Shawn Moe – IBM Informix Lab [email protected]/15337342/Moe IOT_ Informix 2015-v3.pdf · • Less cloud CPU requirements by precomputing values 2.Reduces Latency • Intelligent

© 2015 IBM Corporation

Internet of Things – Business CasesJune 2015

Shawn Moe – IBM Informix [email protected]


Agenda

� What is the Internet of Things?

� Database to Enable IoT – IBM Informix

� Industry Examples

� Additional IoT Examples

� Summary


The Internet of Things (IoT) is the network of physical objects or "things" embedded with electronics, software, sensors and connectivity to enable it to achieve greater value and service by exchanging data with the manufacturer, operator and/or other connected devices.

Courtesy of Wikipedia














Billions of devices, sensors, and chips that are able to communicate via the Internet makes up “The Internet of Things”

8

“The Internet of Things refers to uniquely identifiable objects (things) and

their virtual representations in an Internet-like structure.” – Wikipedia (link)

“The Internet of Things is the network of physical objects that contain

embedded technology to communicate and sense or interact with their

internal states or the external environment.” – Gartner (link)

“The Internet of Things (IoT) represents a new construct in the information and communications technology (ICT) world

that is occupying the minds of IT vendors, service providers, and systems integrators as it represents huge potential for

new streams of revenue and new customers.” – IDC (link)

“The Internet of Things represents an evolution in which objects are capable of interacting with other objects. Hospitals can

monitor and regulate pacemakers long distance, factories can automatically address production line issues and hotels can

adjust temperature and lighting according to a guest's preferences, to name just a few examples.” – IBM (link)


Forecasts call for billions and billions of connected devices

9

“In 2020, Over 30 Billion Connected Devices Will Be

In Use.” – Gartner (link)

“There will be 212 B devices or things connected to

networks by 2020” - IDC (link)

“Driven by reducing price per connection and the

consequent rapid growth in the number of machine-

to-machine (M2M) connections, we expect the

number of connected objects to reach 50bn by 2020

(2.7% of things in the world).” – Cisco (link)

“There are more than 10 billion wirelessly connected

devices in the market today; with over 30 billion devices

expected by 2020..” – ABI Research (link)

“From vehicles and smart phones to containers and

machines – by 2015 more than six billion things will

be connected to the internet.” – Bosch (link)

Business Insider Intelligence: Global Internet Device Installed Base Forecast

© 2015 IBM Corporation10

Internet of Things Topography

Streams

Deep Analytics Zone

Device/Sensors

Smart Gateways

Sensor Analytics Zone


All Internet of Things applications have a common set of requirements

Requirements:

� Quickly and easily provision new sensors

� Create a real-time communication channel with the sensor

� Capture data from the sensor and store it in a time series database

� Provide access to the collected data

� Pay for the service based on usage

Extended Requirements:

� Provide analytics on the data in both real-time and historical data

� Trigger events based on specific data conditions

� Interact with the sensor from business applications and/or from mobile devices


Simple IoT Topography

�Sensors or devices connect directly to the Cloud

�Cloud-based processing performs actions and/or analytics on the data

�Data may or may not be stored


Another Simple IoT Topography

�Low power sensors communicate with Gateway device

�Gateway communicates with the Cloud

�Gateway may perform some actions before sending data to the cloud

�Cloud-based analytics and processing


Agenda

� What is the Internet of Things (IoT)?




� Summary


1. Reduces Cost

• Less cloud storage by filtering and aggregating data locally

• Less cloud CPU requirements by precomputing values

2. Reduces Latency

• Intelligent gateways detect and respond to local events in real-time vs waiting for transfer to the cloud

3. Security

• Gateways allow customers to capture and get value from their sensors without sending data to the cloud

Over time more and more of the processing will move from the cloud to edge (gateway) devices

Why Use a Gateway Device in an IoT Environment?

15


� Aggregates, sorts, and filters data for consolidated view

� Enables local processing and action for optimized business results

� Reduces latency, bandwidth, and cost of the backend cloud

� Reduces storage requirements

� Performs 80% simple operations locally

� Handles real-time and context computing on the edge

� Delivers analytics and insights for data-in-motion

� Enables immediate action

Gateways and Real-Time Analytics are Key to the Growth of IoT

16


Why have a database at the Edge?

�Common operations at the edge are:–Sorting, aggregation, duplicate elimination, data cleansing, correlation, prediction, etc…

�Response time is often critical–Must be fast enough to store and respond to events in near real-time

�A distributed view of data across devices may be required

�All characteristics of a DBMS….

17


EmbeddedDevelopers

Informix: Intelligent Database

for Internet of Things•Real-time data analytics

•Time-series data management

•Hybrid database support

•Embedded database for edge

devices

•Rapid web and mobile application

development

Edge Devices

Gateways

Smart HomeSmart AutoMobile

Devices

One Database -For the Cloud & Edge Devices

IBM Informix: Intelligent Database for Internet of Things


Why IBM Informix?

� Native support for time series and spatial/GIS data

� Sensor data always contains time series data

� Easily embedded into a device

� Install footprint less than 50 MB and memory required as low as 64 MB

� Hands-free administration

� Storage and retrieval capability

� Analytics built into the database

� Supports JSON/BSON and SQL apps simultaneously in the same database

� Scales-out across multiple devices

� Informix Warehouse Accelerator improves the performance sensor analytics to

speed complex queries for relational and Time-Series up to 1000x faster.

� Extremely high levels of performance and availability, distinctive capabilities in

data replication and scalability, and no administrative overhead.

Informix is Uniquely Positioned for Sensor Analytics

19

Sensor data is TimeSeries data!!!


Massive volume of device dataTons of Real-time data generated at different rates from sensor devices and RFID tags need to be stored and queried with storage savings and high performance

Understanding where things areHandle location-based data, querying spatial attributes of the data, combining spatial and non-spatial data within the same query

Making sense of device-generated dataAbility to apply analytics to get insights and actionable information from sensor-based data; being able to understand behavior, patterns and trends of sensor data

Acting on different data structure and needsAbility to address the complexity of storing and querying both structured and unstructured data, ability to effectively handle mixed data requirements and needs

Fast deployment, dynamic and uncertain environmentAbility to deploy IoT services fast; ability to offer security and privacy in devices and clouds, always-on and fast platform; ability to scale out and respond under uncertain environment

1

2

3

4

5

Time Series / Smart Sensor Data AwarenessDatabase with smart built-in timeseries technology that offers high storage savings, query performance and flexibility for time-stamped sensor data

Location-Based Data and ServicesDatabase with smart built-in spatial data technology for speed and intelligence at storing and querying location-based data

Real-time Analytics on Time Series DataDatabase that enables extreme performance technology for analytic queries on large amount of time-stamped sensors data

Hybrid Database: Structured and Unstructured DataDatabase that enables of structured (SQL) and unstructured (NoSQL) data, ability to handle SQL and JSON/BSON; ability to handle mixed workload with flexibility and performance

Always-On, Robust Embedded Database for CloudDatabase for deployment in both Clouds and Devices with highest levels of Availability, Security, Performance, Robustness, Scalability and Flexibility to cope with the uncertainties, rapid service deployment and SLA needs in embedded systems, cloud and shared platforms

Addressing Database Challenges to Leveraging the IoT


Agenda





� Summary


Example of Healthcare Monitoring for Assisted Living

Informix Historian

Operational Analytics

Notification to Assisted Living Central Monitoring Station Change patients medication, closer

monitoring, prevent stroke

1111

2222

3333

Patient/Care giver

Hundreds of patients

Thousands of devices

Locally Act Upon Insights

Data Consolidation Gateway

Sensor Data Input

Display Alerts and Recommended Actions

4444

5555

Collection and analysis of data for all devices across

assisted living facilities

Assisted Living Corporation changes food sodium usage based on trend of high blood pressure

Filter critical and life-saving data

Blood pressure threshold exceeded

• Embedded at device/gateway

• Local decision making at Facility

• Leverage all data: NoSQL/SQL & Timeseries data


Changing Business Models – Health Care Monitoring

Automatic sensors to monitor well being

Pendants, shower & bath buttons

Activity sensors – rising in the morning, taking meds, using the fridge

Bed & Chair sensors for inactivity monitoring

Outside alarms to alert neighbors

1

2

3

4

5


Example of Connected Home

Informix Historian

Operational Analytics

Adjusts HVAC settings during peak times or vacant hours

Remotely control settings unexpected weather changes

1111

2222

3333

• Embedded at device/gateway

• Filtering and consolidation

• Local decision making

• NoSQL/SQL & Timeseries data

Collect data from multiple devices

Collect temperature data

Locally Act Upon Insights

Data Consolidation Gateway

Display Alerts and Recommended Actions

4444

5555

Analysis of data across all facilities

(authorized access)Compare to citywide energy usage, adjust settings or prepare for alternate energy resources

Thousands of data points from all household devices consolidated

Temperatures compared with thousands of monthly averages

Sensor Data Input

© 2015 IBM Corporation•25

Sensors in appliances proactively monitor efficiency

and failure conditions, and provide insight into usage

�High cost of warranty claims: $11 per phone call, $150 per service call, 2-3 truck rolls per repair, 70 million devices

� Limited service opportunities

� Lack of customer insight and limited repeat purchases

Business Challenges� Improve repeat purchases by understanding who buyers are, usage

patterns, and refresh indicators

� Improve service revenue opportunities by offering more proactive monitoring and response

�Reduce cost of warranty claims through pre-visit triage and diagnosis

�Sell usage data to partners to provide automated replenishment of supplies

� Link to smart meters for energy optimization

�Provide insight into features customers don’t use

Opportunities for Innovation

Streaming data from appliance or retrofit plug adapter

Real time analytics to determine alert conditions

Notification sent to repair and support, and/or customer

Historical analytics to determine failure patterns, and optimizations

Automated work orders generated for specific failure conditions

Commerce interface for selling supplies & accessories (incl. partners)

1

2

3

4

5

6

Reduce Product

Maintenance Cost

Improve Revenue

via Customer Insight

Changing Business Models – Home Appliance Monitoring


Changing Business Models – Vehicle Maintenance

One of the largest freight railroad companies is collecting data throughout its railroad network from sensors and detectors to reduce maintenance costs and increase revenues.

Data is collected and consolidated and then sent back to an Informix database for storage and analysis.

Analytics is performed on an ongoing basis to determine potential repair needs and to determine operational effectiveness.

Sensor data is then sent to the cloud where the data is used for online analytical processing

There is a variety of data collected by various sensors. Over the next few year expected to have ~30 different types of sensor types, including video data collection

1

2

Data Center3

4

5


Changing Business Models – Vehicle Tracking

Use map overlays to show hotspots of traffic density

Data is collected and consolidated and then sent back to an Informix database for storage and analysis.

Phone app to access data for your vehicle

Sensor data is then sent to the cloud where the data is used for online analytical processing

Vehicle resident sensing – Monitors engine operation including fluid levels, tire pressure, seat belt, air bag usage

Onboard help using GPS for weather and traffic conditions

Remote starting in bad weather situations

1

2

3

4

5

6

7


Changing Business Models – Network Monitoring

Monitoring network gateways and routers

Performance checks and cross correlation of similar devices on the network

Look for possible failure nodes before they happen

Congestion monitoring & prediction

Perform re-routing as desired

1

2

3

4

5


Projected Value from Top Ten Connected Applications in 2020

Application Value

Connected Car $600 billion

Clinical Remote Monitoring $350 billion

Assisting Living $270 billion

Home and Building Security $250 billion

Pay-As-You-Drive Car Insurance

$245 billion

Car Monitoring & Usage $225 billion

Smart Meters $105 billion

Traffic Management $100 billion

Electric Vehicle Charging $75 billion

Building Automation $40 billion

Source:http://www.globaltelecomsbusiness.com/article/2985699/Connected-devices-will-be-worth-45t.html(2013)


Agenda





� Summary


The Internet of Things (IoT) Creates Opportunities Everywhere

Bus Route Optimization

Neonatal Monitoring

Smart Power Grids

Building PowerManagement

Network Management BorderControl

Water PollutionMonitoring

Food Safety

Data CenterSecurity

Predict Impact on Infrastructure

DerailmentAvoidance/ Planned

Rail Maintenance

Pharmaceuticals Monitoring

Device, Machine or Sensor data: (noun) information which was automatically created from a computer process, application, chip or other machine without the intervention of a human


IoT Creates Opportunities Across Industries

Banking Healthcare Automotive Retail Transport E&U

Monetize

Cash replacement

solutions

Mobile

Banking

Paid home care family

services

Pay-per-drive car rental

Cash replacement

Sensor enabled Loyalty cards

Paid Alerts to travellers

Congestion charging

Pay-per-use energy

Optimize

Optimized Cash

management

ER Bed Resource

Mgmt

Component predictive

replacement

Fleet mgmt

Delivery and stock replenishment

optimization

Store layout optimization

Smart Cities Traffic mgmt

Airport Management

Delay non-essential supply

during peak loads

Extend

Banking the un-banked

Biometrics

Smarter

Subsidies

Life style monitoring

In-car Movies, Music, Games

Highly Automated

Driving

Smart Vending Machines

Delivery Lockers

Mobility Services

Smart home services

Control

Remote ATM Management

Dynamic Authorization

Remote Hospital

environment Mgmt

Remote

Drive-train optimization

Store energy mgmt

Store parking mgmt

Dynamic price labels

Crowd mgmt

Timetable mgmt

Asset mgmt

Remotely control consumer

devices


Automotive Examples – Connected Car


Energy & Utilities Examples – Smart Energy


Retail Examples – Smart Commerce


Healthcare Examples – Smart Health


Agricultural Examples – Smart Food


Industrial Examples – Smart Manufacturing


Agenda





� Summary


Why is Technology Adopted?

40

It offers clear improvement in at least one of the following areas: security, safety, efficiency, or quality of life

A solution is compelling not because it uses technology X or Y, but rather it is through the improvement of life


Technology shift: Consumers lead the Internet of ThingsToday Tomorrow Integration

A few connected devices per person…

Almost every device that consumers own will be connected, and many new ones will be created to leverage

the value created by consumer connections.

Cross-platform integrators will

connected devices and automate

personal activity:Door Lock

Dishwasher

Clothes Washer

Clothes Dryer

Window Lock

Garage Door

Toothbrush

Garden Moisture

Coffee Maker

Home Lights

Examples:

Ifttt.comZapier.com

� Just as consumers have led enterprises in embracing new mobile services, we believe

they will lead the adoption of connected devices & integrated services


Focus on the Person, not the Technology…

“I am convinced that if we are to get on the right side of the world revolution, we as a nation must undergo a radical revolution of values. We must rapidly begin to shift from a "thing-oriented" society to a "person-centered" society.

When machines and computers, profit motives and property rights are considered more important than people, the giant triplets of racism, materialism, and militarism are incapable of being conquered.”

Martin Luther King, Jr.

April 4, 1967

© 2015 IBM Corporation43

Thank You

Shawn Moe – [email protected]

Documents

Shawn Moe – IBM Informix Lab [email protected]/15337342/Moe IOT_ Informix 2015-v3.pdf · • Less cloud CPU requirements by precomputing values 2.Reduces Latency • Intelligent