Prime Number Cluster

Overview

I wanted to push the limits of prime number computation — not just with better algorithms, but by building a small home cluster to split the work across multiple machines.

Goals

• Implement and benchmark multiple prime-finding algorithms (trial division, Sieve of Eratosthenes, segmented sieve)
• Set up a small local network cluster to distribute computation
• Measure and compare real-world performance gains from parallelization

Tools & Skills Used

• Python — algorithm implementation, multiprocessing, socket communication
• Linux — server configuration, SSH, network setup
• Networking — LAN configuration, process coordination across nodes

What I Learned

The biggest bottleneck was network latency and coordination overhead — for smaller ranges, a single optimized local process beat the cluster. But for very large ranges, distributing segments across machines made a real difference. This taught me the real tradeoffs in distributed systems design.