Melbourne liveability

Group project for Cluster and Cloud Computing, 2022

As a team of 5, we implemented a web-app deployed on the cloud to explore liveability in Melbourne, presenting data on maps and charts. We explored social engagement and opportunity liveability indicators using a combination of government datasets and tweets that we harvested. We deployed 4 VMs on the Melbourne Research Cloud using Ansible, with a 3-node CouchDB cluster.

• source
• site demo video
• Ansible deployment video

Architecture

Technology

• Ansible: used for deployment of VMs on the Melbourne Research Cloud, creation of volumes and security groups, installation of dependencies
• Docker: used for CouchDB instances and twitter harvesters
• Svelte: front-end framework making use of MapBox and D3.js to present interactive charts and maps
• Python: used for tweet harvesting (stream and search Twitter APIs) and sentiment analysis using VaderSentiment
• Flask: used to implement ReST API
• Gunicorn: application server for Flask API
• Nginx: web server
• CouchDB: NoSQL database used in a 3-node cluster