Cloud-Dependent
Every message depends on a remote connection.
Cloud broker
Unreliable WAN
Edge devices & PLCs
Webinar
Introducing EMQX Enterprise 6.3: More Control, Stronger Security, and Long-Term Support → EMQX EDGE
Industrial edge
MQTT data backbone
Unify industrial data locally across factories, devices, systems, and remote sites—and bridge it reliably to cloud, analytics, and AI.
QUICK INSTALL
curl -fsSL emqx.sh/edge | bashCloud & Data Platform
EMQX Cloud AWS IoT Azure IoT Google Cloud
EMQX Edge
Connect Process Stream Bridge
Edge Data Sources
EMQX NeuronGatewaysDevicesSCADA
Powering industrial data for leading manufacturers
< 5 ms
Local latency
100K+
Concurrent edge connections
1M+ msg/s
Multi-core throughput
< 10 MB
Memory footprint under load
Why EMQX Edge
Cloud-only messaging leaves the edge exposed
Edge sites need a local MQTT data backbone and reliable synchronization—even when the cloud connection drops.
Edge-First
Local operations continue while data synchronizes upstream.
Cloud & data platform
EMQX Edge (Local UNS)
Local devices & apps
Capabilities
Connect, process, stream, and bridge edge data
01 / CONNECT
Run full MQTT at the edge
Connect devices, gateways, and applications through a high-performance MQTT broker built for constrained edge environments.
- MQTT 5.0 & MQTT 3.1.1 compliant
- MQTT over QUIC & Sparkplug B aware
- TCP, TLS, and WebSocket connectivity
Edge broker
LightweightMQTT 5.0
Sensor cluster
38 nodes
Factory PLC
line-7
Fleet gateway
edge-12
sensors/#
Local topic tree
Client
$ nanomq_cli pub -t "topic" \
-q 2 -u nano -L 100 \
-m test -h broker.emqx.io \
-p 1883
$ nanomq_cli sub -t t -q 1 \
-h broker.emqx.io -p 188302 / PROCESS
Process streams where they happen
Integrate with eKuiper to filter, aggregate, transform, and route MQTT data before it consumes bandwidth or reaches cloud systems.
- SQL-based stream processing through eKuiper
- Windowed aggregation and payload normalization
- Local actions, alerts, and MQTT republishing
eKuiper integrationLive pipeline
Filter
Noise removed
Aggregate
1s window
Transform
Normalize
SELECT avg(temperature) AS temp_avg
FROM mqtt_stream
WHERE quality = 'good'
GROUP BY tumblingwindow(ss, 1)
Rule EngineProcess locally before bridging
03 / STREAM
Persist edge data with MQTT stream
Turn selected MQTT topics into persistent local streams. Query history by time range and replay data for troubleshooting, analytics, and recovery.
- Non-intrusive capture keeps live forwarding fast
- Ring-buffer batching and Parquet persistence
- Time-range queries through dashboard and REST API
factory/plus/1
sensors/temp
site/wind/+/data
MQTT stream
Ring buffer
Parquet
Query
REST API
Replay
04 / BRIDGE
Bridge edge data to any MQTT cloud
Synchronize curated local topic trees with central MQTT platforms through resilient, multiplexed data bridges.
- Bidirectional bridges to MQTT cloud platforms
- MQTT over QUIC with fast reconnection and TCP fallback
- Selective forwarding, topic mapping, and QoS control
Edge-to-cloud bridgeMultiplexed
EdgeLocal cache
BufferSyncRetry
CloudUnified view
Bi-directional bridge with persistent buffering
Deployment
Put a reliable MQTT backbone at every edge
A compact C runtime for Linux, Windows, and Docker across x86, ARM, and RISC-V targets.
Explore deployment optionsEdge gateway
Run on armhf and arm64 Linux gateways close to devices and production lines.
Industrial PC
Deploy on x86_64 industrial PCs for higher connection and message volumes.
Docker
Use the official image for repeatable deployments across edge fleets.
Benefits
Engineered for demanding edge environments
Ultra-low latency
Keep local message dispatch below 5 ms for responsive shop-floor and automation control loops.
High throughput
Handle more than one million messages per second with an asynchronous actor architecture built for multi-core systems.
Ultra-Lightweight
Start with a boot footprint as low as 200 KB and use less than 10 MB of memory under load.
Network resilient
Queue and retransmit messages automatically through intermittent cellular and satellite WAN.
Secure by design
Protect local data with TLS/SSL, multiple authentication methods, and fine-grained access control lists.
Easy to operate
Manage through a built-in web dashboard, REST APIs, and native Prometheus telemetry metrics.
EMQX Edge decouples local operations from unreliable upstream networks and synchronizes data when connectivity returns.
01
Broker messages locally with low latency.
02
Keep local devices and applications connected.
03
Cache upstream messages at the edge.
04
Reconnect over TCP or QUIC and retransmit automatically.
Use cases
One edge broker. Many distributed architectures.
Factory-floor data hub
Aggregate MQTT data from multiple production lines and EMQX Neuron protocol converters into one local broker.
Site-Level UNS
Create an autonomous local MQTT namespace and synchronize selected topics upstream to enterprise historians.
Connected vehicles
Run local messaging in vehicle edge units and bridge telematics over unstable cellular networks.
Smart utilities
Aggregate meter, solar inverter, EV charging, and substation field data close to distributed assets.
Data centers
Keep local infrastructure telemetry flowing while publishing power, cooling, and equipment data to cloud analytics.
Remote edge sites
Preserve operational telemetry on oil rigs, wind turbines, and mining equipment over high-latency satellite uplinks.
Unified namespace
One data backbone. A distributed UNS across every site.
EMQX Neuron connects industrial assets. EMQX Edge creates a local UNS and keeps it available. EMQX Enterprise or cloud unifies data across sites, applications, analytics, and AI.
Field assets
Devices & PLCs
Sensors · Gateways · Machines
Protocol converter
EMQX Neuron
Modbus, OPC UA, S7 to MQTT
Local UNS broker
EMQX Edge
Connect · Process · Stream · Bridge
Analytics & AI
Cloud & data platform
EMQX Cloud · AWS · Azure
Frequently asked questions
EMQX Edge is a lightweight, high-performance MQTT data backbone for the industrial edge. It connects local MQTT data, integrates with eKuiper for processing, persists selected topics with MQTT stream, and bridges data to enterprise or cloud MQTT platforms.
EMQX Neuron connects industrial equipment through protocols such as Modbus, Siemens S7, and OPC UA, then converts its data to MQTT. EMQX Edge receives that MQTT data and brokers, processes, persists, and bridges it across the site and to the cloud.
EMQX Edge supports MQTT 5.0 and MQTT 3.1.1, including all three MQTT quality of service levels (QoS 0, 1, and 2), topic aliasing, user properties, and shared subscriptions.
Yes. Local MQTT communication continues independently. Bridge caching stores configured upstream messages and retransmits them when connectivity returns, while MQTT stream can preserve selected topic data for local queries.
Both are lightweight MQTT brokers. EMQX Edge adds a multi-threaded actor architecture, built-in MQTT stream persistence and queries, eKuiper integration, MQTT over QUIC bridging, and a built-in dashboard, REST API, and Prometheus metrics for industrial edge deployments.
EMQX Edge runs on Linux, Windows, and macOS across x86, ARM, and RISC-V architectures, with an official Docker image for container deployments.
Build your edge MQTT data backbone today.
Deploy in minutes with the Linux quick install, downloadable packages, or Docker.