Dear Reader:
Welcome to the Edge Computing issue of the Journal of Innovation! Edge Computing is quickly
becoming an essential component of Industrial IoT (IIoT). As IIoT is being adopted for a wide range
of uses to include mission-critical equipment monitoring and running autonomous machinery in real
time, it is no longer acceptable to send data from devices to remote platforms and await control
commands. As a result, new paradigms in compute deployment are essential. Edge Computing
addresses this need for alternative compute deployment scenarios by moving compute closer to the
data sources – allowing a faster and more reliable sense-analyze-response cycle. Having data locally
processed also means that not all data needs to be sent remotely – saving bandwidth and energy.
Edge Analytics is pushing the concept further by having analytical models and even business
applications deployed in the edge at scale for near real-time predictions.
This edition of the Journal of Innovation brings you a wide range of topics on Edge Computing: use
cases, requirements, connectivity, architecture and operational aspects to support large scale
deployments. In particular:
• An overview of work and activities happening within the Edge Computing Task Group in
formulating a formal definition for Edge and Edge Computing in “Where is the Edge of the Edge
of Industrial IoT?”
• “Device Ecosystem at the Edge - Manufacturing Scenario” delves into different types of edge
devices in the context of Smart Manufacturing Systems which include machines, assets, devices
and people.
• In “Edge Intelligence: The Central Cloud is Dead – Long Live the Edge Cloud!,” the authors analyze
a set of edge computing-related use cases to identify requirements and technology gaps.
• A discussion of architecture for machine learning at the edge and platform tiers for Industrial IoT
– embodying classical lambda architecture and Linux Foundation’s EdgeX Foundry™ in “Industrial
IoT Edge Architecture for Machine and Deep Learning.”
• In “A Knowledge Graph Driven Approach for Edge Analytics,” the authors describe a novel
approach to the deployment of an edge analytics framework that addresses several key
operationalization challenges such as support, maintainability, and management.
• In “A Practical Guide to Using the Industrial Internet Connectivity Framework” the authors discuss
how the IIC-developed framework helps designers understand the many connectivity standards
and choose the right one for their applications.
We hope you enjoy this edition and do let us know if there are any topics you’d like to see covered
in the future. The next edition of Journal of Innovation will focus on Artificial Intelligence. Stay
tuned!
Journal of Innovation Editorial Team Leaders:
Edy Liongosari
Chief Research Scientist
Accenture
Thought Leadership Task Group Co-Chair
Industrial Internet Consortium
Mark Crawford
Standards Strategist
SAP Strategic IP Initiatives
Architecture TG Co-Chair,
Standards TG Co-Chair &
Thought Leadership Task Group Co-Chair
Industrial Internet Consortium
i
September 2017