TETRiS: A Multi-Application Run-Time System for Predictable Execution of Static Mappings

Andrés Goens; Robert Khasanov; Jeronimo Castrillon; Marcus Hähnel; Till Smejkal; Hermann Härtig

doi:10.1145/3078659.3078663

TETRiS: A Multi-Application Run-Time System for Predictable Execution of Static Mappings

Andrés Goens, Robert Khasanov, Jeronimo Castrillon, Marcus Hähnel, Till Smejkal, Hermann Härtig

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

For embedded system software, it is common to use static mappings of tasks to cores. This becomes considerably more challenging in multi-application scenarios. In this paper, we propose TETRiS, a multi-application run-time system for static mappings for heterogeneous system-on-chip architectures. It leverages compile-time information to map and migrate tasks in a fashion that preserves the predictable performance of using static mappings, allowing the system to accommodate multiple applications. TETRiS runs on off-the-shelf embedded systems and is Linux-compatible. We embed our approach in a state-of-the-art compiler for multicore systems and evaluate the proposed run-time system in a modern heterogeneous platform using realistic benchmarks. We present two experiments whose execution time and energy consumptions are comparable to those obtained by the highly-optimized Linux scheduler CFS, and where execution time variance is reduced by a factor of 510, and energy consumption variance by a factor of 83.

Original language	English
Title of host publication	Proceedings of the 20th International Workshop on Software and Compilers for Embedded Systems
Editors	Sander Stuijk
Place of Publication	New York, NY, USA
Publisher	ACM Association for Computing Machinery
Pages	11–20
Number of pages	10
ISBN (Print)	9781450350396
DOIs	https://doi.org/10.1145/3078659.3078663
Publication status	Published - 12 Jun 2017
Event	20th International Workshop on Software and Compilers for Embedded Systems - St. Goar, Germany Duration: 12 Jun 2017 → 13 Jun 2017 Conference number: 20

Publication series

Name	SCOPES '17
Publisher	Association for Computing Machinery

Conference

Conference	20th International Workshop on Software and Compilers for Embedded Systems
Abbreviated title	SCOPES 2017
Country/Territory	Germany
City	St. Goar
Period	12/06/17 → 13/06/17

Keywords / Materials (for Non-textual outputs)

run-time
Heterogeneous
adaptivity
symmetry
multi-application
MPSoC

Access to Document

10.1145/3078659.3078663Licence: All Rights Reserved

https://dl.acm.org/doi/10.1145/3078659.3078663Licence: All Rights Reserved

Cite this

Goens, A., Khasanov, R., Castrillon, J., Hähnel, M., Smejkal, T., & Härtig, H. (2017). TETRiS: A Multi-Application Run-Time System for Predictable Execution of Static Mappings. In S. Stuijk (Ed.), Proceedings of the 20th International Workshop on Software and Compilers for Embedded Systems (pp. 11–20). (SCOPES '17). ACM Association for Computing Machinery. https://doi.org/10.1145/3078659.3078663

@inproceedings{15683b6331944d31ab3b07302dc54ae1,

title = "TETRiS: A Multi-Application Run-Time System for Predictable Execution of Static Mappings",

abstract = "For embedded system software, it is common to use static mappings of tasks to cores. This becomes considerably more challenging in multi-application scenarios. In this paper, we propose TETRiS, a multi-application run-time system for static mappings for heterogeneous system-on-chip architectures. It leverages compile-time information to map and migrate tasks in a fashion that preserves the predictable performance of using static mappings, allowing the system to accommodate multiple applications. TETRiS runs on off-the-shelf embedded systems and is Linux-compatible. We embed our approach in a state-of-the-art compiler for multicore systems and evaluate the proposed run-time system in a modern heterogeneous platform using realistic benchmarks. We present two experiments whose execution time and energy consumptions are comparable to those obtained by the highly-optimized Linux scheduler CFS, and where execution time variance is reduced by a factor of 510, and energy consumption variance by a factor of 83.",

keywords = "run-time, Heterogeneous, adaptivity, symmetry, multi-application, MPSoC",

author = "Andr{\'e}s Goens and Robert Khasanov and Jeronimo Castrillon and Marcus H{\"a}hnel and Till Smejkal and Hermann H{\"a}rtig",

year = "2017",

month = jun,

day = "12",

doi = "10.1145/3078659.3078663",

language = "English",

isbn = "9781450350396",

series = "SCOPES '17",

publisher = "ACM Association for Computing Machinery",

pages = "11–20",

editor = "Sander Stuijk",

booktitle = "Proceedings of the 20th International Workshop on Software and Compilers for Embedded Systems",

note = "20th International Workshop on Software and Compilers for Embedded Systems, SCOPES 2017 ; Conference date: 12-06-2017 Through 13-06-2017",

}

Goens, A, Khasanov, R, Castrillon, J, Hähnel, M, Smejkal, T & Härtig, H 2017, TETRiS: A Multi-Application Run-Time System for Predictable Execution of Static Mappings. in S Stuijk (ed.), Proceedings of the 20th International Workshop on Software and Compilers for Embedded Systems. SCOPES '17, ACM Association for Computing Machinery, New York, NY, USA, pp. 11–20, 20th International Workshop on Software and Compilers for Embedded Systems, St. Goar, Rhineland-Palatinate, Germany, 12/06/17. https://doi.org/10.1145/3078659.3078663

TETRiS: A Multi-Application Run-Time System for Predictable Execution of Static Mappings. / Goens, Andrés; Khasanov, Robert; Castrillon, Jeronimo et al.
Proceedings of the 20th International Workshop on Software and Compilers for Embedded Systems. ed. / Sander Stuijk. New York, NY, USA: ACM Association for Computing Machinery, 2017. p. 11–20 (SCOPES '17).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - TETRiS: A Multi-Application Run-Time System for Predictable Execution of Static Mappings

AU - Goens, Andrés

AU - Khasanov, Robert

AU - Castrillon, Jeronimo

AU - Hähnel, Marcus

AU - Smejkal, Till

AU - Härtig, Hermann

N1 - Conference code: 20

PY - 2017/6/12

Y1 - 2017/6/12

N2 - For embedded system software, it is common to use static mappings of tasks to cores. This becomes considerably more challenging in multi-application scenarios. In this paper, we propose TETRiS, a multi-application run-time system for static mappings for heterogeneous system-on-chip architectures. It leverages compile-time information to map and migrate tasks in a fashion that preserves the predictable performance of using static mappings, allowing the system to accommodate multiple applications. TETRiS runs on off-the-shelf embedded systems and is Linux-compatible. We embed our approach in a state-of-the-art compiler for multicore systems and evaluate the proposed run-time system in a modern heterogeneous platform using realistic benchmarks. We present two experiments whose execution time and energy consumptions are comparable to those obtained by the highly-optimized Linux scheduler CFS, and where execution time variance is reduced by a factor of 510, and energy consumption variance by a factor of 83.

AB - For embedded system software, it is common to use static mappings of tasks to cores. This becomes considerably more challenging in multi-application scenarios. In this paper, we propose TETRiS, a multi-application run-time system for static mappings for heterogeneous system-on-chip architectures. It leverages compile-time information to map and migrate tasks in a fashion that preserves the predictable performance of using static mappings, allowing the system to accommodate multiple applications. TETRiS runs on off-the-shelf embedded systems and is Linux-compatible. We embed our approach in a state-of-the-art compiler for multicore systems and evaluate the proposed run-time system in a modern heterogeneous platform using realistic benchmarks. We present two experiments whose execution time and energy consumptions are comparable to those obtained by the highly-optimized Linux scheduler CFS, and where execution time variance is reduced by a factor of 510, and energy consumption variance by a factor of 83.

KW - run-time

KW - Heterogeneous

KW - adaptivity

KW - symmetry

KW - multi-application

KW - MPSoC

U2 - 10.1145/3078659.3078663

DO - 10.1145/3078659.3078663

M3 - Conference contribution

SN - 9781450350396

T3 - SCOPES '17

SP - 11

EP - 20

BT - Proceedings of the 20th International Workshop on Software and Compilers for Embedded Systems

A2 - Stuijk, Sander

PB - ACM Association for Computing Machinery

CY - New York, NY, USA

T2 - 20th International Workshop on Software and Compilers for Embedded Systems

Y2 - 12 June 2017 through 13 June 2017

ER -

Goens A, Khasanov R, Castrillon J, Hähnel M, Smejkal T, Härtig H. TETRiS: A Multi-Application Run-Time System for Predictable Execution of Static Mappings. In Stuijk S, editor, Proceedings of the 20th International Workshop on Software and Compilers for Embedded Systems. New York, NY, USA: ACM Association for Computing Machinery. 2017. p. 11–20. (SCOPES '17). doi: 10.1145/3078659.3078663

TETRiS: A Multi-Application Run-Time System for Predictable Execution of Static Mappings

Abstract

Publication series

Conference

Keywords / Materials (for Non-textual outputs)

Access to Document

Fingerprint

Cite this