A Semi-Partitioned Model for Mixed Criticality Systems

Hao Xu; Alan Burns

doi:10.1016/j.jss.2019.01.015

A Semi-Partitioned Model for Mixed Criticality Systems

Hao Xu, Alan Burns

Computer Science

Research output: Contribution to journal › Article › peer-review

Abstract

Many Mixed Criticality algorithms have been developed with an assumption that
lower criticality-level tasks may be abandoned in order
to guarantee the schedulability of higher-criticality tasks when the criticality level of the system changes.
But it is valuable to explore means by which all of the tasks remain schedulable through these criticality level changes.
This paper introduces a semi-partitioned model for a multi-core platform that
allows all of the tasks to remain schedulable if
only a bounded number of cores increase their criticality level. In such a model,
some lower-criticality tasks are allowed to migrate instead of being abandoned.
Detailed response time analysis for this model is derived.
This paper also introduces possible approaches for establishing migration routes.
Together with related previous work, an appropriate semi-partitioned model for mixed
criticality systems hosted on multi-core platforms is recommended.

Original language	English
Pages (from-to)	51-63
Number of pages	13
Journal	Journal of Systems and Software
Volume	150
Early online date	9 Jan 2019
DOIs	https://doi.org/10.1016/j.jss.2019.01.015
Publication status	Published - 1 Apr 2019

Keywords

Real-time
Mixed Criticality

Access to Document

10.1016/j.jss.2019.01.015

Journal
© 2019 Published by Elsevier Inc. This is an author-produced version of the published paper. Uploaded in accordance with the publisher’s self-archiving policy.
Accepted author manuscript, 586 KBLicence: CC BY-NC-ND

Cite this

@article{6f5816d07808465e8e73f5438ebb8ac5,

title = "A Semi-Partitioned Model for Mixed Criticality Systems",

abstract = "Many Mixed Criticality algorithms have been developed with an assumption thatlower criticality-level tasks may be abandoned in orderto guarantee the schedulability of higher-criticality tasks when the criticality level of the system changes.But it is valuable to explore means by which all of the tasks remain schedulable through these criticality level changes.This paper introduces a semi-partitioned model for a multi-core platform thatallows all of the tasks to remain schedulable ifonly a bounded number of cores increase their criticality level. In such a model,some lower-criticality tasks are allowed to migrate instead of being abandoned.Detailed response time analysis for this model is derived.This paper also introduces possible approaches for establishing migration routes.Together with related previous work, an appropriate semi-partitioned model for mixedcriticality systems hosted on multi-core platforms is recommended.",

keywords = "Real-time, Mixed Criticality",

author = "Hao Xu and Alan Burns",

year = "2019",

month = apr,

day = "1",

doi = "10.1016/j.jss.2019.01.015",

language = "English",

volume = "150",

pages = "51--63",

journal = "Journal of Systems and Software",

issn = "0164-1212",

publisher = "Elsevier",

}

TY - JOUR

T1 - A Semi-Partitioned Model for Mixed Criticality Systems

AU - Xu, Hao

AU - Burns, Alan

PY - 2019/4/1

Y1 - 2019/4/1

N2 - Many Mixed Criticality algorithms have been developed with an assumption thatlower criticality-level tasks may be abandoned in orderto guarantee the schedulability of higher-criticality tasks when the criticality level of the system changes.But it is valuable to explore means by which all of the tasks remain schedulable through these criticality level changes.This paper introduces a semi-partitioned model for a multi-core platform thatallows all of the tasks to remain schedulable ifonly a bounded number of cores increase their criticality level. In such a model,some lower-criticality tasks are allowed to migrate instead of being abandoned.Detailed response time analysis for this model is derived.This paper also introduces possible approaches for establishing migration routes.Together with related previous work, an appropriate semi-partitioned model for mixedcriticality systems hosted on multi-core platforms is recommended.

AB - Many Mixed Criticality algorithms have been developed with an assumption thatlower criticality-level tasks may be abandoned in orderto guarantee the schedulability of higher-criticality tasks when the criticality level of the system changes.But it is valuable to explore means by which all of the tasks remain schedulable through these criticality level changes.This paper introduces a semi-partitioned model for a multi-core platform thatallows all of the tasks to remain schedulable ifonly a bounded number of cores increase their criticality level. In such a model,some lower-criticality tasks are allowed to migrate instead of being abandoned.Detailed response time analysis for this model is derived.This paper also introduces possible approaches for establishing migration routes.Together with related previous work, an appropriate semi-partitioned model for mixedcriticality systems hosted on multi-core platforms is recommended.

KW - Real-time

KW - Mixed Criticality

U2 - 10.1016/j.jss.2019.01.015

DO - 10.1016/j.jss.2019.01.015

M3 - Article

SN - 0164-1212

VL - 150

SP - 51

EP - 63

JO - Journal of Systems and Software

JF - Journal of Systems and Software

ER -