Table 5 Average makespan ratio and percentage makespan reduction measured by running K-means and Random Forest on datasets of both balanced and imbalanced shape
From: Block size estimation for data partitioning in HPC applications using machine learning techniques
Algorithm | Metric | Best time | Average time | Worst time |
|---|---|---|---|---|
K-means | Makespan ratio | \(0.99 \pm 0.03\) | \(1.17 \pm 0.05\) | \(1.53 \pm 0.07\) |
Makespan reduction (%) | \(-1.03\% \pm 0.02\) | \(14.27\% \pm 0.04\) | \(34.44\% \pm 0.06\) | |
Random Forest | Makespan ratio | \(0.99 \pm 0.01\) | \(1.03 \pm 0.05\) | \(1.10 \pm 0.05\) |
Makespan reduction (%) | \(-0.56\% \pm 0.04\) | \(3.74\% \pm 0.06\) | \(9.44\% \pm 0.09\) |