Analyzing the Overhead of Self-Optimization through Task Migration within a Decentralized Task Control Mechanism for Dependable System-on-Chip Architectures

This article presents our concept of an artificial hormone system for realizing a completely decentralized self-organizing and real-time capable task control mechanism using self-X properties. Besides the fundamentals of the prior hormone concept and the implementation model, we present latest results of our research: specification and analysis of an accurate timing model according the self-optimization procedure, using global task migration across a redundant multi-core SoC architecture. Furthermore we validate and compare the overhead in task migration towards the performance gain in execution, leading to first assessment in efficiency and profitability of the migration based optimization schemes.