Power BI Embedded: powerful

analytics in your application

Jakub Wawrzyniak
[email protected]
Sponsors
Jakub Wawrzyniak
• Data Architect & Developer Team Leader at TIDK
• PhD student – Institute of Computing Science, Poznań University of Technology, Laboratory of Algorithm
Design and Programming Systems (algorithm design, complexity analysis, combinatorial optimization,
tasks scheduling)
• Designer and creator of solutions for public administration
• Member of the Organizing Committee of the Game Industry Conference and the Applied Data Science
conference
• Member of the Game Research Association of Poland
• Passionate about games and Japanese culture

e-mail: [email protected] - [email protected]

Agenda
• Introduction

• Embedded analytics

• Power BI Embedded overview

• Demo
 Power BI: experience your data
Any data, any way, anywhere

Web

Mobile
Out-of-the-box SaaS content packs

Real time dashboards & interactive reports Excel

Cloud data
Natural Language query
Embedded
Custom visualizations

Native Office 365 integration Cortana

Power BI
On-premises
“……”
data
Power BI family
Author Share and collaborate Large scale deployments

Power BI Desktop Power BI Pro Power BI Premium

Free data analysis Cloud-based modern Dedicated capacity for

and report authoring tool business analytics solution increased performance
Analytics in your app- Consider your options
Build it
You’re savvy, but building and
maintaining your app
is complex enough

Buy the best fit solution

A fully-managed embedded BI
platform-that’s built for you and
won’t hold you back
Why embedded analytics?

For your business For your developers For your customers

Achieve faster time to market Spend time focusing on your Generate insights through
with your application by product instead of developing easy-to-navigate data-exploration
minimizing development efforts analytics features based on their requirements from
anywhere
Stand out in the market place by Rapidly meet customer report
harnessing the power of a world- demands without development Help customers make quick,
class analytics’ engine informed decisions in context
Leverage seamless integration
Get started for free with with other Microsoft technologies Provide actionable insights
Power BI Desktop like Visual Studio all users including business
executives, power users,
and analysts
 World-class analytics’
Provide insights in real-time and platform with true PaaS
context solution and complete cloud
support

Connect to countless
Why Develop on an open

Power BI
and fully extensible
data sources so that platform with fully
you can expose insights documented SDKs and
to your customers
Embedded API libraries

Take advantage of our existing

Custom Visual library or dream Embed consistent,
up the right visual easy-to-navigate visualization
for your customers experiences across any device
Power BI Embedded is a PaaS resource
in Azure Query Duration

Scale up
Power BI Embedded Pillars

Embed Automate
Seamlessly embed Power BI into Automating enterprise and multi
your application, including all tenant apps scenarios
reporting capabilities, natural Azure elasticity
language support and self-service
BI
Visual embedding + custom layout
Scenario

ISV wants to embed visuals in his app, and

have the ability to optimize size and location.

Enabled capabilities
Show/ hide specific visuals in a page.

Set their size and location in real-time.

Update visibility and sizes during session.

Visuals level API

Mobile support
Scenario
ISV wants to embed reports on his

mobile app, and give his users a better

experience when engaging with reports.

Enabled capabilities
Embed report pages in mobile layout
Support ‘Swipe’ & ‘Zoom’ events
Bookmarks API
Scenario

ISV wants to use bookmarks on his embedded

reports.

Enabled capabilities
Get bookmarks list

Apply bookmarks – on load, during session

Show/Hide/ Nav pane

‘Bookmark applied’ event

Capture state as bookmark (Personalization)

Data connectivity

Same as PowerBI.com
Direct Query | Cached | Real-time | Cloud | On premises
How to start with Power BI Embedded?

Create Provision Embed Deploy

Analyst Developer Developer End users

Power BI Desktop Setup embedded REST & JS API Embedded reports

solution
Row - level security

Report + Token = Query + Predicate

Defines Roles Asserts Role Membership Query executes with correct scope
Roles define predicates Asserts Identity

Role: Contoso
Predicate:  [Company] = “Contoso” roles: “Contoso” Select * FROM [Sales] WHERE [Company] = “Contoso”

Role: Company roles: “Company”

Predicate:  [Company] = USERNAME() username:  “Contoso” SELECT * FROM [Sales] WHERE [Company] = “Contoso”
Define roles on Power BI desktop
1. Go to ‘Manage Roles’

2. Create a new ‘Role’ and define the rule

Automate – ALM APIs
Update data
Assign
Import source Update
workspace to
PBIX connection+ Report
Capacity
Set Parameters

Clone and rebind Clone Embed

Create the ‘golden’ workspace dashboards content in
customer’s content your SaaS app
workspace

Once your ‘golden analytical content’ workspace is available

Embedding
for your
customers
 Application Application
Authorization pattern User
1
Authentication

Access to Power BI is through a

“Master Account”
• AAD Account internal to your Application
application 4 5
Front End
• Has a Power BI Pro license (Browser)
Gives token to Gives token
• Owns the App Workspace where content the user’s to Power BI
is stored browser

Application is responsible for Authenticating 3

Calls GenerateToken API
as the Master user
the user
• Can use any authentication method Application
Power BI
Backend
Application authorizes access to Power BI
using token created with the GenerateToken 2 AAD
API (Master user)
Login and
receive AAD
Token
Capacity based licensing
Total cores
(Backend + DirectQuery/live
Capacity Node frontend) Backend Cores Frontend Cores connection limits Peak renders/hour
A1 1 v-core(s) 0.5 core(s), 3-GB RAM 0.5 cores 0 5 per second 1 – 300

A2 2 v-core(s) 1 core(s), 5-GB RAM 1 cor(e) 10 per second 301 – 600

A3 4 v-core(s) 2 core(s), 10-GB RAM 2 core(s) 15 per second 601 – 1200

A4 8 v-core(s) 4 core(s), 25-GB RAM 4 core(s) 30 per second 1201 – 2400

A5 16 v-core(s) 8 core(s), 50-GB RAM 8 core(s) 60 per second 2401 – 4800

A6 32 v-core(s) 16 core(s), 100-GB RAM 16 core(s) 120 per second 4801 – 9600
Jakub Wawrzyniak
[email protected]
http://tidk.pl
http://aindustry.io
http://dsaas.co