Integrated Analysis of Cache Related Preemption Delays and Cache Persistence Reload Overheads

Syed Aftab Rashid; Geoffrey Nelissen; Sebastian Altmeyer; Robert I. Davis; Eduardo Tovar

doi:10.1109/RTSS.2017.00025

Integrated Analysis of Cache Related Preemption Delays and Cache Persistence Reload Overheads

Syed Aftab Rashid
, Geoffrey Nelissen
, Sebastian Altmeyer
, Robert I. Davis
, Eduardo Tovar

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

14 Citations (Scopus)

Abstract

Schedulability analysis for tasks running on micro- processors with cache memory is incomplete without a treatment of Cache Related Preemption Delays (CRPD) and Cache Persistence Reload Overheads (CPRO). State-of-the-art analyses compute CRPD and CPRO independently, which might result in counting the same overhead more than once. In this paper, we analyze the pessimism associated with the independent calculation of CRPD and CPRO in comparison to an integrated approach. We answer two main questions: (1) Is it benecial to integrate the calculation of CRPD and CPRO? (2) When and to what extent can we gain in terms of schedulability by integrating the calculation of CRPD and CPRO? To achieve this, we (i) identify situations where considering CRPD and CPRO separately might result in overestimating the total memory overhead suffered by tasks, (ii) derive new analyses that integrate the calculation of CRPD and CPRO; and (iii) perform a thorough experimental evaluation using benchmarks to compare the performance of the integrated analysis against the separate calculation of CRPD and CPRO.

Original language	English
Title of host publication	Proceedings - 2017 IEEE Real-Time Systems Symposium, RTSS 2017
Publisher	Institute of Electrical and Electronics Engineers
Pages	188-198
Number of pages	11
ISBN (Electronic)	9781538614143
DOIs	https://doi.org/10.1109/RTSS.2017.00025
Publication status	Published - 1 Feb 2018
Externally published	Yes
Event	38th IEEE Real-Time Systems Symposium, RTSS 2017 - Paris, France Duration: 5 Oct 2017 → 8 Oct 2017 Conference number: 38

Conference

Conference	38th IEEE Real-Time Systems Symposium, RTSS 2017
Abbreviated title	RTSS 2017
Country/Territory	France
City	Paris
Period	5/10/17 → 8/10/17

Funding

Acknowledgments. This paper is supported by NWO Veni Project, “The time is now: Timing Verification for Safety-Critical Multi-Cores”and by the ESPRC grant, MCCps (EP/K011626/1), and also by the Inria International Chair program. EPSRC Research Data Management: No new primary data was created during this study. This work was partially supported by National Funds through FCT (Portuguese Foundation for Science and Technology) and co-financed by ERDF (European Regional Development Fund) under the Portugal2020 Program, within the CISTER Research Unit (CEC/04234); by FCT and the ESF (European Social Fund) through POPH (Portuguese Human Potential Operational Program), under PhD grant SFRH/BD/119150/2016.

Keywords

Cache-Related-Overheads
Real-time-Scheduling
Timing-Analysis
WCRT-Analysis

Access to Document

10.1109/RTSS.2017.00025

Cite this

@inproceedings{a25eb81a956f4efe9de6e10bcd10c6bf,

title = "Integrated Analysis of Cache Related Preemption Delays and Cache Persistence Reload Overheads",

abstract = "Schedulability analysis for tasks running on micro- processors with cache memory is incomplete without a treatment of Cache Related Preemption Delays (CRPD) and Cache Persistence Reload Overheads (CPRO). State-of-the-art analyses compute CRPD and CPRO independently, which might result in counting the same overhead more than once. In this paper, we analyze the pessimism associated with the independent calculation of CRPD and CPRO in comparison to an integrated approach. We answer two main questions: (1) Is it benecial to integrate the calculation of CRPD and CPRO? (2) When and to what extent can we gain in terms of schedulability by integrating the calculation of CRPD and CPRO? To achieve this, we (i) identify situations where considering CRPD and CPRO separately might result in overestimating the total memory overhead suffered by tasks, (ii) derive new analyses that integrate the calculation of CRPD and CPRO; and (iii) perform a thorough experimental evaluation using benchmarks to compare the performance of the integrated analysis against the separate calculation of CRPD and CPRO.",

keywords = "Cache-Related-Overheads, Real-time-Scheduling, Timing-Analysis, WCRT-Analysis",

author = "Rashid, \{Syed Aftab\} and Geoffrey Nelissen and Sebastian Altmeyer and Davis, \{Robert I.\} and Eduardo Tovar",

year = "2018",

month = feb,

day = "1",

doi = "10.1109/RTSS.2017.00025",

language = "English",

pages = "188--198",

booktitle = "Proceedings - 2017 IEEE Real-Time Systems Symposium, RTSS 2017",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

note = "38th IEEE Real-Time Systems Symposium, RTSS 2017, RTSS 2017 ; Conference date: 05-10-2017 Through 08-10-2017",

}

Rashid, SA, Nelissen, G, Altmeyer, S, Davis, RI & Tovar, E 2018, Integrated Analysis of Cache Related Preemption Delays and Cache Persistence Reload Overheads. in Proceedings - 2017 IEEE Real-Time Systems Symposium, RTSS 2017. Institute of Electrical and Electronics Engineers, pp. 188-198, 38th IEEE Real-Time Systems Symposium, RTSS 2017, Paris, France, 5/10/17. https://doi.org/10.1109/RTSS.2017.00025

Integrated Analysis of Cache Related Preemption Delays and Cache Persistence Reload Overheads. / Rashid, Syed Aftab; Nelissen, Geoffrey; Altmeyer, Sebastian et al.
Proceedings - 2017 IEEE Real-Time Systems Symposium, RTSS 2017. Institute of Electrical and Electronics Engineers, 2018. p. 188-198.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Integrated Analysis of Cache Related Preemption Delays and Cache Persistence Reload Overheads

AU - Rashid, Syed Aftab

AU - Nelissen, Geoffrey

AU - Altmeyer, Sebastian

AU - Davis, Robert I.

AU - Tovar, Eduardo

N1 - Conference code: 38

PY - 2018/2/1

Y1 - 2018/2/1

N2 - Schedulability analysis for tasks running on micro- processors with cache memory is incomplete without a treatment of Cache Related Preemption Delays (CRPD) and Cache Persistence Reload Overheads (CPRO). State-of-the-art analyses compute CRPD and CPRO independently, which might result in counting the same overhead more than once. In this paper, we analyze the pessimism associated with the independent calculation of CRPD and CPRO in comparison to an integrated approach. We answer two main questions: (1) Is it benecial to integrate the calculation of CRPD and CPRO? (2) When and to what extent can we gain in terms of schedulability by integrating the calculation of CRPD and CPRO? To achieve this, we (i) identify situations where considering CRPD and CPRO separately might result in overestimating the total memory overhead suffered by tasks, (ii) derive new analyses that integrate the calculation of CRPD and CPRO; and (iii) perform a thorough experimental evaluation using benchmarks to compare the performance of the integrated analysis against the separate calculation of CRPD and CPRO.

AB - Schedulability analysis for tasks running on micro- processors with cache memory is incomplete without a treatment of Cache Related Preemption Delays (CRPD) and Cache Persistence Reload Overheads (CPRO). State-of-the-art analyses compute CRPD and CPRO independently, which might result in counting the same overhead more than once. In this paper, we analyze the pessimism associated with the independent calculation of CRPD and CPRO in comparison to an integrated approach. We answer two main questions: (1) Is it benecial to integrate the calculation of CRPD and CPRO? (2) When and to what extent can we gain in terms of schedulability by integrating the calculation of CRPD and CPRO? To achieve this, we (i) identify situations where considering CRPD and CPRO separately might result in overestimating the total memory overhead suffered by tasks, (ii) derive new analyses that integrate the calculation of CRPD and CPRO; and (iii) perform a thorough experimental evaluation using benchmarks to compare the performance of the integrated analysis against the separate calculation of CRPD and CPRO.

KW - Cache-Related-Overheads

KW - Real-time-Scheduling

KW - Timing-Analysis

KW - WCRT-Analysis

UR - https://www.scopus.com/pages/publications/85046368696

U2 - 10.1109/RTSS.2017.00025

DO - 10.1109/RTSS.2017.00025

M3 - Conference contribution

AN - SCOPUS:85046368696

SP - 188

EP - 198

BT - Proceedings - 2017 IEEE Real-Time Systems Symposium, RTSS 2017

PB - Institute of Electrical and Electronics Engineers

T2 - 38th IEEE Real-Time Systems Symposium, RTSS 2017

Y2 - 5 October 2017 through 8 October 2017

ER -

Integrated Analysis of Cache Related Preemption Delays and Cache Persistence Reload Overheads

Abstract

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this