Cloud Scheduling Datasets
Description
1. Augmented Cloud Scheduling Dataset: This dataset was generated by introducing small random perturbations (jitter) to the original base dataset of 100 rows. Each record contains four attributes: Index, input_size, output_size, and priority. Perturbations were clipped within valid bounds (input_size ∈ [1,99]; output_size ∈ [3,100]; priority ∈ [1,5]) to maintain realism. 2. Synthetic Uniform Cloud Scheduling Dataset: A fully synthetic dataset generated by sampling uniformly across valid ranges: input_size ∈ [1,99]; output_size ∈ [3,100]; priority ∈ [1,5]. The number of samples can be scaled (e.g., 500 rows). 3.Bootstrap-Expanded Cloud Scheduling Dataset: This dataset was created by resampling with replacement from the base dataset to add 500 new rows. Unique Index values were assigned, and the new samples were concatenated with the original dataset. 4.Stratified-Augmented Cloud Scheduling Dataset: A stratified augmentation technique was applied to preserve the priority distribution observed in the original dataset. Within each priority group, small controlled perturbations were added to input_size and output_size, clipped within valid ranges.
Files
Institutions
- Chittagong University of Engineering and Technology