Load testing on Stallion
message broker with Race
. By using
this service you can perform a load test on Stallion
cluster in order
to measure system performance over your custom load.
Race configs are as follows, they need to be inside a config.yaml
file.
host: st://@localhost:7025 # stallion host
message: hello # publish message
debug: false # display results in stdout
consumers: 10 # number of consumers
providers: 10 # number of providers
offset: 2 # providers publish interval
topics: # system topics
- snapp
In order to set up race on docker use the following image:
docker pull amirhossein21/stallion-race:v0.1.0
System metrics will be available on localhost:8080/metrics
. Example:
{
"providers": 10,
"consumers": 10,
"failed_publish": 0,
"publish": {
"snapp": 20,
"titan": 15,
"workers": 15
},
"subscribe": {
"snapp": 4,
"titan": 3,
"workers": 3
},
"received_message": {
"snapp": 73,
"titan": 40,
"workers": 44
},
"topics": ["snapp", "titan", "workers"]
}