VyPR Web-based Visualisation Tool

A Javascript-based visualisation tool for the VyPR performance analysis framework.

Runs alongside VyPR's analysis server, using:

• Python with Flask for a server.
• Javascript with Vue.js for front-end.
• Dagre (based on d3) for drawing directed graphs.

Setup

You will need three terminals open. One for the VyPR verdict server, one for the visualisation tool server, and one for the test project whose execution will be visualised.

Terminal 1 - Visualisation Tool

Clone the visualisation tool into a new directory

git clone [email protected]:pyvypr/VyPR-visualisation.git

Navigate into the new directory and set up a virtual environment

virtualenv --python=python2.7 venv

activate it

source venv/bin/activate

and install the requirements pip install -r requirements.txt.

Terminal 2 - Verdict Server

From the root directory of the visualisation tool, clone the VyPR verdict server

git clone [email protected]:pyvypr/VyPRServer.git

Navigate into that directory and set up the databases required by the VyPR verdict server for the visualisation tool to work:

source setup_dbs

Then, set up a virtual environment

virtualenv --python=python2.7 venv

activate it

source venv/bin/activate

and install the requirements pip install -r requirements.txt.

Finally, you need to clone VyPR into the verdict server's root directory since the server relies on some code from it

git clone [email protected]:pyvypr/VyPR.git

You don't need to install any requirements for this.

To launch the verdict server, run

python run_service.py --port 9002 --db verdicts.db --events-db events.db --path <visualisation tool root>/test-project/

Terminal 3 - Generating the visualisation

Navigate to the directory test-project in the root directory of the visualisation tool.

Set up a virtual environment

virtualenv --python=python2.7 venv

activate it

source venv/bin/activate

and install the requirements pip install -r requirements.txt.

Now, clone the distribution of VyPR designed for short-running programs

git clone [email protected]:pyvypr/VyPRLocal-py2.git VyPR

With VyPR cloned, instrument the sample project with respect to the queries provided in VyPR_queries.py with

python VyPR/instrument.py

Then run the instrumented code with

python main.py

This will generate verdict data with respect to the queries in VyPR_queries.py.

The visualisation data is now generated, so you need to launch the visualisation tool to display it. In the terminal in which you set up the visualisation tool, run

python src/launch.py --port 8080 --instrumentation-stream http://localhost:9002/event_stream/instrumentation/ \
--monitoring-stream http://localhost:9002/eventtream/monitoring/