EC Cost Simulator is a Python simulator designed to evaluate the reconstruction cost of erasure coding schemes in distributed storage systems. It was developed as part of research on optimizing repair costs for Reed–Solomon–based storage systems, and was used to evaluate the hybrid Hitchhiker–LRC scheme.
- Simulation of erasure coding schemes:
- Reed–Solomon (RS)
- Hitchhiker (HH)
- Uniform Cauchy LRC (LRC)
- Hybrid Hitchhiker–LRC (HHLRC)
- Exhaustive evaluation of failure combinations
- Computation of normalized reconstruction cost
- Generation of repair cost distribution histograms
The reconstruction cost corresponds to the number of fragments read during repair and is normalized relative to Reed–Solomon:
c_normalized = (fragments_read / s) × 100
where s is the number of data fragments
./run.shThe simulator evaluates repair costs for different coding schemes and produces histogram visualizations of the results.