Here the user can set the CEP engine benchmarking configuration and submit the form once it's done.
MQQT broker
ip address of the MQQT broker to be used during benchmark.EndEventName
topic to be subscribed by the benchmarking instances that signals the end of the benchmark.Instances
benchmarking instance configuration, add new instances and configure them individually, once a new instances is added, you can choose which CEP engine to use.Start Benchmark
submit the form and start the benchmark.
Events
here are defined the input events for the benchmark, each event has a name (the MQQT topic the events are going to be published to) and a set of properties.Properties
each event property is a key-value pair, consisting of name and data type.
Statements
here are defined the output statements, each statement has a name and an EPL query definition.
The Siddhi engine configuration is similar to the Esper configuration, with a few exceptions.
Definition
everything is defined here: the input streams, the queries and the outputs. This is done via the Siddhi query language.Events
andQueries
the names for events and queries used in theDefinition
field need to be filled individually here, if this is not done, the benchmarking instance will not function properly.
Here the user can visualize ongoing and past benchmarks, as well as each instance and its output events.
Every event sent to the MQQT broker is expected to be in a JSON object format.
The event object's fields must match the event properties previously defined in the benchmarking configuration.
E.g. the user defines an event WeatherEvent
with the properties temperature
and humidity
, of types float
and int
respectively. The events sent to the MQQT broker topic WeatherEvent
must have the following JSON format:
{
"temperature": 24.3,
"humidity": 30
}
Note: the only allowed types allowed are string
, number
and boolean
, an event property cannot be of type object
, array
or null
.