evaluation Choose QoS factors Step 3 Check QoS information and discover services Determine Weights Check QoS Information Qos Information of GeoBufferProcess Throughput Performance Availability Reliability et 1 0 Reputation 3 Star Excel Version HasGioSinto http geopw wht_edu cn 8080 wpsiVebPr 04 i httpigeos whu edu cn s080 iwpsiVebPro oso x Figure 3
8. es which have the same functionality but differ in their quality To provide a better QoS Quality of Service it is primarily necessary to identify all possible QoS requirements for geoprocessing Web Services Here we present a set of QoS attributes for geoprocessing services by complementing regular QoS factors with geospatial factors Their definitions are provided in Figure 2 2 1 geoprocessing services comen sat tutor Figure 2 2 1 QoS Factors Temporal Accuracy 2 3 Geoprocessing Service Chain Optimization The optimization of service chain is to composite the available services to gain a good performance under user s given constrains There are two strategies of optimization local selection strategy and global optimization strategy Local strategy selects the concrete service by combining each optimal candidate of subservices Global optimization can maximize the global QoS according to the global requirements the users given Global optimization is the NP hard problem often handled with heuristic optimization algorithm 3 Basic Operation 3 1 Check QoS Drag a concrete service form the left service list panel and drop it into the current working panel as shown in Figure 3 1 1 ray x File Project Model Run QoS Data SensorWeb Help gt bed Ca elxla 3 es os Leja ale WorldWind Service List WorldWind DefauitTab ServiceBuffer x Y GeosampleProcess E Y GeoCleanProces
9. g in Surveying Mapping and Remote Sensing LIESMARS Wuhan University Contact Peng Yue State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing LIESMARS Wuhan University 129 Luoyu Road Wuhan Hubei China 430079 pyue whu edu cn 1 3 License Agreement This program is free software You can redistribute and or modify it under the terms of the GNU General Public License version 2 as published by the Free Software Foundation Free Software Foundation web page http www fsf org This program is distributed without any warranty even without the implied watranty of merchantability or fitness for a particular purpose See the GNU General Public License for more details http www gnu org licenses gpl 2 0 html 2 Basic Conception 2 1 Geoprocessing Service Chain Geoprocessing operations offered via web services provide the means for building complex web based geospatial applications With the advent and evolution of OGC Web Service Program OWS more and more geoprocessing Web services are published under network environment available for spatial analysis and workflow technology has been widely used to chain the scattered services Geoprocessing services chained together can help solve complex geospatial problems 2 2 Quality of Geoprocessing Service When dealing with a geoprocessing task we may meet the need that how to find 1 out the best feasible one among tens of thousands of geoprocessing servic
10. s yw GeoR2VProcess Y GeoRPlaneProcess Y RasterFlowlineProcess Y RasterModeProcess Y GeoDbselectProcess y RasterWatershedProcess Y GeoBufferProcess M Y GeoNetallocProcess Y GeoRThinProcess Y RastermedianProcess Y GeoRCostProcess Y GeoSelectProcess yw GeoRStatisticsProcess RasterContourProcess Y RasterTerraflowaccumulaj Y RasterTransectProcess Fl Y GeoDistanceProcess RasterGensigsetProcess Y GeoUnivarProcess Y RasterColorsProcess Y RasterParamscaleProces Y RasterRandomProcess Y GeolZcProcess Y GeoTransformationProces Y RasterFlowdensityProces Y GeoGeneralizeProcess Y RasterGrowProcess Y GeoNetsteinerProcess Y RasterDescribeProcess Y RasterBasinsfillProcess Y RasterLakeProcess Figure 3 1 1 A Concrete Service Select the service node and then click menu QoS Information Single Service We can check the QoS information related with this service from the popup dialog as shown in Figure 3 1 2 m QoS factors Availability The detail information about performance Reliability Reputation Data Quality Time 5 For the given benchmark data the execution time is 985 175 Figure 3 1 2 Performance of Service Click QoS factor on the left side panel we can check the detail information of 3 each QoS factor on selected service 3 2 Data Provenance There are two types of data quality
11. service mode Click Bind Model button E and select the service model and then click Add button in the 13 popup dialog We should select concrete service in the following dialog and then bind the model parameters to service parameters As shown in Figure 3 3 7 Name Process Set Process BuildPolylinesProcess Model Parameters FirstinputData Bind Parameters InputData Model Parameter Bind Parameter Figure 3 3 7 Change service binding After binding all the services in the service chain click menu QoS Chain QoS the bind result will display on the WorldWind Move you mouse to the connecting lines you will get the QoS value of this service chain Click on the service icon you can get the QoS information about the service As shown in Figure 3 3 8 14 File Project Model Run QoS Data SensorWeb Help walo e x 1 z eL alalle Lalo WorldWind Service List WorldWind Default Tab Layers lv Stars x Atmosphere NASA Blue Marble Image Blue Marble WMS 2004 Z R El El i cubed Landsat _ USDA NAIP USDA NAIP USGS MS Virtual Earth Aerial dl A Layers EN Added Data Added Service http geos whu edu cn 8080 wps Webl _ Default Tab0 _ Default Tab1 l Tome al Altitude 709km Lat 31 4547 Lon 114 6687 Eley 22 meters Downloading Figure 3 3 8 Now you can click menu
12. t 28 6461 Lon 116 8072 Elev 20 meters Downloading Figure 4 3 1 Check QoS Information 4 4 Run the Service Chain Run Run to run the service and you will get your result you wanted The result can be added to the WorldWind click e button and then click Load in WorldWind in the popup dialog As show in Figure 4 4 1 File Project Model Run QoS Data SensorWeb Help a aoe le 7 Seal S rea al Pa ajaja exa 3 wal gt i alal i alo Worlawind Service List WorldWind Default Tab Layers Stars Atmosphere NASA Blue Marble Ima Blue Marble WMS 201 i cubed Landsat USDA NAIP lo El El El El RE USDA NAIP USGS MS Virtual Earth Aeria v E il gt Layers Added Data 2879b9a0 cd3c 1031 bdee fo Added Service http geos whu edu cn 8080M 50 Km AS 4 Ni b Altitude 715km Off Globe Downloading Figure 4 4 1 Display Result Data 18
13. timization First you should do is to build your own service chain Click the Add Model button OF toolbar to add an abstract service model set parameters of model as J HE shown in Figure 3 3 1 Then click Add Flow button L on the toolbar to chain the abstract models together we can get an abstract service chain as shown in Figure 3 3 2 Abstract Po Parameter List Complex Parameter NIRBand Complex Parameter Figure 3 3 1 Create service model Figure 3 3 2 Service chain model Having built your service chain it s time to do the optimization namely to select a best candidate service set to run the service chain Click menu QoS Chain Evaluation open optimization wizard as shown in Figure 3 3 3 10 service Chain Optimization ocal Selection Global Optimization Step 1 Set the keyword for each abstract service Abstract Services RasterMapcalcProcess Keywords RasterMapcalcProcess Step 2 Set the QoS filters for each abstract service optional Qos Factor The availability of the service The reliability of the service gt 00 iw The reputation of the service Candidate Service List Version RasterMapcalcProcess http geos wht e RasterMapcalcProcess http geopw wnh RasterMapcalcProcess hitpigeopw whu Figure 3 3 3 Local Selection 1 First is Local Selection part Step 1 is to set the keywords for each abstract service to get its candidate Select item of Abstract
14. ubject Experence Method gt Analytic Hierarchy Process AHP Step 5 Set the constraints of whole service chain optional QoS Factor The availability of the service chain The reliability of the service chain z The reputation of the service chain gt Optimization Method Figure 3 3 4 Global Optimization Click Finish button we can get the optimization result as shown in Figure 3 3 5 In the result dialog we can see the model and its best candidate Click OK button the selected candidate will be binded to abstract service model If the checkbox of Show Chain in WorldWind is selected the optimization result will visualize in the WorldWind as shown in Figure 3 3 6 12 Service Chain Optimization Result a Optimization Result Model Name service Name service Url RasterMapcalcProcess RasterMapcalcProcess http geopw whu edu cn 80804WwpsiWebProcessingService RasterBinaryProcess RasterBinaryProcess http geopw whu edu co 8080 hnpsAWYebProcessingService RasterColorsProcess RasterColorsProcess http geopw whu edu cno 30804wpsiWebProcessingservice Show Chain QoS in WorldWind Figure 3 3 5 Optimization Result Zaoyano SEE tasterBinaryProcess m TE Macheng Jingmen fingcheng Jinolina Wuhan Huangzhou ruaniin Ezhou Zhicheng mm asterMapcaicProcess manning julin Figure 3 3 6 QoS Visulization What more you can bind other concrete service to you
15. valuation wizard as shown in Figure 3 2 1 Choose factors we want to use in the service evaluation process 205 Evaluation Step 1 Choose the QoS factors using in the evaluation process Determine Weights Check QoS Information Performance The performance of a web service represents how fast a service request can be completed Availability Availability is the probability that the web service is ready for immediate consumption Reliability Reliability represents the ability of a web service to perform its required functions C Reputation The reputation represents the general estimation how good the service is C Data Quality The quality of spatial data after geoprocessing by feedback Figure 3 2 1 Choose QoS factors Then click Next button for third step in this part we need to determine the weight of each QoS factors There are two way to set the weights Subjective Experience Method and Analytic Hierarchy Process AHP If we choose the former one we need to give a number for each QoS factors according our own experience as shown in Figure 3 2 2 If we choose the AHP we should click the cell of Fuzzy Pairwise Comparison Table and then select the importance degree for each two factors the table cell will display the degree value automatically As shown in Figure 3 2 3 oS Evaluation Choose QoS factors Step 2 Determine the weights of each QoS factor Determine Weights Check QoS Information a Subjective Experien
Download Pdf Manuals
Related Search