Anti-aging scheduling in single-server queues: A systematic and comparative study

• Published in: Journal of communications and networks Vol. 23; no. 2; pp. 91 - 105
• Main Authors: Liu, Zhongdong, Huang, Liang, Li, Bin, Ji, Bo
• Format: Journal Article
• Language: English
• Published: 한국통신학회 01-04-2021
• Subjects: 전자/정보통신공학
• ISSN: 1229-2370; 1976-5541
• DOI: 10.23919/JCN.2021.000005

The age of information (AoI) is a new performance met ric recently proposed for measuring the freshness of information ininformation-update systems. In this work, we conduct a systematicand comparative study to investigate the impact of scheduling poli cies on the AoI performance in single-server queues and provideuseful guidelines for the design of AoI-efficient scheduling poli cies. Specifically, we first perform extensive simulations to demon strate that the update-size information can be leveraged for achiev ing a substantially improved AoI compared to non-size-based (orarrival-time-based) policies. Then, by utilizing both the update-sizeand arrival-time information, we propose three AoI-based policies.
Observing improved AoI performance of policies that allow ser vice preemption and that prioritize informative updates, we fur ther propose preemptive, informative, AoI-based scheduling poli cies. Our simulation results show that such policies empiricallyachieve the best AoI performance among all the considered poli cies. However, compared to the best delay-efficient policies (suchas shortest remaining processing time (SRPT)), the AoI improve ment is rather marginal in the settings with exogenous arrivals. In terestingly, we also prove sample-path equivalence between somesize-based policies and AoI-based policies. This provides an intu itive explanation for why some size-based policies (such as SRPT)achieve a very good AoI performance. KCI Citation Count: 0