Time-dependent priority disciplines: A heavy traffic analysis

Carra, Damiano; Michiardi, Pietro
Research Report RR-11-251


heavy traffic



 


 

 

 

regime, and derive closed form expressions to evaluate the relativefinite buffers for both

 

difference - in terms of average waiting times - of requests that belong

 

to different priority classes.

 

Our analysis allows to derive the expected waiting time of requests in single

 

server queueing systems, under the assumption of

 

non-preemptive and preemptive cases. Additionally, we extend our results to

 

a distributed scenario composed of an interconnected set of a large number

 

of unreliable single server queueing systems.

 

We then describe a number of applications and illustrate how to use our

 

theoretic foundation to analyze and tune their performance. The accuracy of

 

our analysis is validated through a set of experiments that we performed in a

 

synthetic setting.

 

 

 

In this work, we revisit the theory underlying a range of time-dependent

 

priority disciplines, and extend it to include the requirements of a class of

 

applications that has not been studied in the past. Specifically, we target applications

 

and services in which a scarce resource, or a fraction thereof, has

 

to be awarded to a large number of concurrent requests. We thus consider an


Type:
Rapport
Date:
2011-03-02
Department:
Data Science
Eurecom Ref:
3350
Copyright:
© EURECOM. Personal use of this material is permitted. The definitive version of this paper was published in Research Report RR-11-251 and is available at :

PERMALINK : https://www.eurecom.fr/publication/3350