Abstract
Efficient data-structure selection plays a pivotal role in optimizing computational resources, particularly in constrained IoT-edge environments. This study presents a comparative performance analysis of five search-tree variants: binary search tree (BST), red-black tree (RBT), segment tree (ST), ternary tree (TT), and n-ary tree (NT), under synthetic workloads varying key distribution, input size (10,000–50,000), and operation mix. We measured average search, insertion, and deletion metrics. Results show that ST delivers superior performance for search-dominant and range queries; RBT maintains the most consistent update performance under dynamic conditions; and NT achieves the lowest deletion overhead, making it ideal for hierarchical data purging. These quantitative insights provide clear guidance for selecting the optimal search-tree structure when balancing operation cost against resource constraints. Importantly, while rooted in algorithmic performance analysis, these findings have practical relevance for sensor networks and smart applications, where time-efficient data management is essential in resource-constrained environments.