Distributed Voting and Polling System

This project presents a distributed voting and polling system designed to operate in a decentralized and fault-tolerant environment. The system allows users to create polls and cast votes through a REST-based web service, while maintaining replicated application state across multiple nodes. 

The core challenge addressed is maintaining replicated poll state in the presence of concurrent updates, node crashes, and network partitions. To address this, the system adopts a Conflict-free Replicated Data Type (CRDT) model, allowing nodes to process updates independently while guaranteeing eventual convergence of replicas. Update dissemination is implemented through a best-effort replication strategy combined with periodic anti-entropy synchronization, ensuring that updates missed during temporary disconnections are eventually propagated. 

To improve dependability, each node persists updates using a write-ahead log (Write-Ahead Log (WAL)) and periodically creates checkpoints, allowing recovery after crashes with no loss of locally persisted updates. A heartbeatbased failure suspicion mechanism is also included to monitor node reachability and identify potentially failed or disconnected peers. 

The system prioritizes availability and partition tolerance over strong consistency, making it suitable for distributed environments where network reliability cannot be guaranteed. The final implementation demonstrates how fundamental distributed systems concepts such as replication, consistency trade-offs, and fault tolerance can be applied in a practical application.