High-level language (HLL) virtual machines (VMs) (HLL VMs) are becoming more widespread as we share remote computer processing in unique new ways. Instead of focusing on technical VM architectural elements, this paper highlights the methods and results for a systematic study of literature describing HLL VM research.

Out of 128 research papers, the authors undertook a novel systematic mapping of various technical aspects of the most popular VMs. The goal was to define the methods used for classifying and categorizing the most investigated functionalities/features/characteristics of HLL VMs. The results of the study show that Java virtual machines (JVMs) are the most widely used within the academic community. The authors also found that most VM product focus has been on optimizing their execution environments, improving their memory management capabilities, and tailoring them to resource-constrained settings.

The authors describe the research mapping protocol they devised for conducting the studies (the process could also be applied to other formal studies). Their protocol requires identifying domain-specific categories, and then analyzing the frequency distribution of those areas across the research papers. For HLL VMs, the authors identified optimization, garbage collection, debugging, memory leak tolerance, new language construct, profiling, aspect-oriented programming, embedded systems, security, real-time, distributed computing, fault tolerance, resource sharing, and testing. They then identified a year-wise distribution of selected publications on the most investigated categories. The authors further classified the publications by their distribution across publication type and online electronic access. Finally, they produced a visual map that shows the distribution of HLL VM research by category, implementations employed, and year of publication.