Analysis of a priority queueing system with the enhanced fairness of servers scheduling

A multi-server queueing system with two types of customers is analyzed. Both types of customers have own reserved servers and also there is a pool of servers that can be used by both types of customers. Type 1 customers have almost preemptive priority with flexible use of a priority depending on the...

Full description

Saved in:
Bibliographic Details
Published in:Journal of ambient intelligence and humanized computing 2024, Vol.15 (1), p.465-477
Main Authors: Lee, Seokjun, Dudin, Alexander, Dudina, Olga, Kim, Chesoong
Format: Article
Language:English
Subjects:
Artificial Intelligence
Buffers
Cognitive radio
Computational Intelligence
Customer services
Customers
Engineering
Markov chains
Original Research
Preempting
Queuing theory
Robotics and Automation
Scheduling
Servers
User Interfaces and Human Computer Interaction
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A multi-server queueing system with two types of customers is analyzed. Both types of customers have own reserved servers and also there is a pool of servers that can be used by both types of customers. Type 1 customers have almost preemptive priority with flexible use of a priority depending on the current number of customers of this type in an infinite buffer. Type 2 customers do not have an input buffer and are lost in cases of absence of available servers upon arrival or expelling (forced termination) from service by Type 1 customers. A multi-dimensional Markov chain, which defines the dynamics of the considered system under the fixed total number of servers, numbers of reserved servers, a pattern of the correlated arriving processes, service rates, and the thresholds defining the mechanism of expelling Type 2 customers from service, is analyzed. A numerical example of solving an optimization problem based on the obtained results is considered. Results can be used for the elaboration of enhanced protocols of servers scheduling in many real-world systems including cognitive radio networks with channels leasing.
ISSN:1868-5137
1868-5145
DOI:10.1007/s12652-022-03903-z