Abstract

The proliferation of resource-constrained environments—such as Internet of Things (IoT) devices, embedded systems, and edge computing platforms—has necessitated the evolution of application profiling techniques tailored for low-power and limited-capacity hardware. This paper presents a comprehensive review and analysis of contemporary advancements in profiling strategies aimed at optimizing performance and energy efficiency in such settings. It begins by establishing the foundational concepts of static and dynamic profiling, examining the key performance indicators that guide optimization, and addressing the critical trade-offs between profiling overhead and measurement accuracy. The study highlights innovative tools and methodologies, including lightweight profilers, context-aware instrumentation, and adaptive techniques that dynamically respond to runtime constraints. Furthermore, it explores real-world use cases in IoT and embedded systems, implementation strategies for integrating profiling into modern development workflows, and the persistent challenges related to tool compatibility and deployment. The paper concludes by identifying emerging research directions, such as AI-assisted profiling and cross-platform standardization, emphasizing their potential to advance profiling efficacy and support robust system design. Overall, the findings underscore the vital role of profiling in enabling high-performance, energy-conscious applications in environments where optimization is paramount.