An overview of fault tolerance techniques and energy consumption for real time embedded system

The progress of technology has led to an increase in processor performance, but this has also made them more vulnerable to faults These faults can be transient or permanent. A transient fault happens for a short period time and then disappears without caused physical damage to the processor. On the contrary, a permanent fault disables a processor permanently, it impacts on system reliability and increases the power consumption. Due to the emergence of various fault-tolerance techniques maintaining time constraints has become more difficult. Redundancy is one strategy employed for this (spatial and temporal redundancy). Since embedded systems require the satisfaction of several constraints such as energy which is considered as a limited resource (battery size and recharge interval) that must be conserved as much as possible, and while fault tolerance increases its consumption, dynamic power and energy management must be taken into account to ensure the dependability of these systems. There have been numerous proposed task distribution/scheduling policies: on the one hand techniques for ensure timing and reliability and on the other techniques for enforce reducing power/energy consumption while also satisfying timeliness and reliability simultaneously This paper gives an overview of research projects that use fault-tolerance methods while taking into account timing, reliability, power/energy in critical real-time embedded systems’ design.