The software engineering community has produced numerous tools, techniques, and methodologies for practitioners to analyze and optimize memory usage during software execution. However, little is known about the actual needs of programmers when analyzing memory behavior and how they use tools to address those needs. We conducted an exploratory study (i) to understand what a programmer needs to know when analyzing memory behavior and (ii) how a programmer finds that information with current tools. From our observations, we provide a catalog of 34 questions programmers ask themselves when analyzing memory behavior. We also report a detailed analysis of how some tools are used to answer these questions and the difficulties participants face during the process. Finally, we present four recommendations to guide researchers and developers in designing, evaluating, and improving memory behavior analysis tools.

Detecting and repairing memory issues is still a challenging task. One reason is that understanding a program's memory usage involves a diverse and related set of dynamic and static aspects. Over the years, multiple tools have been proposed to assist practitioners in these activities. However, detailed information about how a tool helps users when analyzing memory usage is missing.This article introduces Vismep, an interactive visualization prototype to help programmers analyze Python applications' memory usage, and presents an exploratory study to understand the behavior and perception of users when using Vismep. As a result, we reported five information needs when participants analyze memory consumption and how they use Vismep to satisfy these needs. Besides, participants positively perceived Vismep due to their valuable views and high overall usability.