Reinforcement learning in queues

Ayesta, Urtzi

doi:10.1007/s11134-022-09844-w

Cited by 2 publications

(1 citation statement)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Markov decision processes, MDP, have been used to model queuing systems and achieve optimal control under static and dynamic environments [18], [19]. In an MDP the current feedback, delayed or not, is known to the agent [20].…”

Section: Related Workmentioning

confidence: 99%

Load Balancing in Compute Clusters With Delayed Feedback

Tahir

Alt

Rizk

et al. 2023

IEEE Trans. Comput.

View full text Add to dashboard Cite

Load balancing arises as a fundamental problem, underlying the dimensioning and operation of many computing and communication systems, such as job routing in data center clusters, multipath communication, Big Data and queueing systems. In essence, the decision-making agent maps each arriving job to one of the possibly heterogeneous servers while aiming at an optimization goal such as load balancing, low average delay or low loss rate. One main difficulty in finding optimal load balancing policies here is that the agent only partially observes the impact of its decisions, e.g., through the delayed acknowledgements of the served jobs. In this paper, we provide a partially observable (PO) model that captures the load balancing decisions in parallel buffered systems under limited information of delayed acknowledgements. We present a simulation model for this PO system to find a load balancing policy in real-time using a scalable Monte Carlo tree search algorithm. We numerically show that the resulting policy outperforms other limited information load balancing strategies such as variants of Join-the-Most-Observations and has comparable performance to full information strategies like: Join-the-Shortest-Queue, Join-the-Shortest-Queue(d) and Shortest-Expected-Delay. Finally, we show that our approach can optimise the real-time parallel processing by using network data provided by Kaggle.

show abstract

Section: Related Workmentioning

confidence: 99%