2015
DOI: 10.1016/j.peva.2015.06.006
|View full text |Cite
|
Sign up to set email alerts
|

QD-AMVA: Evaluating systems with queue-dependent service requirements

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
11
0

Year Published

2015
2015
2023
2023

Publication Types

Select...
4
3
2

Relationship

4
5

Authors

Journals

citations
Cited by 16 publications
(11 citation statements)
references
References 33 publications
0
11
0
Order By: Relevance
“…Single-server fixed-rate station are instead captured by the assignment α k (n k ) = 1. Several other load-dependent scaling factors α k (n k ) have been used in the literature [28,29].…”
Section: Lld Scaling Factorsmentioning
confidence: 99%
See 1 more Smart Citation
“…Single-server fixed-rate station are instead captured by the assignment α k (n k ) = 1. Several other load-dependent scaling factors α k (n k ) have been used in the literature [28,29].…”
Section: Lld Scaling Factorsmentioning
confidence: 99%
“…To further motivate the need for novel analysis methods for LLD models, we illustrate with a toy example some numerical difficulties that affect two state-of-the-art methods for load-dependent analysis: MVA-LD [16] and an AMVA algorithm for load-dependent models with default settings [29].…”
Section: Motivating Examplementioning
confidence: 99%
“…Multi-server queues are dealt with using the multi-server method in (Rolia and Sevcik 1995) where heterogeneous FCFS scheduling follows the standard AMVA-FCFS approximation (Bolch et al 2006, Section 10.2). Multi-server PS is dealt with using the QD-AMVA approximation (Casale et al. 2015).…”
Section: Mva Solvermentioning
confidence: 99%
“…To produce these metrics, we consider using both the Bard-Schweitzer (BS) approximation, which we already used in Section 5.3 and the recently proposed AMVA-QD method, which is an approximate algorithm for mean performance metrics in load-dependent models [5]. For the previous case, we scale the demands by the minimum of the number of jobs and number of CPUs to approximate the multi-core behavior of the server.…”
Section: Explaining Observed Performancementioning
confidence: 99%