Analytics & ML · Study deck

IoT Edge Models: Processing Levels

Keep the first design small.

Data Dora is your guide for this deck.

edgepatternsmodel

Data Dora, the module guide, in a scene from this chapter.

After studying this chapter

Learning objectives

You will be able to:

Explain the Seven-Level IoT Reference Model: Describe how data flows from physical devices through edge processing to cloud analytics
Distinguish OT vs IT Layers: Classify which levels handle operational technology (data in motion) versus information technology (data at rest)
Apply Level 3 Processing Functions: Implement evaluation, formatting, decoding, distillation, and assessment at the edge
Design for Data Accumulation: Determine appropriate storage strategies at Level 4 based on data characteristics

Cloud-centric traffic depends on a central service, while an edge-centric route lets selected immediate decisions stop at a governed local node.

Major section

Start With the Story

Sending every frame away would be slow and costly.

An edge system does some work near the camera before it sends a result.
Bandwidth means how much data a link can carry in a given time.
This short trial often reveals where the first useful decision should live.
The best location is not always the nearest one.

Major section

IoT Reference Model Overview

The IoT Reference Model provides a structured framework for understanding where data processing occurs in IoT systems.

Devices create observations, connectivity moves them, and Level 3 evaluates, formats, distils, or assesses them before Level 4 makes data durable.
The choice is therefore about dependency, not whether the cloud exists.

Seven-layer IoT Reference Model diagram showing Level 1 Physical Devices at bottom through Level 7 Business Collaboration at top, with Level 3 Edge Computing highlighted as the transformation layer performing evaluation, formatting, distillation, and assessment — IoT Reference Model Seven Layers with Edge Computing Functions

Major section

IoT Reference Model Overview (continued)

With that path established, the timeline in Figure: Diagram comparing response times across IoT Reference Model shows what changes when a response stops locally or continues upward.

The local route stops at Level 3 for a response measured in tens of milliseconds, whereas the cloud route crosses accumulation and higher-level processing and can take seconds or longer.
The comparison connects level placement to the chapter's running constraints—latency, bandwidth, and the amount of context required by the decision.
Every device crosses the shared bus to reach the central service, so a lost upstream path also removes its immediate decision point.

Diagram comparing response times across IoT Reference Model levels

Major section

Level 2 Connectivity

Level 2 contains communications and connectivity -- providing reliable, timely information transmission.

This includes transmissions from devices (Level 1) to the network, across the network, and into the edge computing layer (Level 3).
A gateway changes protocol and trust responsibilities, while a direct IP path moves more of those responsibilities onto the device.

Networking and connectivity in IoT systems

Major section

Edge Bandwidth Savings Calculator

The gateway tests, filters, and groups the data before it sends a smaller stream to the cloud.

That point must also keep the meaning of each event.
A lower byte count is useful only if the cloud can still make its decision and the gateway logs what it dropped or grouped.

Edge IoT architecture and processing overview

Deck summary

Key takeaways

Sending every frame away would be slow and costly.

The IoT Reference Model provides a structured framework for understanding where data processing occurs in IoT systems.
With that path established, the timeline in Figure: Diagram comparing response times across IoT Reference Model shows what changes when a response stops locally or continues upward.
Level 2 contains communications and connectivity -- providing reliable, timely information transmission.
The gateway tests, filters, and groups the data before it sends a smaller stream to the cloud.

Retrieval practice

Recall check 1 of 2

Data Dora says: answer from memory, then check your reasoning.

Q1A gateway converts device protocols to MQTT, checks a temperature safety threshold, and averages readings over 5-minute windows. At which IoT Reference Model level do these occur?

ALevel 1, Physical Devices, because sensors originate readings before gateway conversion, checks, or averaging

BLevel 4, Data Accumulation, because the stored averages are the first meaningful stage of processing.

CLevel 5, Data Abstraction, because MQTT may feed applications after the gateway performs edge processing

DLevel 3, Edge (Fog) Computing, where data is formatted, evaluated, and distilled at the network edge

Show answer

Answer: D Protocol conversion, threshold checks, and windowed aggregation are Level 3 edge (fog) computing functions.

Retrieval practice

Recall check 2 of 2

Data Dora says: answer from memory, then check your reasoning.

Q2A team calls a gateway transform and a stored history the same layer. Which map separates the work?

AOperational technology begins only after accumulation has stored every reading.

BData accumulation means the physical sensor that first samples the signal.

COperational technology transports and transforms; accumulation stores; higher layers use.

DApplications alone own raw device transport, transforms, and storage.

Show answer

Answer: C The model separates moving data, durable accumulation, and later information use.

Print reference

Answers

Answer key.

D · Protocol conversion, threshold checks, and windowed aggregation are Level 3 edge (fog) computing functions.
C · The model separates moving data, durable accumulation, and later information use.