Big Data Orientation

Lab 3 – Working with NoSQL Data in Microsoft Azure

Overview
Azure supports many services for working with NoSQL data. In this lab, you will explore NoSQL data using
the Cosmos DB service – a highly scalable data store for key-value table, JSON document, and graph data.

Before You Start

To complete this lab, you will need the following:
• A web browser
• A Windows, Linux, or Mac OS X computer

Important: Before you can perform the lab exercises, you must complete the previous labs in this course.

Exercise 1: Working with a Key-Value Table

In this exercise, you will use Cosmos DB to work with key-value data in Azure Table storage.

Provision Azure Cosmos DB for Table Storage

To get started, you must provision an instance of Azure Cosmos DB.

1. In the Microsoft Azure portal, in the menu, click New. Then in the Databases menu, click Azure
Cosmos DB.
2. In the Azure Cosmos DB blade, enter the following settings, and then click Create:
• ID: Enter a unique ID for your Cosmos DB service
• API: Table (key-value)
• Subscription: Select your Azure subscription
• Resource Group: Select the resource group you created in the previous labs
• Location: Select any available region
 • Pin to dashboard: Unselected
3. In the Azure portal, view Notifications to verify that deployment has started. Then wait for the
Cosmos DB service to be deployed (this can take a few minutes.)

Create a Table
Cosmos DB supports Azure Table Storage for key-value pairs.

1. View the blade for the Cosmos DB instance you created for table storage, and click the Data
Explorer tab.
2. In the Data Explorer, click New Table.
3. In the Add Table pane, enter the following details and click OK:
• Table Id: Employees
• Storage capacity: Fixed (10 GB)
• Throughput: 2000
4. In the Data Explorer, expand the new Employees table and select Entities.
5. In the Entities tab, click Add Entity.
6. In the Add Table Entity pane, Enter the following property values (clicking + Add Property as
required) and then click Add Entity.
Property Name Type Value

PartitionKey String Marketing

RowKey String [email protected]

FirstName String Dan

LastName String Drayton

7. Add a second entity with the following properties:
Property Name Type Value

PartitionKey String Marketing

RowKey String [email protected]

FirstName String Joann

LastName String Chambers

PhoneNumber String 555-123-4567

8. Add a third entity with the following properties:
Property Name Type Value

PartitionKey String Human Resources

 RowKey String [email protected]

FirstName String Rosie

LastName String Reeves

PhoneNumber String 555-321-7654

Query a Table
Now that you have some data in your table, you can query it.
1. In the Entities pane, click Add new clause.
2. Set the following properties for the clause:
And/Or Field Type Operator Value

PartitionKey String = Marketing

3. Click  Run to run the query and apply the filter. Note that only the rows where the
PartitionKey value is “Marketing” are returned.
4. Click Add new clause to add a second clause to the query.
5. Set the following properties for the second clause:
And/Or Field Type Operator Value

And LastName String = Chambers

6. Click  Run to run the query and apply the filter. Note that only the row for Joann Chambers is
returned.

Exercise 2: Working with Documents

In this exercise, you will use Azure Cosmos DB to work with a collection of JSON documents.

Provision Azure Cosmos DB for Document Storage

To get started, you must provision an instance of Azure Cosmos DB.

1. In the Microsoft Azure portal, in the menu, click Create a recource. Then in the Databases menu,
click Azure Cosmos DB.
2. In the Basics tab of the Azure Cosmos DB blade, enter the following settings, and then click Review
+ Create:
• Subscription: Select your Azure subscription
• Resource Group: Select the resource group you created in the previous labs
• Account NameID: Enter a unique ID for your Cosmos DB service
• API: SQL
• Location: Select any available region
• Pin to dashboard: Unselected
 3. On the Review + Create tab, click Create.

4. In the Azure portal, view Notifications to verify that deployment has started. Then wait for the
Cosmos DB service to be deployed (this can take a few minutes.)

Create a Document Collection

Cosmos DB supports Document DB databases for collections of JSON documents.

1. View the blade for the Cosmos DB instance you created for document storage, and click the
Data Explorer tab.
2. In the Data Explorer, click New Container.
3. In the Add Container pane, enter the following details and click OK:
• Database id: DocID

• Container id: EmployeeDocs

• Partition Key: /id

• Throughput: 2000
4. In the Data Explorer, expand the new EmployeeDocs collection and select Documents.
5. In the Documents tab, click New Document.
6. Expand the EmployeeDocs node, and select the Documents node.
7. In the document editor, click the New Document button on the toolbar.
8. Replace the default JSON document with the following JSON and then click Save:
{
"id": "1",
"dept": "Marketing",
"email": "[email protected]",
"firstName": "Dan",
"lastName": "Drayton"
}

9. Click New Document and create a second document with the following JSON:
{
"id": "2",
"dept": "Marketing",
"email": "[email protected]",
"firstName": "Joann",
"lastName": "Chambers"
}

10. Click New Document and create a third document with the following JSON:
{
"id": "3",
"dept": "Human Resources",
"email": "[email protected]",
"firstName": "Rosie",
"lastName": "Reeves"
}
Query a Document Collection
Now that you have a collection of documents, you can query it: 1.

In the Data Explorer, click the New SQL Query button on the toolbar.

2. In the Query 1 tab, modify the existing query to use the following SQL statement:

SELECT c.firstName, c.lastName

FROM c
WHERE c.dept = 'Marketing'

3. Click  Execute Query to run the query. Note that only the documents for Dan Drayton and
Joann Chambers are returned.

Exercise 3: Working with a Graph

In this exercise, you will use Azure Cosmos DB to work with a Graph.

Provision Azure Cosmos DB for Graph Storage

To get started, you must provision an instance of Azure Cosmos DB.

1. In the Microsoft Azure portal, in the menu, click New. Then in the Databases menu, click Azure
Cosmos DB.
2. In the Azure Cosmos DB blade, enter the following settings, and then click Create:
• ID: Enter a unique ID for your Cosmos DB service
• API: Gremlin (graph))
• Subscription: Select your Azure subscription
• Resource Group: Select the resource group you created in the previous labs
• Location: Select any available region
• Pin to dashboard: Unselected
3. In the Azure portal, view Notifications to verify that deployment has started. Then wait for the
Cosmos DB service to be deployed (this can take a few minutes.)
Create a Graph Collection
Cosmos DB supports Graph collections that you can manipulate and query using the Gremlin language.

1. View the blade for the Cosmos DB instance you created for document storage; and on the
Overview page, click + Add Graph.
2. In the Add Graph pane, enter the following details and click OK:
• Database Id: GraphDB
• Graph Id: EmployeeGraph
• Storage capacity: Fixed (10 GB)
 • Throughput: 2000
3. On the Keys tab for your graph, note the Primary Key – you will also need this to connect to
your graph from a client app.

Install Node.JS and Run a Script to Populate the Graph

Although you can create vertices in a graph using the Data Explorer interface in the Azure portal, it lacks the
ability to edit vertices and create edges between them. You will therefore use a Node.JS script to populate
your graph.

1. Open a new tab on your browser and navigate to https://nodejs.org/en/download/.

2. Download and install the latest recommended version of Node.JS for your platform (Windows,
Mac OS X, or Linux). Accept all the defaults for the installation.
3. In the folder where you extracted the lab files for this course, in the graph folder, open
config.js in a code or text editor.
4. In config.js, replace the value for the config.endpoint variable with
<cosmosdbsvc>.gremlin.cosmosdb.azure.com (where <cosmosdbsvc> is the name of your
CosmosDB Gremlin service) and the config.primaryKey variable with the Primary Key value
you noted in the Azure portal. Then save and close config.js.
5. Open a command window in the graph folder, and run the following command to install the
gremlin-secure Node package (which is required to connect securely to a Gremlin data
source):
npm install gremlin-secure

6. After the package has been installed, enter the following command to Node app, which will
populate your graph: node app

7. Wait for the app to finish, and then press CTRL + C to exit it and close the command window.
Explore the Graph
Now that your graph contains some vertices and edges, you can explore it in the Data Explorer.

1. Return to the browser tab containing the Azure portal, and view the Data Explorer tab for the
Cosmos DB instance you created for graph data.
2. In Data Explorer, expand EmployeeGraph and click Graph.
3. In the Graph pane, click Load graph. If the graph doesn’t load after a minute or so, in the
GRAPHS pane, click .
4. Note that the graph contains employees that are related to one another based on which
employees follow one another on social media.
5. Select the [email protected] vertex, and note its properties, which include a label indicating
that this vertex represents a person, and an age property
6. Select [email protected] and note that this person does not have an age property.
7. In the filter text box, replace the default g.V() filter with the following Gremlin expression,
and then click Apply Filter.
g. V().hasLabel('person').has('age')
8. View the vertices that are returned, which include all vertices that have a label of person and
an age property.
9. Apply a new filter based on the following expression:
g. V().hasLabel('person').has('age').values('firstName')

10. Note that this returns a JSON string containing the firstName values for each person with an
age property.
11. Apply a new filter based on the following expression:
g. V('[email protected]').outE('follows').inV()

12. Note that this returns the people that are followed by [email protected].
13. Apply a new filter based on the following expression:
g. V('[email protected]').outE('follows').inV().outE('follows').inV()

14. Note that this returns the people that are followed by the people who are followed by
[email protected] (which includes Dan himself, because he follows Joann, who in turn follows
him!)