The evaluation of scientific simulation applications is crucial for their efficiency and effectiveness. Managing input and output (I/O) operations in seismic simulations represents a significant challenge. The Fletcher Reverse Time Migration (RTM) algorithm is a prime example, where synchronous I/O operations pause computation to store data on disk, which is necessary for saving simulation checkpoints. This approach leads to idle CPU and GPU periods, significantly affecting the simulation time, especially in long-term or high-resolution simulations. This paper examines the performance of the Fletcher RTM algorithm, with a particular focus on the impact of I/O operations. It explores the potential for overlapping computation with I/O to optimize execution, allowing continued data processing while asynchronously saving checkpoints. This strategy can reduce idle periods and enhance efficiency, particularly in environments where efficient management of compute-intensive tasks and data I/O operations is essential, such as modern data centers, cloud computing, and high-performance computing (HPC) systems.