Evaluation of Static and Dynamic Task Mapping Algorithms in NoC-Based MPSoCs

Ewerson Carvalho,  C├ęsar Marcon,  Ney Calazans,  Fernando Moraes
PUCRS


Abstract

Mapping task is an important issue in MPSoCs design. The state of the art of task mapping algorithms is performed at design time, which is known as static mapping. Nonetheless, applications running in MPSoCs may execute a varying number of simultaneous tasks, and even applications can be defined after system design, enforcing a scenario that requires dynamic task mapping. Static mappings have as main advantage the global view of the system, while dynamic mappings normally provide a local view, which considers only neighbors of the mapping task. This work aims to evaluate the pros and cons of static and dynamic mapping solution. Due to the global system view, it is expected that static mapping algorithms achieve superior performance (latency, congestion, energy consumption), which is confirmed by the results presented herein. As dynamic scenarios are a trend in present MPSoC designs, the cost of dynamic mapping algorithms must be known, and directions to improve the quality of such algorithms should be provided without increasing execution time. This quantitative comparison between static and dynamic mapping algorithms is the main contribution of this work.