A practical method for quickly improving performance and reducing answer time through the selection of hot loops based on the input data

A quick resolution of the performance problem depends on having precise methodology/tool for exploring the relationship between performance and computation analysis. Also, it is necessary to take into account parameters that have an impact on the methodology process. Input data is an important and a main parameter which can involve the performance gain and the answer time. In this paper, we propose a strategy which allows us to quickly improve performance by (1) focusing on the hot part of code and (2) taking into account the input data. Our methodology provides a precise static/dynamic analysis for the selected part of code. It guides user to apply the best transformation for the hot part of code. Whatever the number and the size of input data, the size of the application, our methodology provides in a short time the best transformation for the selected part of code.