ADAPTIVE HYBRID OPTIMIZATION OF SORTING ALGORITHMS FOR HETEROGENEOUS COMPUTING ARCHITECTURES

In the current era of exponential data growth, driven by the proliferation of digital systems and large-scale analytics, the problem of efficient data ordering has become one of the central challenges in computer engineering. Sorting, as a fundamental operation, lies at the heart of data processing workflows in databases, machine learning pipelines, and real-time systems. Classical implementations of sorting algorithms—such as QuickSort, MergeSort, and RadixSort—exhibit competitive performance in sequential environments but degrade significantly when scaled to modern heterogeneous computing platforms.