TY - GEN
T1 - An efficient randomized algorithm for real-time process scheduling in PicOS operating system
AU - Helmy, Tarek
AU - Fatai, Anifowose
AU - Sallam, El Sayed
PY - 2010
Y1 - 2010
N2 - PicOS is an event-driven operating environment designed for use with embedded networked sensors. More specifically, it is designed to support the concurrency in intensive operations required by networked sensors with minimal hardware requirements. Existing process scheduling algorithms of Pi-cOS; a commercial tiny, low-footprint, real-time operating system; have their associated drawbacks. An efficient, alternative algorithm, based on a randomized selection policy, has been proposed, demonstrated, confirmed for efficiency and fairness, on the average, and has been recommended for implementation in PicOS. Simulations were carried out and performance measures such as Average Waiting Time (AWT) and Average Turn-around Time (ATT) were used to assess the efficiency of the proposed randomized version over the existing ones. The results prove that Randomized algorithm is the best and most attractive for implementation in PicOS, since it is most fair and has the least AWT and ATT on average over the other non-preemptive scheduling algorithms implemented in this paper.
AB - PicOS is an event-driven operating environment designed for use with embedded networked sensors. More specifically, it is designed to support the concurrency in intensive operations required by networked sensors with minimal hardware requirements. Existing process scheduling algorithms of Pi-cOS; a commercial tiny, low-footprint, real-time operating system; have their associated drawbacks. An efficient, alternative algorithm, based on a randomized selection policy, has been proposed, demonstrated, confirmed for efficiency and fairness, on the average, and has been recommended for implementation in PicOS. Simulations were carried out and performance measures such as Average Waiting Time (AWT) and Average Turn-around Time (ATT) were used to assess the efficiency of the proposed randomized version over the existing ones. The results prove that Randomized algorithm is the best and most attractive for implementation in PicOS, since it is most fair and has the least AWT and ATT on average over the other non-preemptive scheduling algorithms implemented in this paper.
UR - http://www.scopus.com/inward/record.url?scp=84878538896&partnerID=8YFLogxK
U2 - 10.1007/978-90-481-3660-5_20
DO - 10.1007/978-90-481-3660-5_20
M3 - Conference contribution
AN - SCOPUS:84878538896
SN - 9789048136599
T3 - Advanced Techniques in Computing Sciences and Software Engineering
SP - 117
EP - 122
BT - Advanced Techniques in Computing Sciences and Software Engineering
PB - Springer Publishing Company
ER -