Living life on the edge. Martin Gale Master Inventor and CTO for Industrial Sector, IBM UK and Ireland @martinjgale
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Living life on the edge.Martin Gale Master Inventor and CTO for Industrial Sector, IBM UK and Ireland
@martinjgale
In the beginning, we did things the hard way.
Most expensiveLeast expensive
Connected sensors moved the boundaries of enterprise systems and “the edge” became a thing.
Remote sites Headquarters
e.g. MQTT
Edge of the network
There are some important factors we learned from this.
Impact on enterprise systems
design and capacity
The cost of remote communications
Latency between sense and respond.
Good old fashioned common
sense.
Capable devices, capable edge.
HeadquartersRemote sites
e.g. MQTT
Shift towards actionable insight. More efficient use of comms.
Autonomy of remote sites. More possibilities.
What we do at the edge varies in complexity.
Complexity
Filtering/aggregation
Media analytics
Com
pute
pow
er Stream analytics
Data mediation and flow control
Application/device
integration
Very few IoT reference architectures break the edge down.
Control Code Local applications
Analytics Models
Sensor SDKs Back-end SDK
Application Runtime
Wide Area Network
Local Area Network
Assets
Sensors and Actuators
Loca
l Mon
itorin
g an
d
Man
agem
ent
Rem
ote
Mon
itorin
g an
d M
anag
emen
t
Phys
ical
Sec
urity
Devic
e Pl
atfo
rm
Secu
rity
Device
Device installation and maintenance
Device and asset management
Technology
Organisational
Who can access the running box?
How?
Who installs and fixes the device?
How do we apply fixes remotely?
What do we build our applications
out of?
What type of device is really needed?
The edge applies to multiple areas of “Industry 4.0”.
Flexible manufacturing Internet of People and Things
Autonomous systems Analytics and cognitive
Internet of services
In manufacturing, halted production costs big money.
Many manufacturers only know when production will halt when it happens!
per minute£10ks 5-10
minutes a day
Incremental use of the edge helps reduce risk of lost production.
Spot the trend and automatically
respond
Understand the trend and its context
A hybrid of analytics on the cloud and edge enable a smarter factory.
sound, vibration, temperature
Synthesis of readings indicates head will break
within 24 hours
Triggers creation of work order for that night’s planned
maintenance window
Re-routes high priority work order to another cell in
the meantime
Thank you!
Related Documents
