FIWARE

Presentation@ Eclipse IoT Days

Grenoble, April 28 2016 Gilles Privat, Orange Labs

http://fiware.org
http://lab.fiware.org
The FIWARE Programme (originally FI-PPP)

 Goal: capture opportunities derived from the new wave tourism

of digitalization of life and businesses that is coming
Transport,
e-government
Mobility

 Strategy: Build an ecosystem that will work as catalyst and

Logistics
for capturing the opportunities, engaging data
providers and entrepreneurs
Smart …
Energy Grid
 Pillars:
• FI-WARE : a generic, open standard platform which eHealth
serve the needs of developers in multiple domains
• FIWARE Lab : a meeting point where innovation happens
and data providers plus entrepreneurs can be engaged
• FIWARE Ops : the suite of tools easing deployment and
operation of FI-WARE instance nodes
• : a program that funds developers and
entrepreneurs, and ignites roll-out of the ecosystem
• FIWARE mundus : reach a global footprint, opening to
regions that share the same vision and ambition

1
2
Understanding FIWARE (Open Standard Platform)
(advanced OpenStack-based Cloud + rich library of Generic Enablers)

Driven by
implementation

Sustainability
ensured

3
Why an open standard platform is required

 Avoid vendor lock-in:

• Standard Southbound APIs for sensor providers
• Standard Northbound APIs offered to applications
• Portability across platform providers
• Interoperability of apps on top of different providers
 Larger community of developers (critical mass,
economies of scale)
• True innovation
• More competition, leading to cost savings
 Not any standard is enough
• Modularity
• Allow different business models
• Integration with standard open data platform
• Non-intrusive (smooth integration with legacies)
4
Open Source. Avoid vendor lock-in
You have the control

5
6
FIWARE Generic Enablers (GEs)
 A FIWARE Generic Enabler (GE):
 set of general-purpose platform functions available through APIs
 Building with other GEs a FIWARE Reference Architecture
 FIWARE GE Specifications are open (public and royalty-free)

 FIWARE GE implementation (FIWARE GEi):

 Platform product that implements a given GE Open Spec
 There might be multiple compliant GEis of each GE Open Spec
 One open source reference implementation of each
FIWARE GE (FIWARE GEri):
• Well-known open source license
• Publicly available Technical Roadmap updated in every release
 Available FIWARE GEis, GEris and incubated enablers are
published on the FIWARE Catalogue

7
FIWARE major differential features
• Federation of infrastructures (private/public
Cloud regions)
• Automated GE deployment
Data/Services • Complete Context Management Platform
Delivery • Integration of Data and Media Content

• Easy plug&play of devices using multiple protocols

IoT • Automated Measurements/Action Context updates

• Visualization of data (operation dashboards)

Apps • Publication of data sets/services

• Easy incorporation of advanced 3D and AR features

Web UI • Visual representation of context information

• Security Monitoring
Security • Built-in Identity/Access/Privacy Management

• Advanced networking (SDN) and middleware

I2ND • Interface to robots

8
catalogue.fiware.org

9
FIWARE Chapters

10
Platform  Ecosystem

Open Standard Sustainable Open

Platform Innovation Ecosystem

11
A live instance of FIWARE Not only platform, tools…this is
available to developers for powered by data!
free experimentation

12
 FIWARE Lab: the “meeting point”
where innovation takes place
App Customers and Data providers Entrepreneurs, Developers
• Connect to entrepreneurs • Develop once for a large market
• Put their data at work • Easily meet potential customers
• Bring new innovative services to end • Marketing, promotion
users • Ability to test with real data and end
• Be more efficient users
• Social Reputation • Simple yet powerful APIs that
accelerate product development

 4,2 M€ promotion campaign

• Campus Party events
• Startup Weekend events
• Chambers of Commerce
• 870 K€ in prizes
FIWARE Technology Providers  100 M€ of funding devoted to
• “Coopetitive” approach entrepreneurs in phase 3 of
the FIWARE program
• Connect to entrepreneurs: jointly
exploit the opportunities

13
Extending the FIWARE Lab offering for
service providers and developers
 ff
• 12 nodes in Europe
providing up to 3000+
cores, 16TB+ Ram,
750TB+ HD

• Creation of nodes in
Mexico (1000+ cores)
and Brazil
 4,2 M€ promotion campaign
• • Level
Campus 1 Party
and events
Level 2
• support
Startup Weekend events
for the nodes
• Chambers of Commerce
• 870 K€ in prizes
• Showcases for
 100 M€ of funding devoted to
developers,
entrepreneurs in phase 3 of
infrastructures,
the FIWARE program smart
businesses
FIWARE Lab nodes around the world

15
FIWARE Instances
 Future Internet Applications run on top of “FIWARE Instances” that are built by
“FIWARE Instance Providers” upon:
• selection of FIWARE GEis (products) from the FIWARE Catalogue.
• assembly of selected FIWARE GEis with proprietary added-value products.

Future Internet App

FIWARE Catalogue

FIWARE Instance

assemble…

FIWARE GE implementations proprietary product

16
Building the FIWARE ecosystem: the vision

FIWARE provider A

creation of dedicated
GE instances

Applications
Applications
Use of Global shared
GE instance

FIWARE provider B

FIWARE Catalogue

17
What does FIWARE bring to
Smart Cities?

18
Why standards are relevant for Smart Cities
 The current lack of standards means an impediment for
the development of Smart Cities
• The target market for solution and services is not large
enough to attract investment
• Solutions and services become tailor made, therefore
expensive.
• Cities get locked-in to solution/application providers
 Pillars of the FIWARE programme support the creation of
a sustainable ecosystem:
• The FIWARE platform brings the necessary standards
• The FIWARE Lab becomes the meeting point where cities
meet entrepreneurs and innovation takes place
• The FIWARE Acceleration programme helps to attract a first
wave of developers (startups/SMEs)
• The FIWARE mundus programme helps to expand globally
• The FIWARE Ops suite of tools ease the task to deploy
FIWARE instances

19
Open and Agile Smart Cities initiative

 Launch of the Open and Agile Smart at CeBIT:

• Denmark: Copenhagen, Aarhus and Aalborg
• Finland: Helsinki, Espoo, Vantaa, Oulu, Tampere, Turku
• Spain: Valencia, Santander, Málaga, Sevilla
• Portugal: Porto, Lisbon, Fundão, Palmela, Penela and
Águeda
• Belgium: Brussels, Ghent and Antwerp
• Italy: Milan, Palermo and Lecce
• Brazil: Olinda (Recife), Anapólis (Goiás), Porto Alegre
(Rio Grande do Sul), Vitória (Espírito Santo), Colinas de
Tocantins (Tocantins) and Taquaritinga (São Paulo)

 Other countries and cities welcome to join !

 Some of them already in the pipeline

(*) Ordered by country and date of incorporation

20
From silos to platforms
From clusters to ecosystems :
Cross-fertilization of data
 within a given environment (building, plant, city)…
 between vertical application domains (energy, transportation, logistics, etc…)

22
 Modularity; integration with legacies
Domain-specific platforms = FIWARE + specific enablers

SMART SMART City SMART Factory

Agrifood Apps Apps Apps

Domain
specific
enablers

FIWARE GEs

23
Example: FIWARE for Smart Energy
 FIWARE Sustainable Smart City Malmö, Sweden
 FIWARE Smart Region Horsens, Denmark & Madrid
 FIWARE X-border Virtual Utility, Aachen
 FIWARE for the Energy Marketplace in Terni, Italy
 FIWARE for Power management, Ireland

GE and DSE integration

24
FIWARE as a multi-sided IoT platform
IoT Applications providers

end users

developers

IoT data
providers
Data Networks & IoT Infrastructures
aggregators operators

data providers data providers data providers individual

data providers
25
Raising the level of data abstraction in IoT
infrastructures
 Beyond device and protocol abstraction!
 Capturing the invariants in target environment instances

 Abstracting all relevant physical entities in the environment

› rooms, places ( akin to context entities in context middleware)
› non-connected appliances and legacy systems
› passive items

 Providing higher layers of abstraction

› virtual entities based on properties and categories (intrinsic)
› entity & device instance groups (extrinsic and ad hoc)

pivotal layer
27
Consolidating multiple sources of data
 Primary data may come from many types of sources:
• peer and lower-level infrastructures
• networked sensor systems
• external information systems
• crowdsourcing (individual smartphone sensors)

 They are attached by FIWARE to relevant entities

Notify me in case of
What’s the current congestion of street “X”
traffic density
in street “X”?
FIWARE high-level API
{« street X » :
« hasState » : [
« TrafficDensity »
……
]
}

inductive-loop surveillance
vehicle detector driver smartphones cameras
28
Acting upon entities through devices
 IoT devices are acted upon through changes in the state of entites
Backend IoT Applications

GET <Oauth token> PUT <Oauth token>

/cityEntities/street1/presenceState /cityEntities/street1/lamps/”OnState”

FIWARE API FIWARE API

Backend platform

Edge IoT Applications

FIWARE API

Edge platforms

29
Integration with sensor networks
 The FIWARE backend IoT Device Management GE enables creation and
configuration of NGSI IoT Agents that connect to sensor networks

 Each NGSI IoT Agent can behave as Context Consumers or Context

Providers, or both

OMA NGSI API (northbound interface)

FIWARE Context Broker

create/monitor IoT Agent

Manager

IoT IoT IoT FIWARE Backend IoT

Agent-1 Agent-2 Agent-n Device Management

(southbound interfaces)

OneM2M MQTT IETF CoAP

30
 FIWARE IoT & overall Data Management

Applications

CEP BigData Analysis

Other
ContextBroker
data IoT-enabled
sources Context
Simple processing Management

FIWARE API NGSI IoT

Agents
IoT Backend Device
Management
Backend Backend

Native IoT Native IoT

Agent Agent Gateway/Device
Platform GEs
Gateway/Device
Gateway/Device Gateway/Device
Platform GEs
Platform GEs Platform GEs

31
 Example: Smart City platform
RULES OPERATIONA
DEFINITION L KPI GOVERNANCE OPEN DATA PORTALS
TOOL DASHBOARD

Data
Real-time BI BigData Open Data
Quering/Action,
processing ETL Processing publishing
Publish/Subscr

CEP Short-term

Accounting & Payment & Billing

historic Big Data
data
City Service
Services orchestrator

IDM & Auth

Context Context Broker
Inventory
Adapters CKAN
GIS
Real Time
Media
IoT Backend Device
manag
Stream ement
Processing &
 Smart city platform as IoT Edge
abstra
ction
a Data/Knowledge Hub
measures /
commands

 Non-intrusive, open to
Specific Enablers

Generic Enablers
third parties Media
Actuators Sensors Open Data
streams

32
Relationship
with other
Applications
IoT protocols Other context NGSI
information sources
Context Broker Context Entities Mgmt
NGSI
Context Information
Management Layer
NGSI NGSI

Context-IoT
IoT Brokering
Context IoT translation translation
Management Layer conector
IoT Agent IoT Agent

IoT connectivity and

management layer

Entity IoT resources Mgmt

Attribute

IoT device

IoT resource
FIWARE IoT architecture

34
Open Data publication

NGSI
Context
Broker

Context BigData
Source Analysis
s

36
Bringing IoT to the next level
with Linked Data
 IoT systems no longer isolated islands
 They become part of the larger linked data archipelago
Evolution of FIWARE APIs towards
Linked Data & Web of Things
 abiding by linked data principles
• use URIs as names
• use HTTP URIs,
so that people can look up those names
• when someone looks up a URI,
provide useful information
• include links to other URIs,
so that they can discover more things

 Naming and identifying devices

• URI of network interface (HTTP or CoAP) or proxy

 Naming and identifying physical things

• physical things are interfaced or represented through dereferenceable URIs of their own

• device resources may allow to monitor or control things, either directly or indirectly
 Federating IoT infrastructures with Linked Data

IoT data
sources

IoT data IoT data

Networks & IoT Infrastructures
providers sources
Data
aggregators operators

data providers data providers IoT data

IoT data sources
sources
Linked data from the Web of Things

 Narrow waist =REST identifiers shared by different

infrastructures and abstraction layers
› entities are resources, states are subresources, instant values are
representations
› devices are resources, reading from sensors and actuator controls are representations
HTTP or CoAP URIs for all resources and subresources
JSON-LD for semantic mapping of identifiers monitoring
applications
› resource descriptions are hyperlinks  »follow your nose » real-time
control
crowdsourced
data gathering
› no declarative descriptions à la WSDL! applicationsfast -
data
analytics
enablers
enabler
s M2M
backend

REST = HTTP/CoAP URIs + CRUD + hyperlinks

gateways/
reverse proxies
IP devices
Non-IP devices
persons
things
space entities
Join us!
http://fiware.org

http://lab.fiware.org

Follow @Fiware on Twitter !

41