This paper presents a Matlab toolkit for the numerical analysis of the two-level processor sharing queue (2LPS). The job sizes are expressed in terms of acyclic phase type distributions which can approximate any distribution arbitrary well while arrivals occur according to a homogeneous Poisson process. The toolkit provides a simple yet efficient way to find the optimal parametrization of the 2LPS queueing disciplines given the job size distributions and the intensity of the workload. In practice, the tool can be used to configure the 2LPS scheduler for TCP flows. The time complexity of the solution depends on the cube of the number of phases of the distribution describing the flow sizes.

A Matlab Toolkit for the Analysis of Two-Level Processor Sharing Queues

Marin A.
Methodology
;
Rossi S.
Methodology
;
Zen C.
Software
2020-01-01

Abstract

This paper presents a Matlab toolkit for the numerical analysis of the two-level processor sharing queue (2LPS). The job sizes are expressed in terms of acyclic phase type distributions which can approximate any distribution arbitrary well while arrivals occur according to a homogeneous Poisson process. The toolkit provides a simple yet efficient way to find the optimal parametrization of the 2LPS queueing disciplines given the job size distributions and the intensity of the workload. In practice, the tool can be used to configure the 2LPS scheduler for TCP flows. The time complexity of the solution depends on the cube of the number of phases of the distribution describing the flow sizes.
2020
Proc. of the 17th International Conference on Quantitative Evaluation Systems, QEST 2020, Vienna, 31 August 2020 - 3 September 2020
File in questo prodotto:
File Dimensione Formato  
main.pdf

non disponibili

Tipologia: Documento in Post-print
Licenza: Accesso chiuso-personale
Dimensione 179.51 kB
Formato Adobe PDF
179.51 kB Adobe PDF   Visualizza/Apri

I documenti in ARCA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10278/3743023
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? 0
social impact