Scheduling Analysis under Fault Bursts

Abstract

Real time systems must respect their temporal constraints both in nominal and degraded conditions. Environment disturbances cause faults which are revealed by errors during task execution. Therefore, schedulers must be fault tolerant to guarantee no missed deadline. Phenomena like electromagnetic fields disturb real-time systems on a extended period of time. It is difficult to forecast faults and their consequences to build efficient faulttolerant systems. The classical fault models deal with pseudo-periodic faults. They are not made for phenomena extended in time. This paper intends to describe electromagnetic disturbances in a new fault model, named fault burst model. In adequation with the fault burst model, we provide error recovery strategies. Finally, we study the effects of strategies on the schedulability analysis to guarantee fault tolerance when fault bursts occur.

Real-time scheduling algorithm for safety-critical systems on faulty multicore environments

“…For example, electromagnetic interference (EMI) caused by lighting strikes may cause incorrect computation and bit flips in architectural registers. Other examples are automotive and avionics systems exposed to higher EMI levels when passing airport radar or communication facilities (Many and Doose 2011).…”

Real-time scheduling under fault bursts with multiple recovery strategy

“…More recently, there has been a growing interest in dealing with scenarios where faults follow a random pattern over a bounded time window and impose a potentially continuous disturbance to the computational activities [8]- [11]. In that new framework called the fault burst model, the fault distribution within the interval under consideration is not known.…”

“…For example, short voltage fluctuations due to power supply jitters or electromagnetic interference (EMI) caused by short-lived atmospheric effects (e.g., lightning strikes) may cause incorrect computations in the core logic and bit flips in architectural registers. Other examples may be found in automotive and avionics areas, where the real-time system controlling a vehicle or unmanned system may be exposed to higher EMI levels when passing near airport radar and communication facilities [8], [10].…”

Energy-aware Fault-tolerant Scheduling for Hard Real-time Systems

“…As mentioned before, soft errors occur more frequently than hard errors in modern computing systems. While soft errors can occur in both single-core and multi-core platforms, a majority of current researches are focused on single-core platforms [52,39,8,89,131] and only a few on multi-core platforms.…”

“…The advantage of this approach is that the reliability can be quantified and the impacts of DVFS to reliability can also be taken into consideration. However, to precisely identify the parameters for the reliability model can be challenging, especially when faults usually occur in a burst manner [89].…”