Machine to Machine Communication with Microsoft Azure IoT Edge & HiveMQ

Copyright © by HiveMQ. All Rights Reserved.
● Independent contractor
● Strong software
engineering experience
● Certiﬁed Azure
Solutions Architect
● Focused on IoT, from
concept to the
implementation
● Technical blogger
● Speaker
WELCOME
Christoph Schäbel Kresimir Galic
@schaebo @kgalic1
linkedin.com/in/kresimir-galic
-8664a66a/
www.variant-logic.com
● Practical MQTT expert
with multiple years of
experience in the ﬁeld
● HiveMQ Core Developer
● Background in scalable
and reliable distributed
systems and robotics
linkedin.com/in/cschaebel
www.hivemq.com

IoT Edge

WHAT WE WILL TALK ABOUT
• Shifting to the Edge
→ What is edge?
• Azure IoT Edge Concept
→ Cloud managed deployments
→ Containerized environment
• Cloud Managed Deployment of HiveMQ Broker for
M2M Communication

What We Call ‘Edge’?
• Edge of network
• Closer to the devices/machines
• Outside of the cloud, but managed by
the cloud

Shifting To The Edge
• Cloud
→ Globally available, unlimited compute resources
• IoT
→ Signals from sensors and machines, managed centrally by
cloud
• Edge
→ Intelligence oﬄoaded from the cloud to IoT devices
• AI
→ Intelligence capabilities, in the cloud and on the edge

Azure IoT Edge
• Cloud Managed
→ Enables rich management of Azure IoT Edge from Azure
provide a complete solution instead of just an SDK
• Cross Platform
→Containerized environment, enables targeting most popular
edge operating systems such as Windows and Linux
• Portable
→ Enables Dev/Test of edge workloads in the cloud with later
deployment to the edge as part of a continuous integration / continuous
deployment pipeline
• Extensible
→ Deploying custom modules, third party solutions and/or AI

MQTT on
the edge

MQTT
● Lightweight protocol on top of TCP/IP
● Publish / Subscribe pattern
● De-coupling of sender and receiver
● Instant message delivery
● Open standard

MQTT History

MQTT 5 Features
• Compatibility and Portability improvements
• Session & Message Expiry
• Shared Subscriptions
→ Load balancing for clients
→ Multiple clients share the same subscription
• User Properties
→ User deﬁned metadata
→ Reduce bandwidth and costs
• Improved Client Feedback
→ More descriptive reason codes and messages
→ Negative acknowledgements
• Request / Response pattern

HiveMQ ecosystem
• Enterprise MQTT platform
• Different Editions
→ Commercial with enterprise features and up to 24/7 support
→ Open Source “Community” edition
• Open Source client library
→ built for high-performance
• Enterprise integrations
→ Security
→ Kafka
→ ….

Why HiveMQ on the Edge?
• MQTT as open standard
→ Interoperability
→ Vendor neutrality
→ HiveMQ supports 100% MQTT v5
→ Azure has partial MQTT v3.1.1 support
• Independent edge
→ Reliable intra-edge communication, without dependency
on internet connectivity
→ Vendor neutrality
• Control data ﬂow and cost
→ Control which part of the IoT data is sent to the cloud or
analyzed on the edge

Why HiveMQ and IoT Edge?
• Bidirectional messaging
→ Direct Communication of components / machines at the
edge with each other
• Simple deployment and maintenance
→ Managed through Azure Cloud UI
→ Simple concepts
• Intelligent Edges
→ (Pre-)analyze data at the edge
→ Utilize Azure AI technologies
• High throughput on the Edge
→ Maintain high-throughput of messages on the edge,
not every message has to go to the cloud

Use Cases

Intra Machine Communication
• Easy to manage and operate
→ IoT Devices can be provisioned and managed in the cloud
• Eﬃcient intra machine communication
→ Low latency
→ Low overhead
• Early data analysis
→ Machine learning and AI for each machine
• Cost effective
→ Only selected and pre-analyzed data is streamed to the
cloud

Inter Machine Communication
• Reliable
→ Factory stays independent of internet connectivity
→ Each building / location can have own communications hub
• Fast
→ Lower latencies in local network
• Interoperable
→ between machines and vendors
due to standard protocol MQTT v5
• Secure
→ Separate security layers for on-premise and cloud
• Intelligent
→ Computing and analysis right at the edge

Azure IoT Edge
Concepts

Concept - Azure IoT Edge Runtime
• Installs and updates workloads on the
device.
• Maintains Azure IoT Edge security
standards on the device.
• Ensures that IoT Edge modules are
always running.
• Reports module health to the cloud for
remote monitoring.
• Facilitates communication between downstream leaf devices and the IoT Edge device.
• Facilitates communication between modules on the IoT Edge device.
• Facilitates communication between the IoT Edge device and the cloud

Concept - Module
• Module image is a package containing the software that deﬁnes a module.
• Module instance is the speciﬁc unit of computation running the module image on an IoT Edge device.
→ The module instance is started by the IoT Edge runtime
• Module identity is a piece of information (including security credentials) stored in IoT Hub, that is associated to each
module instance.
• Module twin is a JSON document stored in IoT Hub, that contains state information for a module instance.
• SDKs to develop custom modules in multiple languages (C#, C, Python, Java, Node.JS)

Concept - Routing
● Query Language
FROM
/messages/modules/TelemetrySubscri
berModule/outputs/* INTO $upstream

Concept - Device Management

IoT in the Cloud vs Edge

HiveMQ on Azure IoT Edge

Deploying HiveMQ Broker from the Cloud
● Deployment manifest
● Deploying at scale

● Deployment manifest
● Deploying at scale
Deploying Custom Module from the Cloud

Deploying Custom Module from the Cloud
● Subscribing to the dedicated topic
● Topic name can be a part of Module Configuration (Module Twin)
● Using HiveMQ Client Library

Routing Messages and Sending Results to the Cloud
● Sending the message to the module output
● Azure IoT Edge routes decide where the message goes next (another
module, cloud - IoT Hub)
● "TelemetrySubscriberModuleToIoTHub": "FROM
/messages/modules/TelemetrySubscriberModule/outputs/* INTO $upstream"

Machine to Machine Communication
• Clients/Machines can leverage HiveMQ client
library or any other MQTT library
• Using HiveMQ broker to leverage MQTT features
• One example is ‘LastWill’ message to determine whether
machine got disconnected for triggering the alarms
through the cloud
• Full code available on GitHub:
https://github.com/kgalic/IoTEdgeM2MWithHiveMQ
• Broker for machines with additional logic embedded
for preprocessing before alarming another machine

What’s next?

What’s next?
• AI / ML at the edge and in the cloud
→ (Pre-)analyze data at the edge
→ Utilize Azure AI technologies
• Even deeper integration
→ HiveMQ extensions that directly integrate with IoT Edge
• High availability at the edge
→ Deploy HiveMQ clusters
→ Multiple IoT Edge Devices at the same location
• HiveMQ Cloud <-> Azure Cloud
→ Managed MQTT brokers through HiveMQ Cloud
→ Big Data analytics with Azure

Resources
MQTT Essentials Series
Evaluate HiveMQ Broker
Try HiveMQ Cloud
Get Started with MQTT

ANY
QUESTIONS?
Reach out to community.hivemq.com

Machine to Machine Communication with Microsoft Azure IoT Edge & HiveMQ

More Related Content

Similar to Machine to Machine Communication with Microsoft Azure IoT Edge & HiveMQ (20)

More from HiveMQ (20)

Recently uploaded (20)

Machine to Machine Communication with Microsoft Azure IoT Edge & HiveMQ