Performance evaluation of diverse techniques for performance, energy, and temperature efficient task allocation

The task-to-core scheduling problem using Dynamic Voltage and Frequency Scaling (DVFS) for achieving three objectives of performance, energy, and temperature (PET), poses algorithmic challenges as it involves conflicting goals and trade-offs. Some myriad static algorithms have been proposed for solving this problem which can be roughly categorized into three groups: approaches for generating optimal solutions (for smaller sizes problems), complex optimization techniques, and fast heuristics. These algorithms generate multi-dimensional results which can be hardly intelligible. The assessment of these results requires new comparison methods and concise evaluation measures. This paper proposes a set of benchmarks and evaluation procedures for carrying out methodical comparisons of various algorithms for solving the PET-aware task-to-core scheduling problem. The proposed performance measures assist in judiciously comparing these different algorithms and analyzing their results on a unified basis. The goal is also to seek answers as to how good the Pareto-optimal algorithms are compared to fast heuristics tackling the same problem with the same assumptions. At the same time, we are interested in knowing how good both the groups of algorithms are compared to the absolute optimal (at least for small sets of problems). In addition, the paper provides methods for evaluating trade-offs and determining which application and target parameters affect the results (performance, energy consumed and temperature achieved) of these algorithms. Extensive experimentations facilitate a comprehensive comparison of different kinds of algorithms amongst themselves as well as with optimal solutions obtained through Integer Linear Programming as a reference.