An Enhanced Distributed Algorithm for Area Skyline Computation Based on Apache Spark.

Skyline computations are a way of finding the best data points based on multiple criteria for location-based decision-making. However, as the input data grows larger, these computations become slower and more challenging. To address this issue, we propose an efficient algorithm that uses Apache Spark, a platform for distributed processing, to perform area skyline computations faster and more salable. Our algorithm consists of three main phases: calculating distances between data points, generating distance tuples, and computing the skyline. In the second phase, we apply a technique called local partial skyline extraction, which reduces the amount of data that needs to be sent from each executor (a parallel processing unit) to the driver (a central processing unit). The driver then computes the final skyline from the received data and creates filters to eliminate irrelevant points. Our experiments show that our algorithm can significantly reduce the data size and the computation time of the area skyline.