Analyze data by using KQL and anomaly detection

Task 03: Analyze data by using KQL and anomaly detection

Introduction

Once data is streaming into the KQL database, you can analyze it using KQL and configure alerts for anomalies.

Description

In this task, you’ll query streaming data and configure anomaly detection and alerts.

Example scenario

Sudden temperature drops may indicate equipment failure and require immediate attention.

Success criteria

Queries return results and alerts trigger when conditions are met.

Learning resources

Kusto Query Language (KQL) basics
Anomaly detection in Fabric

Key steps

01: Query streaming data and configure anomaly detection

Open Microsoft Edge and go to https://app.powerbi.com/.
If prompted, sign in by using the following credentials:

Setting Value

Username @lab.CloudPortalCredential(User1).Username

Temporary Access Pass (TAP) token @lab.CloudPortalCredential(User1).AccessToken
In the left pane, select the Zava@lab.LabInstance.Id workspace.
In the list of resources, select the thermostat KQL database.

Be sure to select the thermostat KQL database resource and not the thermostat Eventhouse resource.
On the command bar for the database, select Query with code.
On the command bar for the database, select Copilot.

Setting	Value
Username	`@lab.CloudPortalCredential(User1).Username`
Temporary Access Pass (TAP) token	`@lab.CloudPortalCredential(User1).AccessToken`

Submit the following prompt:

 What is the average temperature per day?

In the response from Copilot, select Copy to clipboard>
Paste the query from Copilot into the query window.

Sometimes when you paste responses from the Copilot window, in addition to pasting the comments and code, HTML markup gets pasted.

You must remove the HTML markup tags that appear before or after the comments and code.
On the command bar, select Run.
In the left pane, select the Zava@lab.LabInstance.Id workspace.
On the command bar, select + New item.
Search for Activator.
Select Get data.
In the Data section, select RealtimeDataTo-KQLDB-stream.
Select Next.
[]Select Connect.
Select Finish.
Select New Object.
Configure the Build object pane by using the following information:

Field Value

Object Name Thermostat

Unique identifier Device Id

You may not be able to enter information into these fields until you start seeing data in the Event details pane that appears at the bottom of the window.

It may take a couple of minutes before you start seeing data. If you cannot configure the object, close the Build object pane and select New object to reopen the pane.
Select each of the following properties:
- Battery level
- City
- Temp
- StoreID
- EnqueuedTimeUTC
Select Create.
In the Explorer pane, in the Thermostat node, select Battery level.
In the right pane, select New rule.
Configure the alert rule Condition section by using the following information:

Field Value

Condition Decreases below

Value 10

Occurrence Every time the condition is met
Select Save and Start.
Once we have streaming data which matches with the alert condition we should start seeing Email actions getting triggered.

Field	Value
Condition	Decreases below
Value	`10`
Occurrence	Every time the condition is met

Field	Value
Object Name	Thermostat
Unique identifier	Device Id