MQTT Benchmark

A benchmarking tool for MQTT v3.1 servers. The benchmark covers a wide variety of common usage scenarios.

Running the Benchmark

Just run:

./bin/benchmark-all

or one of the server specific benchmark scripts like:

./bin/benchmark-apollo

The benchmark report will be stored in the reports/$(hostname) directory.

Running the Benchmark on an EC2 Amazon Linux 64 bit AMI

If you want to run the benchmark on EC2, we recommend using at least the c1.xlarge instance type. Once you have the instance started just execute the following commands on the instance:

sudo yum install -y screen
curl https://nodeload.github.com/chirino/mqtt-benchmark/tarball/master | tar -zxv 
mv chirino-mqtt-benchmark-* mqtt-benchmark
screen ./mqtt-benchmark/bin/benchmark-all

The results will be stored in the ~/reports directory.