Data Preparation

IServer Manager provides registration of different data sources to iServer. According to different storage locations of data, iServer now supports the following two methods for registration:

Register user self-managed storage: Also known as IServer Registration Data, register the disk directory, HDFS Directory, and database in the iServer through the Data Registration function in the iServer service page. Enables services provided by the iServer to access and use the data.
Register SuperMap managed storage: This is associated with the iServer through the iServer DataStore package. When you configure a Relational Data Save in the DataStore, a PostgreSQL database is automatically created to store the Dataset. Access data through the iServer Big Data Service, and then download, Upload, and analyze the data.

IServer Registration Data

The user can register the database or the directory where the data file is located with the iServer, so that the service provided by the iServer can access and use the data. The iServer data registry provides four storage types: big data shared directory, Spatial data library, distributed slice library, and Machine Learning model. When the data is Ready, the iServer will automatically list the Datasets that meet the analysis criteria when creating various big data Analysis Tasks.

Sign up for big data file sharing

The iServer Distributed Analysis service provides the ability to process and analyze CSV data and udbx Dataset. You can register shared or Local File directories on the network, as well as HDFS directories, to the iServer. The CSV files, index files, udbx Dataset files, and subfolders in the directory are used by the Distributed Analysis service.

On the iServer service management page, click Data → Data Registration → Register Data Save, set Data Save Type in the Register Data Save panel, and select Big Data File Sharing.
File share types are available in the following two options:

Shared directory: CSV data and udbx data stored locally or in a file share can be registered in iServer for Distributed Analysis or Machine Learning analysis. Among them, the registered CSV data supports modifying the Type of the data.
HDFS Directory: In order to better adapt to large-scale data GIS applications, it is recommended that you use HDFS (Hadoop Distributed File System). It has high fault tolerance and is suitable for large-scale Dataset applications. IServer supports the registration of CSV data stored in HDFS (the registered CSV data supports modifying the Type of data) and index file data.

Enter the path of the share where the registration data is located in Share Directory, or the path of the CSV data in HDFS Directory.
After setting the above parameters, click the Register Data Save button to register the data in the iServer.

Register Spatial Data Library

Log in to the service manager and click Data -- Data Registration. All registered Data Saves are displayed in the Data Save list. Click the storage ID to view detailed storage configuration information.

The specific configuration information is as follows:

Store ID: Create a Unique Identification
Data Save type: Select Spatial Data Library.
Database type: HBase, HDFS, Oracle, PostgreSQL and PostGIS are currently supported.
Server Address: The database connection address.
Database: The database you will use.
Username: Username of the database owner (except HBASE and HDFS).
Password: The password of the database owner (except HBASE and HDFS).
Allow Editing: Currently only registered HBase, HDFS, PostgreSQL, POSTGIS databases support this option. If the Allow Editing "is not checked, the database cannot be used as a storage node for directory service uploading relational data. Check "Allow Editing": The database can be used as a storage node for the directory service to upload relational data. As a storage node for uploaded relational data, follow these guidelines:
- When multiple Allow Editing databases are registered, the database with the least Select Dataset is preferred as the storage node.
- When the Allow Editing database and iServer DataStore are registered at the same time, the Allow Editing database is selected as the storage node first, and the principle of minimum Dataset is followed.
- If only the iServer DataStore is registered, the iServer DataStore is available as a storage node.

Note: During data registration, if the iServer service is not on the same machine as the HBase cluster, you need to add the IP and host name of the machine where each node of the HBase cluster is located to the hosts file of the machine where the iServer service is located.

You can also register the Spatial Data database by modifying the data directory service Config File. For specific operations, add the following configuration information in the iserver- datacatalog. XML:

<datastore> 
<datastoreType>SPATIAL</datastoreType> 
<name>postgresql2</name>
<type>POSTGRESQL</type>   
  <connectionInfo>  
    <dataBase>postgres2</dataBase>   
    <engineType>POSTGRESQL</engineType>   
    <password>iserver</password>   
    <server>192.168.17.116</server>   
    <user>iserver</user>   
    <connect>false</connect>   
    <exclusive>false</exclusive>   
    <openLinkTable>false</openLinkTable>   
    <readOnly>false</readOnly>  
  </connectionInfo> 
</datastore>

Where, datastoreType, name and type are required parameters, and dataBase, password, server and user in conectionInfo are also required parameters.

Distributed Slice Library

If you want to store Map Tiles in a distributed way, you need to register the distributed slice library with the iServer before creating a distributed slice task. Currently, the supported types of distributed slicing libraries are MongoDB and OTS. The following describes the process of adding MongoDB storage in detail:

On the iServer service management page (http://supermapiserver:8090/iserver/manager), click Data → Data Registration → Registration Data Save, and set the following parameters:
Storage ID: User-defined storage ID name, Unique Identification created for database
Data Save Type: Select Distributed Slice Library.
Tile Storage Type: Select MongoDB. Click Add MongoDB Server Address.
Server Address: refers to the address of the MongoDB service in the format of { IP }: { port }.
Add Replication Set: If checked, copies are created for the data in the Distributed Slice Library. Through the replication set mechanism, the automatic failover of MongoDB storage system can be realized, and the availability of data can be improved.
Database Name: The database used to store Map Tiles. By default, a database named smtiles is used.
User Name: The user who has permission to access the database. That is, the user name and password you set for the database you will be using before you enter them.
Password: The password that has been set.
Click the Register Data Save button below to complete the data registration.

Machine Learning Model

The iServerMachine Learning service can provide capabilities such as regression prediction for Vector Dataset in UDB, and image classification and Object Detection for Image Dataset. You can register a Model File in iServer for use with the Machine Learning service.

On the iServer service management page (http://supermapiserver:8090/iserver/manager), click Data → Data Registration → Registration Data Save, and set the following parameters:

Store ID: Customize the store ID name.
Data Save Type: Select Machine Learning Model.
Model File Path: Fill in the directory path where the SDM file is located, such as/home/supermap/model/binary _ building.

Click the Register Data Save button below to complete the data registration.

iServer DataStore

The iServer DataStore is an Application that allows you to quickly create and associate a Data Save with an iServer. When you configure a Relational Data Save in the DataStore, a PostgreSQL database is automatically created to store the Dataset. Access data through the iServer Big Data Service, and then download, Upload, and analyze the data.

Environmental construction

The default port of DataStore is 8020. Please modify the firewall configuration first so that the port can be used. If your DataStore will be installed on a Windows system, verify that vcredist 2013 is installed, and if not, the installation files are provided in the DataStore package. Located in the DataStore package \ support \ vcredist \ vcredist _ x64-2013.exe.

In the bin folder of the iServer Data Store package, Start service;
Enter the address: HTTP://{ IP }: 8020 in the browser to access the DataStore configuration wizard, and enter the user name and password of the iServer administrator;

Specify the content directory to store data, logs, backup files, and so on. The default is Data Store installation directory \ data. Note: The content directory of DataStore does not support paths containing Chinese characters and spaces;

Select the Data Type to be created. Currently, relational data, binary file, slice cache, and Spatio-temporal data are supported; the storage capacity of binary data and slice data depends on the MongoDB database, and MongoDB needs to be installed on the local machine. After selecting Create Binary Data Save or Slice Data Save, specify the installation path of native MongoDB, for example, D: \ mongodb;
Click Finish and wait for the configuration to succeed.
After the configuration is successful, the "Configuration Status" page will prompt that the relationship component of iServerDataStore has been successfully configured to iServer. Click "Data Directory Service" on this page, log in on the jumped page, and click "dataimport" to Select data and upload;

Select the corresponding data in the Data Import interface and click the "Upload" button to upload the data to the DataStore. Support the import of UDB Datasource, CSV File, Workspace, Excel File and GeoJson File, and the import format of udbDatasource and Workspace shall be *.zip compressed file.

After the Dataset is uploaded successfully, a unique dataID will be returned, as shown in the figure above, which is used to represent the data package of the uploaded file. Click the Data ID to view the Detail Description information of the data package. The successfully uploaded Dataset will also appear in the "Dataset List" "of the relationship/datasets resource.

After Upload Data, you can view the uploaded data by opening the newly configured iServer DataStore Datasource in the desktop Application.

IServer Registration Data

iServer DataStore

Related topics