The Impact of Varying Memory Region Numbers and Nesting on RTSJ Execution Time

摘要

Developing a real-time system in Java requires awareness of memory behaviour in addition to program logic. In the same context, region-based memory management has certain advantages over garbage collection in terms of predictability; however, the use of regions is, in many ways, arbitrary, since it is left to the developer to decide on the number of required regions prior to program execution. In theory, a memory model with a large number of regions will negatively impact the execution time of software. This paper presents an initial, exploratory empirical study of varying the number of un-nested and nested regions on the execution times of RTSJ. RTS 2.2 Java code was used to allocate varying numbers of integer objects (100 to 2500) into regions and the execution times were then recorded. Results suggested that more regions do actually lead to increases in execution time, and when nested regions were used, execution times increased proportionately more. The empirical nature of the study presented highlights the importance of careful choice of region numbers; we point to future work as a result.