A clock synchronization strategy for minimizing clock variance at runtime in high-end computing environments

Terry Jones; Gregory A. Koenig

doi:10.1109/SBAC-PAD.2010.33

A clock synchronization strategy for minimizing clock variance at runtime in high-end computing environments

Terry Jones, Gregory A. Koenig

Intelligent Systems and Facilities

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

11 Scopus citations

Abstract

We present a new software-based clock synchronization scheme that provides high precision time agreement among distributed memory nodes. The technique is designed to minimize variance from a reference chimer during runtime and with minimal time-request latency. Our scheme permits initial unbounded variations in time and corrects both slow and fast chimers (clock skew). An implementation developed within the context of the MPI message passing interface is described and time coordination measurements are presented. Among our results, the mean time variance among a set of nodes improved from 20.0 milliseconds under standard Network Time Protocol (NTP) to 2.29 μsecs under our scheme.

Original language	English
Title of host publication	Proceedings - 22nd International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2010
Pages	207-214
Number of pages	8
DOIs	https://doi.org/10.1109/SBAC-PAD.2010.33
State	Published - 2010
Event	22nd International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2010 - Petropolis, Brazil Duration: Oct 27 2010 → Oct 30 2010

Publication series

Name	Proceedings - 22nd International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2010

Conference

Conference	22nd International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2010
Country/Territory	Brazil
City	Petropolis
Period	10/27/10 → 10/30/10

Keywords

Clock synchronization
MPI
Programming tools
Supercomputing
System software
Time service

Access to Document

10.1109/SBAC-PAD.2010.33

Cite this

Jones, T., & Koenig, G. A. (2010). A clock synchronization strategy for minimizing clock variance at runtime in high-end computing environments. In Proceedings - 22nd International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2010 (pp. 207-214). Article 5644949 (Proceedings - 22nd International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2010). https://doi.org/10.1109/SBAC-PAD.2010.33

Jones, Terry ; Koenig, Gregory A. / A clock synchronization strategy for minimizing clock variance at runtime in high-end computing environments. Proceedings - 22nd International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2010. 2010. pp. 207-214 (Proceedings - 22nd International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2010).

@inproceedings{4cd5e458304441909d7798f936e66a76,

title = "A clock synchronization strategy for minimizing clock variance at runtime in high-end computing environments",

abstract = "We present a new software-based clock synchronization scheme that provides high precision time agreement among distributed memory nodes. The technique is designed to minimize variance from a reference chimer during runtime and with minimal time-request latency. Our scheme permits initial unbounded variations in time and corrects both slow and fast chimers (clock skew). An implementation developed within the context of the MPI message passing interface is described and time coordination measurements are presented. Among our results, the mean time variance among a set of nodes improved from 20.0 milliseconds under standard Network Time Protocol (NTP) to 2.29 μsecs under our scheme.",

keywords = "Clock synchronization, MPI, Programming tools, Supercomputing, System software, Time service",

author = "Terry Jones and Koenig, {Gregory A.}",

year = "2010",

doi = "10.1109/SBAC-PAD.2010.33",

language = "English",

isbn = "9780769542164",

series = "Proceedings - 22nd International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2010",

pages = "207--214",

booktitle = "Proceedings - 22nd International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2010",

note = "22nd International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2010 ; Conference date: 27-10-2010 Through 30-10-2010",

}

Jones, T & Koenig, GA 2010, A clock synchronization strategy for minimizing clock variance at runtime in high-end computing environments. in Proceedings - 22nd International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2010., 5644949, Proceedings - 22nd International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2010, pp. 207-214, 22nd International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2010, Petropolis, Brazil, 10/27/10. https://doi.org/10.1109/SBAC-PAD.2010.33

A clock synchronization strategy for minimizing clock variance at runtime in high-end computing environments. / Jones, Terry; Koenig, Gregory A.
Proceedings - 22nd International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2010. 2010. p. 207-214 5644949 (Proceedings - 22nd International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2010).

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

TY - GEN

T1 - A clock synchronization strategy for minimizing clock variance at runtime in high-end computing environments

AU - Jones, Terry

AU - Koenig, Gregory A.

PY - 2010

Y1 - 2010

N2 - We present a new software-based clock synchronization scheme that provides high precision time agreement among distributed memory nodes. The technique is designed to minimize variance from a reference chimer during runtime and with minimal time-request latency. Our scheme permits initial unbounded variations in time and corrects both slow and fast chimers (clock skew). An implementation developed within the context of the MPI message passing interface is described and time coordination measurements are presented. Among our results, the mean time variance among a set of nodes improved from 20.0 milliseconds under standard Network Time Protocol (NTP) to 2.29 μsecs under our scheme.

AB - We present a new software-based clock synchronization scheme that provides high precision time agreement among distributed memory nodes. The technique is designed to minimize variance from a reference chimer during runtime and with minimal time-request latency. Our scheme permits initial unbounded variations in time and corrects both slow and fast chimers (clock skew). An implementation developed within the context of the MPI message passing interface is described and time coordination measurements are presented. Among our results, the mean time variance among a set of nodes improved from 20.0 milliseconds under standard Network Time Protocol (NTP) to 2.29 μsecs under our scheme.

KW - Clock synchronization

KW - MPI

KW - Programming tools

KW - Supercomputing

KW - System software

KW - Time service

UR - http://www.scopus.com/inward/record.url?scp=78650758041&partnerID=8YFLogxK

U2 - 10.1109/SBAC-PAD.2010.33

DO - 10.1109/SBAC-PAD.2010.33

M3 - Conference contribution

AN - SCOPUS:78650758041

SN - 9780769542164

T3 - Proceedings - 22nd International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2010

SP - 207

EP - 214

BT - Proceedings - 22nd International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2010

T2 - 22nd International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2010

Y2 - 27 October 2010 through 30 October 2010

ER -

Jones T, Koenig GA. A clock synchronization strategy for minimizing clock variance at runtime in high-end computing environments. In Proceedings - 22nd International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2010. 2010. p. 207-214. 5644949. (Proceedings - 22nd International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2010). doi: 10.1109/SBAC-PAD.2010.33

A clock synchronization strategy for minimizing clock variance at runtime in high-end computing environments

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this