Optimizing System Quality of Service through Rejuvenation for Long-Running Applications with Real-Time Constraints
This work addresses reliability and performance issues for safety-critical real-time systems, but it is incremental as it builds on existing rejuvenation methods with a new scheduling algorithm.
The paper tackles the problem of optimizing software rejuvenation timing for long-running real-time applications to maximize reliability, longevity, and availability while minimizing delays, achieving reductions in application delays of 9.01% and 14.24% compared to existing scheduling algorithms.
Reliability, longevity, availability, and deadline guarantees are the four most important metrics to measure the QoS of long-running safety-critical real-time applications. Software aging is one of the major factors that impact the safety of long-running real-time applications as the degraded performance and increased failure rate caused by software aging can lead to deadline missing and catastrophic consequences. Software rejuvenation is one of the most commonly used approaches to handle issues caused by software aging. In this paper, we study the optimal time when software rejuvenation shall take place so that the system's reliability, longevity, and availability are maximized, and application delays caused by software rejuvenation is minimized. In particular, we formally analyze the relationships between software rejuvenation frequency and system reliability, longevity, and availability. Based on the theoretic analysis, we develop approaches to maximizing system reliability, longevity, and availability, and use simulation to evaluate the developed approaches. In addition, we design the MIN-DELAY semi-priority-driven scheduling algorithm to minimize application delays caused by rejuvenation processes. The simulation experiments show that the developed semi-priority-driven scheduling algorithm reduces application delays by 9.01% and 14.24% over the earliest deadline first (EDF) and least release time (LRT) scheduling algorithms, respectively.