See more

ni.com
Big Analog Data™ Solutions
The Often Overlooked Big Data
ni.com
Our Mission
We equip engineers and scientists with tools that
accelerate productivity, innovation, and discovery.
ni.com
3
Graphical System Design
A platform-based approach for measurement and control
ni.com
4
Science, Discovery, and Innovation . . .
Experimental
Science
Theoretical
Science
Computational
Science
Data-Intensive
Science
Jim Gray, The Fourth Paradigm
Internet of
Things
Industrial
Internet
ni.com
Intelligent
Systems
Big Analog Data™
Solutions
Programmable
World
Cyberphysical
Systems
Smarter
Planet
5
M2M
Industrie
4.0
Data-Intensive….  Phases
Cyber Physical,
Prescriptive Analytics
Optimized,
Predictive Analytics
Automated,
Descriptive Analytics
Manual
Big Analog Data
Big Analog Data
Data
Paper Based
ni.com
Big Analog Data
6
Big Data Characterized
Variety
Another “V” NI is seeing . . .
Mix of structure and format
Volume
Visibility
Large amounts
Access from disparate geographic locations
Velocity
High speed, high sample rates
Value
Importance of analysis, which was
previously limited by technology
ni.com
7
Big Data Characterized
Sources of Big Data:
Industry/IT sources



Enterprise apps: ERM, CRM, HR
IT data: events, logs, inventories
Process and control
New/emerging data sources


Social data, behaviors, sentiments
Tweets, posts, comments
Engineering/scientific data sources


Physical world: analog phenomenon
DAQ, A/D
Big Analog Data™ Sources
ni.com
8
Big Data Characterized
Big Analog DataTM Sources
is Big Data derived from the physical analog world
is sourced from nature, people, and electrical and mechanical devices and machines
is digitized via A/D conversion
Location
Sound
Time
Temperature
Pressure
Voltage
Current
Radio Signals
Magnetism
Moisture
Particulates
Vibration
Acceleration
Velocity
Video
ni.com
Light
Motion
9
Wind
Big Analog Data™ Applications Characterized
Smart Power Grid
Measurements:
5TB / Month
Turbine Test
Measurements:
10TB / Day
Jet Engine
Measurements:
20TB / Hour
Physics Experiment
Measurements:
40TB / Second
ni.com
10
T.M.S. Bradicich, PhD
Big Analog Data™ Sources
Big Analog Data™Sources are Everywhere
… the oldest, fastest, and biggest
of all the other Big Data… combined
ni.com
11
Big Analog Data™ Sources
… the oldest, fastest, and biggest
of all the other Big Data… combined
ni.com
12
Big Analog Data™ End-To-End Solution Architecture
Decision Making and Control
Data Acquisition and
Analysis Systems
(Test, Monitoring, Logging, Control)
Measurement HW and Firmware
IT Infrastructure
The Edge
Sensors/Actuators
Analysis and Visualization Software
Data Aggregation and Management
ni.com
13
(Big Data Analytics, Mining)
Edge IT
(Local, Remote, Cloud)
Corporate/
Federated IT
Big Analog Data™ End-To-End Solution Architecture
Data Acquisition and
Analysis Systems
IT Infrastructure
(Test, Monitoring, Logging, Control)
NI Hardware and FPGA Firmware
The Edge
Sensors/Actuators
(Big Data Analytics, Mining)
Edge IT
(Local, Remote, Cloud)
Corporate/
Federated IT
NI Software
Acquire
DataFlow:
Real-Time
InMotion
EarlyLife
AtRest
Analyze
Engineering,Scientific,andBusinessAnalytics
Present
Visualization
System Management/RASM1
Deployment,Health,Monitoring,andServiceability
1
ni.com
RASM:Reliabilty,Availabilty,Serviceabilty,Manageabilty
14
Archive
Example Big Analog Data™ End-to-End Solution
Fleet-wide Online Asset Condition Monitoring
Monitoring Systems
Vibration
Sensor
Turbines
(Critical)
Oil
Sensor
IT Infrastructure
The Edge
Sensors
NI CompactRIO
Vibration
Sensor
Wired
or Wireless
Leak Detection
Sensor
Balance of Plant
Assets
Pressure
Sensor
Wireless
Access
Point
NI CompactRIO
Temperature
Sensor
10,000
machines
ni.com
30,000 sensors
Business Network
2,000+ devices
NI InsightCM
Enterprise
Server
~60 plant servers
15
1 Federated
M&D center
Example Big Analog Data™ Solution Architecture
Jet Engine Test Cell
Corporate IT
Edge Servers
IT Infrastructure
The Edge
Data Acquisition and
Analysis Systems
(NI PXI)
Sensors
ni.com
16
Example Big Analog Data™ Solution Architecture
Sensors/Actuators
Data Acquisition and
Analysis Systems
The Edge
Scientific Research (Seismic)
IT Infrastructure
Emitting surface
Recording surface
- PXI
- FPGA
- LabVIEW
ni.com
- DIAdem
- DAQ
- Timing/Synch
17
- GPUs
- Multicore x86 Servers
- Vector Math
Example Big Analog Data™ Solution Architecture
Product Manufacturing Test
IT Infrastructures
The Edge
Data Acquisition
and Analysis
Systems
Sensors
Visibility
Access from disparate
geographic locations
This annihilates former ‘time and place dependencies’
Images Courtesy of Adventure Technologies, Agarwal, Kulak, Weir
ni.com
18
Example Big Analog Data™ Solution Architecture
ni.com
Sensors
Data Acquisition and
Analysis Systems
64 sensor signals per tram
3 NI CompactRIO
systems per tram
19
The
Edge
Rail In-Vehicle Monitoring
IT Infrastructure
1 Ethernet switch per tram,
1 industrial PC per tram
Example Big Analog Data™ Solution Architecture
Smart Grid - Synchrophasor/Phasor Measurement Units (PMUs)
IT Infrastructure
Data Acquisition and
Analysis Systems
Sensors
The
Edge
Edge
Servers
ni.com
20
Example Big Analog Data™ Solution Architectures
Embedded Monitoring & Control Applications
ni.com
21
Example Big Analog Data™ Solution Architectures
Test Applications
ni.com
22
Big Analog Data™ End-to-End Solution Architecture
Data Acquisition and
Analysis Systems
IT Infrastructure
(Test, Monitoring, Logging, Control)
NI Hardware and FPGA Firmware
The Edge
Sensors/Actuators
(Big Data Analytics, Mining)
Edge IT
(Local, Remote, Cloud)
Corporate/
Federated IT
NI Software
Acquire
DataFlow:
Real-Time
InMotion
EarlyLife
1) Engineering Insight
2) Scientific Insight
3) Business Insight
It’s all about the data – and the
insight derived from the data:
ni.com
AtRest
23
Archive
Big Analog Data™ End-to-End Solution Architecture
Data Acquisition and
Analysis Systems
IT Infrastructure
(Test, Monitoring, Logging, Control)
NI Hardware and FPGA Firmware
The Edge
Sensors/Actuators
(Big Data Analytics, Mining)
Edge IT
(Local, Remote, Cloud)
Corporate/
Federated IT
NI Software
1. RIO
FPGA
2. System
CPUs
3.Edge
IT CPUs
4. Back-End
IT CPUs
Acquire
4 processingDatadomains
Flow:
for data analytics:
Real-Time
InMotion
EarlyLife
AtRest
Archive
Data Insight “Spectrum of Value”
Data analytics executed throughout the data flow
ni.com
24
Big Analog Data™ End-to-End Solution Architecture
Data Acquisition and
Analysis Systems
IT Infrastructure
(Test, Monitoring, Logging, Control)
NI Hardware and FPGA Firmware
The Edge
Sensors/Actuators
(Big Data Analytics, Mining)
Edge IT
(Local, Remote, Cloud)
Corporate/
Federated IT
NI Software
1. RIO
FPGA
2. System
CPUs
3.Edge
IT CPUs
4. Back-End
IT CPUs
Acquire
4 processingDatadomains
Flow:
for data analytics:
Trade Off:
ni.com
Real-Time
InMotion
EarlyLife
AtRest
Faster Insight – vs.-- Deeper Analytics
25
Archive
The Internet of Things and
Big Analog Data™ Solutions
Commercial
Smart factory
Three supply side benefits of “perpetual connectivity”
to product and consumer
Physical assets
Transportation
Electronic
devices
Vending machines
Home appliances
Monitor
Maintain
Motivate
Continuous knowledge
of the condition and
usage of product
in the market
Convenient upgrades,
patches, fixes
and predictive
failure analysis
Compel consumer to
purchase up-sells,
accessories, etc.
Automobiles
Consumer
ni.com
26
Big Analog Data™ Solutions
Mapping to The Internet of Things:
IT Infrastructure
Data Acquisition and
Analysis Systems
Sensors/Actuators
Commercial
(Analytics, Data Buses, Archieve)
Edge IT
(Local, Remote, Cloud)
Smart factory
Corporate/
Federated IT
Physical assets
Transportation
Electronic
devices
Vending machines
Home appliances
Automobiles
Consumer
Acquire
DataFlow:
ni.com
Real-Time
InMotion
27
EarlyLife
AtRest
Archive
Manufacturing and Big Analog Data™ Solutions
The Fourth Industrial Revolution
First production line,
Cincinnati slaughterhouses (1870)
First mechanical loom (1784)
First Industrial Revolution
Water and steam
powered mechanical
production
End of 18th century
Second
Industrial Revolution
Electric power and
mass production
Start of 20th century
Third
Industrial Revolution
Electronics and
IT automation
Data Intensive
Computers,
Communications,
Control
“Industrial Internet”
“Smart Factories”
“Cyber-physical
Systems”
Start of 1970s
Source DKFI 2011
ni.com
Fourth
Industrial Revolution
28
Today
time
complexity
First programmable logic
controller, Modicon 084 (1969)
Manufacturing and Big Analog Data™ Solutions
Complex Data Acquisition, Control, and Global Supply Chain Communications
Smart Factories and IoT drive efficiencies by
“immediate and perpetual visibility“ into manufacturing operations:
This cuts out ‘middle men’ in the system
This annihilates former ‘time and place dependencies’
Production Machines
Smart-Grid
Heavy Equipment
Test Cells
Hardware in
the Loop (HIL)
Big Physics
Metal Fabrication
Test and
Measurements
ni.com
29
Gaining Insight from Data
“Time-Based” and “Place-Based” Considerations
3 Key
“Place-Based”
Considerations:
2.
Quadrant of
highest efficiency
Place
Independence
Seek to move to the upper right
quadrant for maximum efficiency
1. The Location of Acquisition
2. The Location of Analysis
3. The Location of Awareness
Time
Independence
5 Key
“Time-Based”
Considerations:
3.
Acquire
DataFlow:
Real-Time
InMotion
EarlyLife
AtRest
Data Insight “Spectrum of Value”
ni.com
1.
30
Archive
Gaining Insight from Data
“Time-Based” and “Place-Based” Considerations
Place
Independence
Quadrant of
highest efficiency
Visibility
Access from disparate
geographic locations
NI sees “Visibility” as a key
characteristic of big data, which enables
both Time and Place Independence
Time
Independence
e.g. IT Infrastructures, Cloud,
Smart Devices, IoT
1. The Location of Acquisition
2. The Location of Analysis
3. The Location of Awareness
5 Key
“Time-Based”
Considerations:
Acquire
DataFlow:
Real-Time
InMotion
EarlyLife
AtRest
Data Insight “Spectrum of Value”
ni.com
31
Archive
Considerations: Enabling Technology Vectors
Data
Management
Systems
Management
Data
Analytics
capture, filing,
transmission,
retrieval, aging,
data formats,
visualization
deployment,
configuration,
health monitoring,
security, RASM,
remote access
algorithms,
prognostics,
alarming,
distributed
intelligence
Considerations: Customer Value
Data
Management
Systems
Management
Data
Analytics
capture, filing,
transmission,
retrieval, aging,
data formats,
visualization
deployment,
configuration,
health monitoring,
security, RASM,
remote access
algorithms,
prognostics,
alarming,
distributed
intelligence
Robust, end-to-end test and validation (V&V)
Reduces customer’s Integration Risk
Decreases customer’s Time to Value
ni.com/big-analog-data
ni.com