Ffmk: A fast and fault-tolerant microkernel-based system for exascale computing

Carsten Weinhold; Adam Lackorzynski; Jan Bierbaum; Martin Küttler; Maksym Planeta; Hannes Weisbach; Matthias Hille; Hermann Härtig; Alexander Margolin; Dror Sharf; Ely Levy; Pavel Gak; Amnon Barak; Masoud Gholami; Florian Schintke; Thorsten Schütt; Alexander Reinefeld; Matthias Lieber; Wolfgang E. Nagel

doi:10.1007/978-3-030-47956-5_16

Ffmk: A fast and fault-tolerant microkernel-based system for exascale computing

Carsten Weinhold^*, Adam Lackorzynski, Jan Bierbaum, Martin Küttler, Maksym Planeta, Hannes Weisbach, Matthias Hille, Hermann Härtig, Alexander Margolin, Dror Sharf, Ely Levy, Pavel Gak, Amnon Barak, Masoud Gholami, Florian Schintke, Thorsten Schütt, Alexander Reinefeld, Matthias Lieber, Wolfgang E. Nagel

^*Corresponding author for this work

The Rachel and Selim Benin School of Engineering and Computer Science

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

1 Scopus citations

Abstract

The FFMK project designs, builds and evaluates a system-software architecture to address the challenges expected in Exascale systems. In particular, these challenges include performance losses caused by the much larger impact of runtime variability within applications, hardware, and operating system (OS), as well as increased vulnerability to failures. The FFMK OS platform is built upon a multi-kernel architecture, which combines the L4Re microkernel and a virtualized Linux kernel into a noise-free, yet feature-rich execution environment. It further includes global, distributed platform management and system-level optimization services that transparently minimize checkpoint/restart overhead for applications. The project also researched algorithms to make collective operations fault tolerant in presence of failing nodes. In this paper, we describe the basic components, algorithms, and services we developed in Phase 2 of the project.

Original language	English
Title of host publication	Lecture Notes in Computational Science and Engineering
Publisher	Springer
Pages	483-516
Number of pages	34
DOIs	https://doi.org/10.1007/978-3-030-47956-5_16
State	Published - 2020

Publication series

Name	Lecture Notes in Computational Science and Engineering
Volume	136
ISSN (Print)	1439-7358
ISSN (Electronic)	2197-7100

Bibliographical note

Publisher Copyright:
© The Author(s) 2020.

Access to Document

10.1007/978-3-030-47956-5_16

Cite this

Weinhold, C., Lackorzynski, A., Bierbaum, J., Küttler, M., Planeta, M., Weisbach, H., Hille, M., Härtig, H., Margolin, A., Sharf, D., Levy, E., Gak, P., Barak, A., Gholami, M., Schintke, F., Schütt, T., Reinefeld, A., Lieber, M., & Nagel, W. E. (2020). Ffmk: A fast and fault-tolerant microkernel-based system for exascale computing. In Lecture Notes in Computational Science and Engineering (pp. 483-516). (Lecture Notes in Computational Science and Engineering; Vol. 136). Springer. https://doi.org/10.1007/978-3-030-47956-5_16

@inbook{99c6e007651f4d5880480e3b304e6400,

title = "Ffmk: A fast and fault-tolerant microkernel-based system for exascale computing",

abstract = "The FFMK project designs, builds and evaluates a system-software architecture to address the challenges expected in Exascale systems. In particular, these challenges include performance losses caused by the much larger impact of runtime variability within applications, hardware, and operating system (OS), as well as increased vulnerability to failures. The FFMK OS platform is built upon a multi-kernel architecture, which combines the L4Re microkernel and a virtualized Linux kernel into a noise-free, yet feature-rich execution environment. It further includes global, distributed platform management and system-level optimization services that transparently minimize checkpoint/restart overhead for applications. The project also researched algorithms to make collective operations fault tolerant in presence of failing nodes. In this paper, we describe the basic components, algorithms, and services we developed in Phase 2 of the project.",

author = "Carsten Weinhold and Adam Lackorzynski and Jan Bierbaum and Martin K{\"u}ttler and Maksym Planeta and Hannes Weisbach and Matthias Hille and Hermann H{\"a}rtig and Alexander Margolin and Dror Sharf and Ely Levy and Pavel Gak and Amnon Barak and Masoud Gholami and Florian Schintke and Thorsten Sch{\"u}tt and Alexander Reinefeld and Matthias Lieber and Nagel, {Wolfgang E.}",

note = "Publisher Copyright: {\textcopyright} The Author(s) 2020.",

year = "2020",

doi = "10.1007/978-3-030-47956-5_16",

language = "אנגלית",

series = "Lecture Notes in Computational Science and Engineering",

publisher = "Springer",

pages = "483--516",

booktitle = "Lecture Notes in Computational Science and Engineering",

}

Weinhold, C, Lackorzynski, A, Bierbaum, J, Küttler, M, Planeta, M, Weisbach, H, Hille, M, Härtig, H, Margolin, A, Sharf, D, Levy, E, Gak, P, Barak, A, Gholami, M, Schintke, F, Schütt, T, Reinefeld, A, Lieber, M & Nagel, WE 2020, Ffmk: A fast and fault-tolerant microkernel-based system for exascale computing. in Lecture Notes in Computational Science and Engineering. Lecture Notes in Computational Science and Engineering, vol. 136, Springer, pp. 483-516. https://doi.org/10.1007/978-3-030-47956-5_16

Ffmk: A fast and fault-tolerant microkernel-based system for exascale computing. / Weinhold, Carsten; Lackorzynski, Adam; Bierbaum, Jan et al.
Lecture Notes in Computational Science and Engineering. Springer, 2020. p. 483-516 (Lecture Notes in Computational Science and Engineering; Vol. 136).

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

TY - CHAP

T1 - Ffmk

T2 - A fast and fault-tolerant microkernel-based system for exascale computing

AU - Weinhold, Carsten

AU - Lackorzynski, Adam

AU - Bierbaum, Jan

AU - Küttler, Martin

AU - Planeta, Maksym

AU - Weisbach, Hannes

AU - Hille, Matthias

AU - Härtig, Hermann

AU - Margolin, Alexander

AU - Sharf, Dror

AU - Levy, Ely

AU - Gak, Pavel

AU - Barak, Amnon

AU - Gholami, Masoud

AU - Schintke, Florian

AU - Schütt, Thorsten

AU - Reinefeld, Alexander

AU - Lieber, Matthias

AU - Nagel, Wolfgang E.

PY - 2020

Y1 - 2020

N2 - The FFMK project designs, builds and evaluates a system-software architecture to address the challenges expected in Exascale systems. In particular, these challenges include performance losses caused by the much larger impact of runtime variability within applications, hardware, and operating system (OS), as well as increased vulnerability to failures. The FFMK OS platform is built upon a multi-kernel architecture, which combines the L4Re microkernel and a virtualized Linux kernel into a noise-free, yet feature-rich execution environment. It further includes global, distributed platform management and system-level optimization services that transparently minimize checkpoint/restart overhead for applications. The project also researched algorithms to make collective operations fault tolerant in presence of failing nodes. In this paper, we describe the basic components, algorithms, and services we developed in Phase 2 of the project.

AB - The FFMK project designs, builds and evaluates a system-software architecture to address the challenges expected in Exascale systems. In particular, these challenges include performance losses caused by the much larger impact of runtime variability within applications, hardware, and operating system (OS), as well as increased vulnerability to failures. The FFMK OS platform is built upon a multi-kernel architecture, which combines the L4Re microkernel and a virtualized Linux kernel into a noise-free, yet feature-rich execution environment. It further includes global, distributed platform management and system-level optimization services that transparently minimize checkpoint/restart overhead for applications. The project also researched algorithms to make collective operations fault tolerant in presence of failing nodes. In this paper, we describe the basic components, algorithms, and services we developed in Phase 2 of the project.

UR - http://www.scopus.com/inward/record.url?scp=85089615234&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-47956-5_16

DO - 10.1007/978-3-030-47956-5_16

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.chapter???

AN - SCOPUS:85089615234

T3 - Lecture Notes in Computational Science and Engineering

SP - 483

EP - 516

BT - Lecture Notes in Computational Science and Engineering

PB - Springer

ER -

Ffmk: A fast and fault-tolerant microkernel-based system for exascale computing

Abstract

Publication series

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this