GET TO KNOW

AZURE DATA STUDIO

BY MARGARITA NAUMOVA

www.inspirit.bg www.inspir-it.no
About the e-book
SQL SERVER

This e-book introduces Azure Data Studio and helps you discover the main ADS features you can use for
your daily job. It starts with installing and using the basic querying features in ADS and goes through
more advanced concepts in an easy and exciting way by using examples and visuals. It goes beyond the
basics explaining your next steps of developing modules and using notebooks. You will understand how
to organize your files in workspace and even customize your ADS environment. The book is a good
starting companion of your ADS journey.

1
About the author
SQL SERVER

Margarita Naumova is a well-known Bulgarian SQL Expert. Magi has achieved the highest possible SQL
Server Technical Certification in the field – Microsoft Certified Master, making her one of the best SQL
Server Experts Worldwide, among the top one percent of IT professionals in the World.
She has almost 20 years of SQL Server and BI technologies experience delivering consulting and training
in Bulgaria and abroad (UK, Finland, Germany, Sweden, Cyprus, Norway) and is a trusted advisor for
major Bulgarian customers in SQL Server Platform Area.
She works in the area of Designing and Architecting SQL Server and BI Platform Solutions, SQL Server
Performance Tuning and Optimization, High Availability and Disaster Recovery SQL Server Solutions, and
Azure Data Platform solution design.
Magi Naumova is a leader and founder of Bulgarian SQL & BI User Group “Let’s SQL Together!” and is a
regular presenter of the largest IT events in Bulgaria and in Europe.
She is a CEO of two companies - Inspirit LTD located in Sofia and Inspir-IT AS located in Oslo. Both are
Data platform companies delivering services in Bulgaria and Norway. She is an author and founder of
SQL Master Academy – a SQL Training Program that has helped hundreds of specialists to feel
knowledgeable in their daily work or to find an inspirational career path in the world of SQL Server.

2
Table of contents

About the e-book ................................................................................................................................ 1

About the author ................................................................................................................................ 2
Introduction to ADS .......................................................................................................................... 4
First steps in ADS – installing and defining connections................................................. 7
Getting further – writing queries in ADS Query Editor ............................................... 16
Mastering Development – using snippets and notebooks .......................................... 28
Getting organized – using workspaces ................................................................................. 39
Going extra mile - ADS Extensions and customizations ............................................... 43
Conclusion .......................................................................................................................................... 52

3
Introduction to ADS

When it comes to tools it is easy to say that we are in the age of developers’ heaven. There are so many
tools it is hard to choose the right one to work with. When you work with SQL Server and Azure, the
thinks are not different. We have SSMS, VS, Visual Studio Code, ML Studio and Azure Data Studio. It’s
time to differentiate Azure Data Studio among the others.

Azure Data Studio goes further then all those tools and can be a great choice for those who work with
data, develop code in T-SQL but also use Python, Scala and R. It works with many data sources and on
different platforms, it comes with Intellisense, code snippets, integration for source control and an
integrated terminal.

Azure Data Studio is a cross platform and cross database tool which offers a modern editor experience
for managing data across multiple sources with fast intellisense, code snippets, source control
integration, and an integrated terminal. Azure Data Studio is engineered with the data platform user in
mind, with built-in charting of query result-sets and customizable dashboards.
(https://azure.microsoft.com/en-ca/updates/azure-data-studio-is-now-available/ )

The Azure Data Studio is not an entirely new tool, it has its predecessor called SQL Operations Studio,
which was developed with an idea of having a cross platform tool to work with SQL Server not only
under OS Windows like SSMS or VS do, but on Linux and MAC as well. As part of the SQL Server Linux
age, starting SQL Server 2017 and above the releasing of the tool back in November 2017, which covers
many aspects of this multi-platform work with monthly releases since then.

This was how the tool officially released back in the late 2017. Many features have been added and
improved since its first release, but a primary focus of the tool remains the same as a best tools
developer can use on Mac and Linux (even Windows) to develop and manage their SQL data stored on
premises and in Azure.

You probably have already gotten the main difference this tool has comparing to the one we already
know but let me emphasize the most important.

4
ADS run not only on Windows but currently installs on macOS and Linux. You can choose the platform
when you download the ADS from the official download page: https://docs.microsoft.com/en-
us/sql/azure-data-studio/download?view=sql-server-ver15

In Azure Data Studio you can connect to multiple data systems, not just SQL Server. These are: All SQL
Server data sources on premise and in Azure cloud including Azure Managed Instance, Azure SQL
database and Azure Synapse (Azure SQL Datawarehouse), as well as Apache Hadoop HDFS, Apache
Spark™, Postgre SQL Servers. You can create more sources, because the tool is extendable, which is yet
another big advantage of it.

Development environment includes many options a data professional or a database developer would
need in his daily job. Some of them are IntelliSense keyword completion, code snippets, code navigation
for creating and deploying T-SQL code, as well as many options for tuning the environment, staring from
choosing color schema, going through organizing connections in groups and ending up with workspaces,
which are very similar to project experience we have in VS, but is more flexible in including many files
located in different places in a single workspace. The editor experience includes working with extended
object and code search, replacing all occurrences of a code, and multiple integrated terminals running at
the same time, which help us avoid switching in out of the editor each time you want to do some
command line work. By using terminals, we can have PowerShell and CMD running, or even an Azure
Command-line interface (CLI) shell quick command.

There’s support for Git in Azure Data Studio inherited from VS Code. ADS ships with a Git source control
manager (SCM). We still need to install Git (version 2.0.0 or later) before these features are available in
ADS.

Extensibility is probably one of the great benefits of ADS. You use JSON for tuning configuration settings
and customizing the environment including workspaces or adding custom snippets, and creating widgets
to show information about your system. But there is more than that, the tool is built with the
extensibility in mind and adding extensions comes as a new and exciting task.

This allows the tool to work as a platform, which we can extend and customize according to our needs.
Extensions come as additional installations we perform from inside the tool by downloading them from
Git, or we can add extensions from VS Code, we can even develop our own extension. Some extensions
examples coming in the product are SQL Server Agent for managing jobs, SQL Server 2019 (Preview)

5
from connecting to Big Data Cluster, Admin Pack for SQL Server containing SQL Server Agent, Import
from text files, SQL Server Profiler and SQL Server dacpac, SQL Server Schema Compare and many more.

6
First steps in ADS – installing and defining
connections

Installation of ADS is rather simple. We need to download the latest product version first. We can find
the link in SSMS 18. Once the ADS is installed on the machine clicking on the command runs it.

After downloading the product from the official download link, we need to just specify the default folder
and the default choice of creating the Azure Data Studio Start menu folder. It is good to perform an
installation of both Git and Tortoise Git first so that we have this support later in the tool.

The default installation directory is located under the local user’s profile. Also, you will need about 400
MB of free space to complete the install.

When we first start the product, we are asked to enable the preview features. It is a very good idea to
click Yes, because we will have more support on new features and extensions.

7
Then we are welcomed in the tool. From its first screen we can start using the ADS by choosing to
customize the appearance, learn how to find and run commands, getting help or just starting a new
connection or a new query.

The left side of the window is the Activity bar, which contains five important features:

• Connections
• Search
• Explorer

8
 • Source Control
• Extensions

All of them have short keys. Right clicking on the activity bar can show or hide them.

Let us start by discovering connections first and then let’s run a simple query in SQL Server database and
in Azure SQL Database.

We can start creating a connection from multiple places in ADS. The most common is by clicking New
Connection button next to SERVERs group in Connections menu

9
Not surprisingly creating a connection requires providing the server name and the credentials. We can
optionally select a specific database. In this case the connection will always be associated with this
database, and we will have it as a default once we need to create a query.

10
A connection can optionally have a name which will appear in the server list. The tool shows a list of all
recent connections performed.

When start working with one server, we define a connection and it appears in the SERVERS group, then
we will be able to find it easy and reuse it. But what if we have many servers, databases from different
servers, and Azure database connections. It could be useful to be able to group them. This is what Server
Groups are in ADS. Let’s add two server groups with names TEST and PROD and then move our newly
created connection to the TEST server group.

To add a new server group, we use the New Group command button next to SERVERs group in
Connections menu, we choose the color and press OK.

11
We create the second server Group PROD with a red color. Then in order to move our existing
connection to newly created PRDO Group we drag the connection and drop it over the target server
group name bar.

We create another connection in the TEST Server Group by right clicking on the header of the Server
group. An alternative is to specify the server group in connection properties. Now we have our
connections ordered and in groups. And as expected exploring a connection shows all objects just like
SSMS does, so I can see tables and columns.

12
What is still missing is Azure connection. The azure connection group allows the ADS to scan the
available servers and databases in a subscription. But before doing so a signing is required and a
registration of the device, which means I need to authenticate my computer using the device login URL
and a user code provided by ADS. Instructions are easy to follow by clicking Copy&Open button.

Once I register my computer, ADS can scan my subscriptions and lists all possible SQL Connections. This
doesn’t mean we can connect yet. I still need to be sure I have my IP address added to Azure SQL Server
which I need to connect to. Security configuration of Azure connections is outside of ADS scope, but you
need to make sure the IP address is added to the SQL Server firewall list, you can do it in azure portal.
The first time I click on the Azure SQL Server or any database it contains, the ADS Opens the connection

13
menu for me asking for the account password. It’s better to save it for having smooth experience with it
later.

Azure connection can be moved to SERVERS group or any other custom created server group. We click
on Add to Servers button on the right of Azure Database Connection name, which opens the connection,
then we type the password and we specify the server group which we want to have that connection in.

We have all connections grouped and places properly now:

14
Now its time to see how we can use these connections to query data and write code.

15
Getting further – writing queries in ADS Query
Editor

Let us continue our journey by performing queries and data analysis by using ADS Query Editor. We
need to understand the concept of server and database dashboards first. Right clicking on the
connection or a database provides us with two commands – Manage and New Query. The Manage
command opens a dashboard, either server or database dashboard depending on what we have right
clicked.

Dashboard is a combination of widgets for a database or a server. The widgets we have in the dashboard
by default are Tasks and Search, and a header information.

16
We can customize the dashboard by including widgets coming from the queries we type or using sample
insight widget we are provided with like the five slowest queries sample widget or table space usage
widget. We can also customize SERVER dashboard by adding metrics to our server to measure resource
utilization or workload intensity. Let us do the thinks one step at a time.

Lets assume we need to perform analysis. The purpose of our analysis is to analyze product prices so we
can decide how to optimize the price or the profit of some of our products. We need to have this
analysis as a persistent module in our database, so we can reuse it. Let’s find tables we will work with
first. Double clicking on the AdventureWorks database in the SEARCH widget will get us to the database
dashboard. We can search by object type there or by object name or by both. To search for tables we
need to type t:, we type v: for views, sp: for stored procedures and f: for functions.

17
The database has many objects, so we need to filter our search by name. Typing a name of the object,
we are looking for immediately after the object type will get us there. Now we got all the tables which
contain product in their name or schema:

18
Search widget provide very good experience when we need to find an object fast. Table
Production.Product looks like the one we need for our analysis, so we use it. The first step is to build the
query and then based on the result we will decide to enclose it into a stored procedure in our database.

By selecting New Query or pressing Ctrl+N we get our T-SQL Query editor. Starting code typing already
feels the difference in the intelisence.

19
It lists key words we can choose from, recognizes schemas, object names and provides a great help on
typing, we can build a simple query to get all the data from Production.Product table.

Select * is probably not the best query we can run, so we need to be sure the table doesn’t have
hundred of columns and to specify just what we are looking for – price, quantity or any other columns
which will be of potential help in ours analysis.

20
We have a very good help in exploring object definitions. GO TO Definition (or pressing F12 key) will
open the CREATE table script in a new query window without using the same connection meaning it is in
a disconnected state.

Peek definition on the other hand will provide us with an in-query editor window of the create table
command, from which we can explore the list of columns, and once we finish our check we can close the
window and we can continue to work in our query editor. I would say this is very convenient and give a
great experience in constructing queries, and we have short key combination for every command.

We can safely run our query and seems we have columns we can use for our analysis. Let us run the query
to explore them. We will continue with constructing the query by selecting just Subcategory, List Price
and quantity. We will make a join with Production.SubCategories table to include Subcategory name. Our
final query looks ready to run. We can press Shift+Alt+F to apply code formatting or we can choose Format
Document from the context menu:

21
Let us explore the results to find out the data and if the query needs some corrections. The results pane
does not have anything surprising at a first look, we have data in grid with execution time statistics.
Pretty much like SSMS experience.

22
BUT let’s look at the buttons on the right side of the results pane. We have several options for exporting
and working with the result set. We have Export to excel, CSV and JSON, everyone of each creates a

23
corresponding file type file and for xlsx and csv it includes column headers in file! Something we don’t
have in SSMS.

24
We can select just one row by clicking on the row header and we can export it to JSON, this will help a
lot in having a correct JSON template of just one row we can use in developing further modules.

25
We have chart button as well; it provides us with a valuable option of exploring the data by using simple
data visualizations. Clicking on chart adds a tab in the results pane showing the result in a visual.

We tune the data direction to vertical and we got a chart showing our number of products and average
of their price by categories. Easy and fast!

Discovering the data shown in the chart lead us to probably defining tree groups of average pricing
levels, which is worthy to perform analysis on. These are the most expensive categories having average
price level between 800 and 1800, middle range price level between 200 and 800, and low-level price

26
range of categories with product price levels up to 200. We can test this and how the data look like by
putting additional HAVING Clause in our query.

The chart shows the quantity and average price of products by categories fitting in the low-level price
range. It gives interesting view of the data and probably a starting point of discovering is it a good idea
to increase the quantity of some products fitting in this range, this could lead to increasing the profit.
We can Copy the image or even save it a png image file by clicking on Save as image button.

We are ready to include our query into a stored procedure and provide it as a module a user can use for
source of data analysis. Query Editor and its result pane helped us to get a proper query construct fast
and easy.

27
Mastering Development – using snippets and
notebooks
GET

We will use code snippets to create a stored procedure, this will help us re-use our fancy query. Code
snippets are block of reusable code which we can insert or load in our code window by using a
command or hot key combination. ADS makes no exception in providing this experience to us. To load a
snippet, we open a new query window (Ctrl+N) and type sql. This opens a list which we can navigate in
and locate a snippet we want to use.

We Select the item to load, and press ENTER, and the code snippet loads into the window:

We choose sqlCreateStoredProc and the snippet inserts into our query window.

28
We need to fill the StoredProcedureName and SchemaName. To make it easy select
StoredProcedureName, right-click, and select Change All Occurrences or type Ctrl+F2. We type the proc
name as GetProductPriceByRangeAndCategories and all StoredProcedureName entries change to
GetProductPriceByRangeAndCategories.

We keep the schema as dbo, so we end up with Dbo.GetProductPriceByRangeAndCategories as

a consistent name used in the entire script

29
The snippet contains placeholder parameters and body text that needs updating. The EXECUTE
statement also contains placeholder text because it does not know how many parameters the
procedure will have. We will parameterize our HAVING Clause such as user could run the procedure for
a price range. Using the same hot key combination of Change All Occurrences we replace parameters
with their new names. We copy our query and paste it in the body of the stored procedure. A useful
feature of query editor is it shows the type of the column once you hover over it. In ours case the
p.ListPrice is of type money, so we need to align the parameters and change the int to money type.

We change the EXECUTE stored proc row and we are ready for creating and testing our stored proc.

30
We run the code up to line 28 to create our stored proc and then we test it by running the EXECUTE
command

31
Great experience up until now and seems that we have completed our task of creating a module in the
database for product price analysis. Let’s get one step further and create a SQL Notebook to tell the
story of our data discovery and module creation.

A notebook is a new term, coming in an open source initiative called Project Jupiter. Notebooks are
created for sharing text code, images, data insights in the same document.

The text in notebook is formatted using MD Language (MarkDown language), which is also used for
producing .md files in GitHub. MD language is a cross-platform language.

Notebooks are featured in Azure Data Studio, which makes them available with T-SQL kernel. This allows
us to create and share documents which contain text, T-SQL code, query results, and even images and
charts to share database insights and to tell a story or to create runbooks.

To create a new notebook we can use the Command Palette. We can run it from the View Menu or by
pressing its hot key Ctrl+Shift+P. The Command palette is very useful in code typing when you need to
find a command.

32
You start typing and the tool provides you with the list of possible commands to choose from. Trying to
find how to create a new notebook leads us to the New Notebook command which we can choose from
the list or its shortkey combination Alt_Windows+N

A new Notebook opens and we are ready to put our story here.

33
Let's take a look at notebook components first. We have Kernel, which specifies the language we will
write the code in this notebook and forces the engine that will execute it. It will be SQL in our new
notebook. Attach to sets the connection from the list of connections we work with in this notebook.
Code and Text add snippets for typing the corresponding content in our notebook. We can use Run Cells
to run the code and Clear Results to delete all the table results which we can potentially have stored in
the notebook.

Our notebook can be stored as an .ipynb file. We can share it to other users, so we need to be sure it is
trusted, meaning there is no malicious code inside.

Creating content in a notebook can be an exciting experience, but it requires getting to know the MD
language or at least some basic concepts of its syntax. Even if it is not that complex it still has a long list
of elements. As diving into them is far beyond the scope of this book, I am going to use just # element
for formatting Headings of the new notebook.

Code snippets allow to paste or type a code, and this comes with a fully functional code editor and
intelisense.

34
A code snippet has a run button on the top left corner, which allows us to run the code of this snippet
only, and even to include the results in the notebook. If we don’t want the results to appear or be
preserved in the code snippets then we can click the Clear Results button in the command line of the
notebook.

35
Combining text and code snippets allow us to create a very simple but still good looking T-SQL notebook

36
As soon as we are ready we can save our notebook with a name Price AnalysisByProduct.ipynb

37
38
Getting organized – using workspaces
GET

Workspace is a great way of storing all the files and queries we work with as a single store group. We
can add more files later, or we can add even a folder to a workspace. This will make the task of working
with files in it easy and we can open any of them by just clicking on it. We have Explorer menu on the
left side of the tool under the Connections and Search menu. We work with Explorer Menu to organize
files and folder in our workspace. We put all the files including JSON and CSV in a folder ProductAnalysis,
so I can add it to the workspace by clicking on the Add Folder button.

We add the folder and then it appears in the list of workspace folders

39
We can click on files and see outline

40
41
It is not necessary all the files of a workspace to be in the same folder. Saving the workspace is simple,
we Select Save Workspace from the File menu and provide name and location.

42
Going extra mile - ADS Extensions and
customizations
GET

Customizing Azure Data Services is one of its greatest benefits making it rather development platform
for working with SQL Databases rather than just a T-SQL tool. We have many options to customize and
extent the product, all of them intuitive and easy to perform. Customizations usually contain editing or
typing JSON configurations not only using T-SQL Scripts. Let’s get through some examples.

Custom code snippet

We discovered code snippets earlier when we created our stored procedure for product price analysis.
We found that using code snippets is practical and convenient in ADS. We can create our own code
snippets. Let us create a snippet for updating statistics of a table.

Adding a simple custom T-SQL Snippet is possible at it is very easy to perform. Let’s create a custom
snippet

By calling the Command Palette (Ctrl+Shift+P) and searching for Configuring User Snippets. We click on
the Preferences: Configure User Snippets command from the list which lead us to type of snippet we
want to create

We type sql which lists the only option to configure a custom sql snippet – to create our snippet code in
sql.json file. We choose the sql.json option and we are ready to type the create statistics snippet code.

43
Snippet is defined in a JSON format and it has the following main elements:

• Prefix – what triggers the snippet and what we see as a sql. Command when we search for sql snippets
• body – this is the t-sql code we type and we can use variables and placeholders in the way the instruction
suggests
• description – short explanation which user will see when it hovers over the snippet

According to the instructions and the elements needed, the JSON code of our Update statistics on a user
table snippet will look like this:

{
"Update tbl stats": {
"prefix":"sqlUpdateTblStats",
"body":[
"UPDATE STATISTICS ${1:SchemaName}.${2:TableName};",
"GO "

44
],
"description":"Updates statistics on a specified table"
}
}

We type the elements in the sql.json file. And all we need to do is to save the file.

We have the snippet listed in our sql snippets list. We can find it by typing sqlu (coming from UDPATE), it
appears first in the list and we can choose it, fill the SchemaNAme and TableName and run the
command to update stats

45
Adding widget

Widgets are yet another customizable element in ADS. They can appear in dashboards as data insights,

providing an at-a-glance view about object status, resource usage, workload intensity or any other
important measure, including those coming from queries on data in tables.

Azure Data Studio has a built-in sample widget to monitor the space used by tables in a database. We
need to just add it to the user configuration settings JSON file.

To do that we start Command Palette and search for User settings command

We open User Settings and search for dashboard until we find Dashboard>Database Widgets.

46
We click on Edits in settings.json to add the widget description. Typing “dashboard.database.widgets”
will help us to find the place in the file where we need to insert the widget elements. We insert the
description and save the settings.json file

{
"name": "Space Used by Tables",
"gridItemConfig": {
"sizex": 2,

47
 "sizey": 1
},
"widget": {
"table-space-db-insight": null
}
},

48
Then it is just a matter of showing the dashboard of our AdventureWorks database to find the new table
space widget there

49
At the time of writing there is one more out of the box widget for analyzing the top five slowest queries
based on Query Store execution statistics. We can still develop our own custom widgets.

Extensions

Extensions is yet another way of customizing ADS. They provide additional features to the product and
are listed in Extensions command on the left side of the screen. While some of the extensions are listed
within Azure Data Studio. Others you’ll need to go to the Visual Studio Code marketplace to get. For
example, here’s the PowerShell extension: https://marketplace.visualstudio.com/items?itemName=ms-
vscode.PowerShell

Just download the *.vsix file and install that.

50
Every extension contains a detailed description, recommendations and an installation button. In the
description we can find a detailed information on what the extension does and how to use it. We can
even develop extension

51
Conclusion
GET

Based on everything we saw Azure Data Studio is a great tool providing data management and exciting
coding experience for database developers. It is best suited for those who Query SQL databases like SQL
Server, Azure SQL Databases including Managed Instance and PostgreSQL. ADS is the tool when you
work on Linux or MAC, why not on Windows also. Its highly extensible, probably not very mature yet,
but still a great option for those seeking light and flexible coding tool.

52