Barebone MQTT server that can run on any stream servers
- Aedes
To install aedes, simply use npm:
npm install aedes
Check Docker docs here
- Full compatible with MQTT 3.1 and 3.1.1
- Standard TCP Support
- SSL / TLS
- WebSocket Support
- Message Persistence
- Automatic Reconnect
- Offline Buffering
- Backpress-support API
- High Availability
- Clusterable
- Authentication and Authorization
$SYS
support- Pluggable middlewares
- Dynamic Topics Support
- MQTT Bridge Support between aedes
- MQTT 5.0 (not support yet)
- Bridge Protocol (not support yet)
Aedes needs on disk dbs like MongoDB and Redis in order to work with clusters. Based on our tests and users reports the best performances/stability are reached when using aedes-persistence-mongodb paired with mqemitter-redis.
Other info:
- The repo aedes-tests is used to test aedes with clusters and different emitters/persistences. Check its source code to have a starting point on how to work with clusters
- aedes-logging: Logging module for Aedes, based on Pino
- aedes-stats: Stats for Aedes
- aedes-cli: Run Aedes MQTT Broker from the CLI
- aedes-protocol-decoder: Protocol decoder for Aedes MQTT Broker
- aedes-server-factory: Create a server instance such as TCP, HTTP, TLS...
- aedes-persistence: In-memory implementation of an Aedes persistence
- aedes-persistence-mongodb: MongoDB persistence for Aedes
- aedes-persistence-redis: Redis persistence for Aedes
- aedes-persistence-level: LevelDB persistence for Aedes
- aedes-persistence-nedb: NeDB persistence for Aedes
- mqemitter: An opinionated memory Message Queue with an emitter-style API
- mqemitter-redis: Redis-powered mqemitter
- mqemitter-mongodb: Mongodb based mqemitter
- mqemitter-child-process: Share the same mqemitter between a hierarchy of child processes
- mqemitter-cs: Expose a MQEmitter via a simple client/server protocol
- mqemitter-p2p: A P2P implementation of MQEmitter, based on HyperEmitter and a Merkle DAG
- mqemitter-aerospike: Aerospike mqemitter
This library is born after a lot of discussion with all Mosca users and how that was deployed in production. This addresses your concerns about performance and stability.
Example benchmark test with 1000 clients sending 5000 QoS 1 messsages. Used mqtt-benchmark with command:
mqtt-benchmark --broker tcp://localhost:1883 --clients 1000 --qos 1 --count 5000
CPU INFO:
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Byte Order: Little Endian
CPU(s): 8
On-line CPU(s) list: 0-7
Thread(s) per core: 2
Core(s) per socket: 4
Socket(s): 1
NUMA node(s): 1
Vendor ID: GenuineIntel
CPU family: 6
Model: 94
Model name: Intel(R) Core(TM) i7-6700HQ CPU @ 2.60GHz
Stepping: 3
CPU MHz: 800.014
CPU max MHz: 3500,0000
CPU min MHz: 800,0000
BogoMIPS: 5199.98
Virtualization: VT-x
L1d cache: 32K
L1i cache: 32K
L2 cache: 256K
L3 cache: 6144K
========= TOTAL (1000) =========
Total Ratio: 1.000 (5000000/5000000)
Total Runtime (sec): 178.495
Average Runtime (sec): 177.845
Msg time min (ms): 0.077
Msg time max (ms): 199.805
Msg time mean mean (ms): 35.403
Msg time mean std (ms): 0.042
Average Bandwidth (msg/sec): 28.115
Total Bandwidth (msg/sec): 28114.678
========= TOTAL (1000) =========
Total Ratio: 1.000 (5000000/5000000)
Total Runtime (sec): 114.404
Average Runtime (sec): 109.022
Msg time min (ms): 0.065
Msg time max (ms): 393.214
Msg time mean mean (ms): 21.520
Msg time mean std (ms): 0.595
Average Bandwidth (msg/sec): 45.896
Total Bandwidth (msg/sec): 45896.306
========= TOTAL (1000) =========
Total Ratio: 1.000 (5000000/5000000)
Total Runtime (sec): 112.769
Average Runtime (sec): 105.524
Msg time min (ms): 0.062
Msg time max (ms): 329.062
Msg time mean mean (ms): 20.750
Msg time mean std (ms): 0.878
Average Bandwidth (msg/sec): 47.464
Total Bandwidth (msg/sec): 47464.271
========= TOTAL (1000) =========
Total Ratio: 1.000 (5000000/5000000)
Total Runtime (sec): 118.587
Average Runtime (sec): 114.190
Msg time min (ms): 0.080
Msg time max (ms): 324.028
Msg time mean mean (ms): 22.558
Msg time mean std (ms): 0.730
Average Bandwidth (msg/sec): 43.832
Total Bandwidth (msg/sec): 43831.927
Benchmark: Mosca
========= TOTAL (1000) =========
Total Ratio: 1.000 (5000000/5000000)
Total Runtime (sec): 264.934
Average Runtime (sec): 264.190
Msg time min (ms): 0.070
Msg time max (ms): 168.116
Msg time mean mean (ms): 52.629
Msg time mean std (ms): 0.074
Average Bandwidth (msg/sec): 18.926
Total Bandwidth (msg/sec): 18925.942
Here is a list of some interesting projects that are using Aedes as MQTT Broker. Submit a PR or an issue if you would like to add yours
- node-red-contrib-aedes: MQTT broker for Node-Red based on Aedes
- Mqtt2Mqtt: Mqtt Bridge between two brokers with UI
- Kuzzle: High performance and full featured IoT backend using MQTT alongside WebSocket and Http protocols
Want to contribute? Check our list of features/bugs
If there are bugs/leaks in production scenarios, we encourage people to send Pull Request and/or reach out maintainers for some paid support.
Thank you to all our backers! ๐
Become a sponsor to get your logo on our README on Github
Licensed under MIT.