C7:Complex Event Processing

1
C7 Complex Event Processing

• Making Sense of Sensor Network Events in Real
  Time

John Doherty
Senior Presales Consultant
2
Agenda
CEP Making Sense of Sensor Network Events in
Real Time

• The World of Sensors
• A Business Case for Sensor Events in Real Time
• Building A Sensor Event Architecture
• Case Studies

3
The world is full of sensors

• Sensor A device that measures or detects a
  real-world condition, such as motion, heat or
  light and converts the condition into an analog
  or digital representation.

4
What are the Top Markets leveraging sensor
technology?

• Energy
• Healthcare
• Logistics
• Nature Climate
• Mobility Traffic
• Telecommunications

5
The Control Loop
Traffic light iseither red orgreen
Traffic lightbecomes greenon car approach
Traffic light adapts to road blocks and traffic
jams
6
Connect Sensor Technology to Business IT Quite a
gap
Sensor Technology Environment

• Large number of small high tech companies
• Closed circuit/systems no open access
• Usually not SOA-aware
• Little standardization
• Sensors can spit out very large volumes of data

AN OPPORTUNITY FOR CEP AND ESB?
GAP
Business IT Environment
7
Agenda
CEP Making Sense of Sensor Network Events in
Real Time

• The World of Sensors
• A Business Case for Sensor Events in Real Time
• Building A Sensor Event Architecture
• Scenarios for Traffic Management

8
The dependency mismatch external vs. internal
integration
104 107
ESB
CEP
9
Reality Check Sensors, Events, Services and Data
are very much related
Data

• Service

• Event

Sensor
10
Making sense of sensor events...
The challenge
The opportunity

• Sensors can provide
• real-time data
• real-life data
• increased speed and agility
• Sensors can be
• very small
• very cheap
• wireless

• Sensor events
• need to be analyzed
• Deltas
• Alerts
• Loss of signal
• are not equal to business events
• Sensors are
• usually not intelligent
• not error-free

You need an infrastructure to transform sensor
events to business events
11
Agenda
CEP Making Sense of Sensor Network Events in
Real Time

• The World of Sensors
• A Business Case for Sensor Events in Real Time
• Building A Sensor Event Architecture
• Scenarios for Traffic Management

12
Building a Sensor Event Architecture
Event EdgeDomain
SensorDomain
BusinessDomain
Sensor
EdgeEventProcessor
BusinessEventProcessor
ESB
Actuator
13
The Progress Portfolio for Sensor-Driven SOA
ESB
TCP/IP
MESSAGING
14
When does Apama make sense in sensor networks?

• Multiple (sensor) data sources that need
  correlation
• Large amounts of time-dependent sensor data
• Analyse Filter events
• Include temporal constraints
• Prediction/anticipation based on sensor feeds
• Intelligent decisions (based on rules not part of
  the sensor network)
• Record replay of sensor events

Traffic Jam Scenario
THEN Activate Traffic Jam ACTUATORS AND
Alert TMC Traffic Centre
time
real-time data streams
15
Design Choices for Architects (1)

• Base concepts
• How quickly do you need to act?
• Do you need to derive values from Event Streams?
• What data do you need to store?
• Do you need to integrate applications across
  Business Silos?
• Platform choice
• Apama
• Sonic
• OpenEdge
• All three?

16
Design Choices for Architects (2)

• IT Architecture
• How will you connect to Apama?
• Will you use an Edge Domain (or not?)
• What about Fault Tolerance?
• Development
• Collaboration (Business with IT)?
• Do you need to keep the state?
• Speed (Time to Market vs Latency)?
• What will the interface look like?

17
Agenda
CEP Making Sense of Sensor Network Events in
Real Time

• The World of Sensors
• A Business Case for Sensor Events in Real Time
• Building A Sensor Event Architecture
• Case Studies

18
Case Studies

• Dynamic Parking
• Optimization of parking space usage
• Health Monitoring
• Walkers swallowing pill to monitor vitals
• City Ring Parking
• Parking registration and billing based on car
  location
• Train Delay Management
• Providing real time service to customers and
  employees

19
Dynamic Parking

• Customer Large telecom operator
• Problem Ineffective use of parking space
  reservation
• Sensors used License Plate Recognition, road
  sensors, mobile phones
• Technology used Apama, Sonic, Business Partner
  Application and existing Reservation system.

20
Health Monitoring

• Customer Nijmegen 4 Day Walk
• Problem Walkers Dying due to dehydration.
• Sensors used RFID, Thermometer, mobile phones
• Technology used Apama, GPS, SMS

21
City Ring Parking

• Customer Local government
• Problem Charging for road-side parking
• Sensors used License Plate Recognition
• Technology used Apama, Sonic, Business Partner
  Application

22
Train Delay Escalation

• Customer Railway operator
• Problem providing train delay information to
  customers employees
• Sensors Actuators used Train, Mobile phone,
  Railpocket
• Technology used Apama, Sonic and Google Maps

23
In Summary Making Sense

• Sensor Networks are growing rapidly
• A SOA Architecture approach is very useful
• Not having a SOA will increase complexity
  significantly
• Real Time will provide significant value but
• No investment without a proper business case -
  Will it Save you Money or Make you Money!
• Avoid re-architecting. Its resource intensive
• Architects have to understand the Real Time
  Paradigm
• Usage of real time paradigm has major impact
• Progress has the technology to make sense of
  sensor network events!

24
?
What Do You Think
25
Thank You !!
jdoherty_at_progress.com
26
(No Transcript)