The user can go through UI To create ADF, stay UI Created in ADF when , Users don't need to download separate IDE, And just through Microsoft Edge perhaps Google Chrome browser . The user login Azure Portal, choice “Data factories” service , adopt Data factories Create in service ADF.
One , establish Data Factory example
open Data factories after , Click on “+ Add”, Create your own data factory instance ：
step1, Fill in Basics Information
stay “Create Data Factory” Panel to start creating data factory instance , Fill in first “Basics” Information ：Subscription（ subscribe ）、 Resource group （Resource group）、 Area （Region）、 name （Name） And version （Version）, Version selection V2.
step2： To configure git
stay V2 In the version , When a user creates a data factory , It can also be configured “Git configuration”, For version control , You can check “Configure Git later”, After creating a data factory instance , Optional configuration git.
step3： Check and create
Check （Review+Create） When there is no mistake , Click on “Create” Button to create Data factory example . When the instance is created , Click on Next Step “Go to resource” Navigate to the data factory page .
Two , The author and the monitor
stay Data factory Of overview On the page , Click on "Authoer & Monitor" Button , This will navigate to Azure Data Factory The user interface （UI） On the page .
ADF Of UI The interface is shown in the figure below , The interface shows several commonly used functions ：Create Pipeline、Create Data Flow etc. .
Because this is the first time we created Data Factory, Creating Pipeline Before , We also need to create connections （connection） And datasets （dataset）.
3、 ... and , Create a connection service
Click on UI On the left side of the interface “Manage” tab , First create a connection , There are two types of connections ：Linked services and Integration runtimes, This article creates Liked Services, because Linked Services Depend on Integration runtimes, therefore , Let's first create Integration runtimes.
1, establish Integration runtimes（IR）
How to create Integration runtimes, Please read ：《ADF Third articles ：Integration runtime and Linked Service》
2, establish Linked Services
stay Connections Choose “Linked Services”, Click on “+New”, Create a new Linked Services：
Different data sources , Different Linked Service, According to the actual data source , Choose the right type of data source , The figure below creates Linked Service The type is SQL Server, Input Name、Connect via integration runtime、Server name、Database name、Authentication type 、 User name and Password.
Be careful ,Connect via integration runtime It was created in the previous section Integration runtimes.
Azure Key Vault It's a storage space , The user stores the password in Azure Key Vault in , Input Key Vault You can extract the information it stores .
Four , establish Dataset
dataset Represents the structure of data storage （schema）, It can represent both data sources , Read data from a data source ; It can also represent data targets , Store data in the data target .
Create a dataset example , It only stores metadata information such as the structure of data storage , Instead of actually storing the actual data . Data is actually stored in dataset Point to the underlying storage object , for instance ,dataset perform SQL Server A table in the instance , So the data is actually stored in this table , and dataset The stored data is the table structure and navigation to the table Linked Service. The same dataset, It can be used as a source of data , It can also be used as a data target for storing data .
Click on “ The pencil ” Corresponding “Author” tab , Enter into Fact Resources Interface , Click on “+”, choice Dataset, Go to create Dataset The interface of
Set up Dataset Properties of , Set up Dataset Of Name, adopt Linked service To get the connection to the source data , adopt Table name To specify the table , Make a proposal to Import schema Set to From conneciton/store.
5、 ... and , establish Pipeline
Create pipes , The pipe is equivalent to a container , You can put one or more Activity Drag and drop into the pipeline .
If you put Activity？ Users don't have to write any code , Just from “Activities” Select what you want from the list Activity, Drag and drop Pipeline in , frequently-used Activity Usually located in “General” Subdirectory .
This article demonstrates Copy data Activity Usage of , from “Move & transform” subdirectories , choice Copy data：
Copy Activity The purpose of this is to take data from a dataset Move to another dataset in .
1, Set up Copy Activity Of Source attribute
Source Property represents the data source ,Copy Activity from Source dataset Get data in ：
2,Copy Activity Of Sink attribute
Sink Property is used to set the data target ,Sink dataset For storing data ：
3,Copy Activity Other properties of
Mapping The Properties tab is used to set Source dataset and Sink dataset Column mapping between , And you can set the conversion of column type .
4, debugging Pipeline
Click on “Debug” For the current Pipeline debug
Here we are , A simple ADF Just create it .
Reference documents ：