Abstract
Computational scalability enables process mining to deliver useful results as event data grow in volume, complexity, and rate of change. This review organizes scalability approaches to help researchers and practitioners match methods to their analytical needs. It combines a systematic search map of the literature with a purposive full-text synthesis of selected studies. The map documents publication coverage, while the synthesis develops a framework around reduction, parallelization, relocation, and reuse of computation. The framework connects these mechanisms to process mining tasks, workload structure, and the results they preserve. It makes preparation, execution, and maintenance costs explicit, helping readers assess when a faster operation improves the intended analysis. Within the selected studies, resource reporting emphasizes runtime, with more limited attention to memory, communication, and energy. These findings support comparisons under stated conditions and identify opportunities for broader evaluation. This review provides a structured basis for selecting compatible methods and designing experiments on reuse, diverse workloads, and sustained operation.