This work addresses the fundamental challenges covered in a number of papers on high performance computing. One of the most important issues is the Memory Wall problem. Standard memory access patterns can cause sparse temporal and spatial locality and thus performance degradation. It is important to get the profile of an application’s memory access to analyze its performance. The main purpose of this study is the dynamic binary analysis tool for data locality analysis. The tslmap profiler was developed for analyzing the performance behavior of applications. It is based on the Valgrind platform intended for creating tools for binary analysis of applications. The tslmap tool allows to get and visualize the profiles helping to measure the metrics of temporal and spatial locality by dynamic analysis of a program’s memory references. Experiments were performed on cluster architectures for plotting the performance of client application as a function of its spatial and temporal scores for certain systems.