StreamingMASSIF

This is the implementation of the StreamingMASSIF platform, a streaming extension of the MASSIF platform.

StreamingMASSIF allows to perform cascading reasoning by combining various components. In its standard configuration it allows to filter meaningful events from a datastream through RDF Stream Processing, abstract the selection through DL reasoning and perform Complex Event Processing ontop of these abstraction.

Check the wikipage for a more in depth explanation on how to use Streaming MASSIF!

How to cite Streaming MASSIF:

@article{bonte2018streaming,
  title={Streaming MASSIF: Cascading Reasoning for Efficient Processing of IoT Data Streams},
  author={Bonte, Pieter and Tommasini, Riccardo and Della Valle, Emanuele and De Turck, Filip and Ongenae, Femke},
  journal={Sensors},
  volume={18},
  number={11},
  pages={3832},
  year={2018},
  publisher={Multidisciplinary Digital Publishing Institute}
}

How to cite MASSIF:

@article{bonte2017massif,
  title={The MASSIF platform: a modular and semantic platform for the development of flexible IoT services},
  author={Bonte, Pieter and Ongenae, Femke and De Backere, Femke and Schaballie, Jeroen and Arndt, D{\"o}rthe and Verstichel, Stijn and Mannens, Erik and Van de Walle, Rik and De Turck, Filip},
  journal={Knowledge and Information Systems},
  volume={51},
  number={1},
  pages={89--126},
  year={2017},
  publisher={Springer}
}

Building and running MASSIF

Requirements:

• Java 9+
• Maven2

Build

To build the MASSIF app, call mvn and get the compiled .jar.

mvn clean compile assembly:single
mv target/massif-jar-with-dependencies.jar .

To build the MASSIF classes (e.g. for usage in higher level apps), call mvn to compile and install the project in the local repository, then add the package to the higher level app dependencies (see pom.xml snippet below):

mvn install -Dmaven.test.skip=true 

<dependency>
    <groupId>be.ugent.idlab</groupId>
    <artifactId>massif</artifactId>
    <version>0.0.1</version>
</dependency>

Run

To run MASSIF, call the compiled .jar from the command line as follows:

java -jar -Dlog4j.configurationFile=webfiles/log4j2.xml massif-jar-with-dependencies.jar

Calling this command will return something like this on the CLI:

21:27:32.313 [main] INFO  idlab.massif.run.Run - MASSIF STARTING
21:27:32.403 [main] INFO  idlab.massif.run.Run - MASSIF Listening on port 9000
21:27:32.403 [main] INFO  idlab.massif.run.Run - Access the MASSIF GUI on  localhost:9000 or register a configuration on localhost:9000/register
21:27:32.414 [main] INFO  idlab.massif.run.Run - MASSIF is ONLINE

Run options:

• -p : TCP port on which massif listens, default: 9000

GUI:

• The MASSIF GUI is available on http://localhost:9000

REST API

The MASSIF allow for direct management through GET/POST calls. Here are the most important path:

• <massif_url>/register: register a configuration
• <massif_url>/stop : stop a certain query
• <massif_url>/configs : get all registered configs

REST API call examples with the cURL tool

List of active configs:

curl -X GET --verbose \
  --url "http://127.0.0.1:9000/configs"

• Note: returns {} if there are no active config.

Stop a config (pre-requisite: config ID, see previous command):

curl -X POST --verbose \
  -H "Content-Type: application/json" \
  -d '19' \
  --url "http://127.0.0.1:9000/stop"

• Note: returns ok

Send a config:

curl -X POST --verbose \
  -H "Content-Type: application/json" \
  -d '{"configuration":{"0":[1],"1":[]},"components":{"0":{"type":"Source","impl":"kafkaSource","kafkaServer":"127.0.0.1:9092","kafkaTopic":"backblaze_smart"},"1":{"type":"Sink","impl":"httpGetSinkCombined","path":"1","config":""}}}' \
  --url "http://127.0.0.1:9000/register"

• Note: returns the config ID

Get a component running metrics (pre-requisite: component ID, see previous commands):

curl -X GET --verbose \
  --url "http://127.0.0.1:9000/monitor/1"