internet of things, from datastream to insight · your internet of things: from data stream to...

29
Your Internet of Things: from data stream to insight Technology Day 2015

Upload: others

Post on 25-May-2020

5 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

Your Internet of Things: from data stream to insight

Technology Day 2015

Page 2: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

Tim Jacobs Geert Baeke

Presenting the presenters

Page 3: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

Machine Learning

Structure of today’s talk

GenerateData

CaptureData

AnalyzeData

VisualizeData

Page 4: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

Agenda

•Data Generation & Capturing

•Data Analytics & Visualization

•Machine Learning

•Conclusions

Page 5: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

We will be flooded by huge amounts of devices, generating even bigger amounts of data.

Page 6: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

Internet of Things

Source: ITU-T Y.2060 / Internet of Things Global Standards Initiative – http://www.itu.int/en/ITU-T/gsi/iot/

The Internet of Things (IoT) is the [dynamic] network of physical objects or "things" embedded with electronics, software, sensors, and network connectivity,

which enables these objects to collect and exchange data.

Place

Time

Thing

Page 7: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

Devices, protocols, OS’s & standards

Device Types

• Specific(“Industrial”)

• Generic (“Consumer”)

Protocols

• WiFi

• RF (ZigBee, Z-Wave, …)

• Bluetooth (BLE, iBeacon, Eddystone)

• Cellular

• Sigfox/Neul/LoRaWAN

OS’s

• Device specific

• *nix variants

• Windows 10 IoT Core

Standards

• HTTP

• XMPP

• CoAP

• AMQP

• MQTT

• STOMP

• DDS

• VSCP

• …

communication application

Page 8: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

IoT deployment architectures

Communication Endpoint

Device 1

Device 2

Device 3

disconnected mode gateway mode standalone mode

Device 1

Device 2

Device 3

Device 1

Device 2

Device 3

Local Gateway

Page 9: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

IoT data reception

• Third party on-premises platforms

• Your own platform

•Public Cloud-based offerings:• Azure IoT/Event Hubs & Stream Analytics• Amazon Kinesis• Citrix Octoblu• …

Page 10: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

Let’s see if this gets you excited…

Demo: data generation & capturing

Page 11: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

Demo: data generation & capturing

RPi2 Weather Station

Blob Storage

Advanced Analytics

PowerBIEnd-user Analytics

Stream Analytics

IoT/Event Hub

Microsoft Azure

Service Bus

Queue

Custom Applications

Page 12: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

Agenda

•Data Generation & Capturing

•Data Analytics & Visualization

•Machine Learning

•Conclusions

Page 13: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

Data: from BI to analytics

Standard reports

Ad-hoc reports

Query & drilldown

Alerts

Statistical analysis

Forecasting/extrapolation

Predictive modeling

Optimization

Degree of intelligence

Va

lue

Sta

nd

ard

An

aly

tics

Ad

van

ced

An

aly

tics

What happened?

How many? How often? Where?

Where exactly is the problem?

What actions are needed?

Why is this happening?

What if these trends continue?

What will happen next?

What is the best that can happen?

Traditional BI questionsTraditional tooling:

ETL Tools, SQL, Reporting Services & friends

New type of questionsNew tooling: NoSQL,

NewSQL, MapReduce, statistical software, ...

Page 14: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015
Page 15: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

How to store?

How to capture?

How to process?

How to visualize?

How to analyze?

How to report on?

Page 16: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

Azure Data Lake

Page 17: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

Demo: data analytics & visualization

Providing insights…. Minion style

Page 18: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

Demo: data analytics & visualization

RPi2 Weather Station

Blob Storage

Service Bus

Queue

Advanced Analytics

Custom Applications

Stream Analytics

Microsoft Azure

IoT/Event Hub

PowerBIEnd-user Analytics

Page 19: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

Agenda

•Data Generation & Capturing

•Data Analytics & Visualization

•Machine Learning

•Conclusions

Page 20: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

Computer

Input Data

Program

Output

Computer

Input Data

Output

Program

TraditionalProgramming

Machine Learning

Machine Learning Paradigm

Page 21: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

SELECT customers.ID

FROM customers

LEFT JOIN orders ON customers.ID = orders.customer

LEFT JOIN products ON products.id = orders.product

GROUP BY customers.ID

HAVING orders.date > GETDATE() – 30

AND orders.category = ‘MOVIES’

AND SUM(orders.price) > 100

OR products.description LIKE ‘%MINION%’

OR products.category LIKE ‘%ANIMATION%’

Hey Geert, the new “Minions” movie is available on BluRay with a 40% discount!“ ”

Page 22: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

SELECT customers.ID

FROM customers

LEFT JOIN orders ON customers.ID = orders.customer

LEFT JOIN products ON products.id = orders.product

GROUP BY customers.ID

HAVING orders.date > GETDATE() – 30

AND orders.category = ‘MOVIES’

AND SUM(orders.price) > 100

OR products.description LIKE ‘%MINION%’

OR products.category LIKE ‘%ANIMATION%’

Hey Geert, the new “Minions” movie is available on BluRay with a 40% discount!“ ”

90

Use Machine Learning to identify your business logic in your data!

Page 23: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

Demo: Machine Learning

Dunno what that guy is talking about, but after this we can have a drink!

Page 24: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

Demo: Machine Learning

Blob Storage

Service Bus

Queue

Advanced Analytics

Custom Applications

Microsoft Azure

PowerBIEnd-user Analytics

RPi2 Weather Station

Stream Analytics

IoT/Event Hub

Page 25: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

Agenda

•Data Generation & Capturing

•Data Analytics & Visualization

•Machine Learning

•Conclusions

Page 26: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

Conclusions

• IoT is more than just hardware, it’s an entire ecosystem.

• Value in data is both in:• Aggregated / historical data

• Realtime data

• Machine Learning helps to go from reactive to proactive decision making based on your data

• Time to value is shorter in the public cloud

Page 27: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

How can Xylos help?

Data Strategy

Data Architecture

Data Analytics

Data Platform

Page 28: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

Xylos Cloud ServicesXylos Cloud Services

Page 29: Internet of Things, from datastream to insight · Your Internet of Things: from data stream to insight Technology Day 2015

Xylos Cloud Services