ScyllaDB Job Interview

Preconditions

1. Make sure that your Docker Agent/Daemon is always running
2. Download ScyllaDB docker image docker pull scylladb/scylla
3. Start scylla server instance docker run --name some-scylla --hostname some-scylla -d scylladb/scylla

Installation and Running

1. Download or clone repository
2. Navigate to the directory where you downloaded the repository (main catalog)
3. Install required packages by typing following command pip install -r requirements.txt
4. Run the program can be run by typing python scylla_stress --durations X, where X is stands for durations of each Cassandra stress command in [0-9]+[smh] format, e.g. 1s 2s 10s 5s 1m
5. Alternatively the program can be run by typing python scylla_stress --number-of-runs-and-duration X Y where X is the number of Cassandra stress commands and Y is their duration in [0-9]+[smh] format, e.g. 5 10s

NOTE: On Linux make sure that you are running script with sudo and also that you are running proper python, e.g. python3

Extra Features

1. There are some extra arguments which can be passed to scylla_stress:
   • export-to-json - will save stress results into /results folder which is in the main project folder
   • cassandra-logs - will show full output from Cassandra stress runs in console and also log them into log file
   • container-name - will allow you to type non-default container name (other than 'some-scylla')
2. Run logs are saved after each run in /logs folder in the main project folder
3. Sample commands with all available arguments:
   python scylla_stress --number-of-runs-and-duration 3 5s --export-to-json --cassandra-logs --container-name some-scylla
python scylla_stress --durations 1s 2s 10s 1m --export-to-json --cassandra-logs --container-name some-scylla