Vectorization of the generalized Born model for molecular dynamics on shared-memory computers

Carlos P. Sosa; Tom Hewitt; Matthew R. Lee; David A. Case

doi:10.1016/S0166-1280(01)00497-3

Vectorization of the generalized Born model for molecular dynamics on shared-memory computers

Carlos P. Sosa, Tom Hewitt, Matthew R. Lee, David A. Case

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

7 Scopus citations

Abstract

Vectorization and performance of the generalized Born solvation model (GB) as it is implemented in the AMBER program is presented in this study. Nonbonded interactions computed within the generalized Born model use the pairwise approximation of Hawkins et al. [J. Phys. Chem. 100 (1996) 19,824], which is in turn a variant of the model proposed by Schaeffer and Froemmel [J. Mol. Biol. 216 (1990) 1045]. The performance of this implementation on CRAY SV1 vector machines is discussed with reference to several proteins. Loop vectorization has shown improvements by as much as a factor of 10. Comparison of the timings of the GB method against simulations using fully hydrated proteins favors the GB method.

Original language	English (US)
Pages (from-to)	193-201
Number of pages	9
Journal	Journal of Molecular Structure: THEOCHEM
Volume	549
Issue number	1-2
DOIs	https://doi.org/10.1016/S0166-1280(01)00497-3
State	Published - Aug 6 2001
Externally published	Yes

All Science Journal Classification (ASJC) codes

Biochemistry
Condensed Matter Physics
Physical and Theoretical Chemistry

Keywords

AMBER
CRAY SV1
Generalized Born
Molecular dynamics
Vector

Access to Document

10.1016/S0166-1280(01)00497-3

Cite this

@article{78fedffcbae746ada3c41ef0e5cd48af,

title = "Vectorization of the generalized Born model for molecular dynamics on shared-memory computers",

abstract = "Vectorization and performance of the generalized Born solvation model (GB) as it is implemented in the AMBER program is presented in this study. Nonbonded interactions computed within the generalized Born model use the pairwise approximation of Hawkins et al. [J. Phys. Chem. 100 (1996) 19,824], which is in turn a variant of the model proposed by Schaeffer and Froemmel [J. Mol. Biol. 216 (1990) 1045]. The performance of this implementation on CRAY SV1 vector machines is discussed with reference to several proteins. Loop vectorization has shown improvements by as much as a factor of 10. Comparison of the timings of the GB method against simulations using fully hydrated proteins favors the GB method.",

keywords = "AMBER, CRAY SV1, Generalized Born, Molecular dynamics, Vector",

author = "Sosa, {Carlos P.} and Tom Hewitt and Lee, {Matthew R.} and Case, {David A.}",

note = "Funding Information: We would like to thank CRAY Inc. for providing all the computing facilities to carry out this research. CPS would also like to thank Vince Graziano for very helpful discussions on compilers issues. DAC acknowledges support from NIH grant RR12255. ",

year = "2001",

month = aug,

day = "6",

doi = "10.1016/S0166-1280(01)00497-3",

language = "English (US)",

volume = "549",

pages = "193--201",

journal = "Computational and Theoretical Chemistry",

issn = "2210-271X",

publisher = "Elsevier BV",

number = "1-2",

}

TY - JOUR

T1 - Vectorization of the generalized Born model for molecular dynamics on shared-memory computers

AU - Sosa, Carlos P.

AU - Hewitt, Tom

AU - Lee, Matthew R.

AU - Case, David A.

N1 - Funding Information: We would like to thank CRAY Inc. for providing all the computing facilities to carry out this research. CPS would also like to thank Vince Graziano for very helpful discussions on compilers issues. DAC acknowledges support from NIH grant RR12255.

PY - 2001/8/6

Y1 - 2001/8/6

N2 - Vectorization and performance of the generalized Born solvation model (GB) as it is implemented in the AMBER program is presented in this study. Nonbonded interactions computed within the generalized Born model use the pairwise approximation of Hawkins et al. [J. Phys. Chem. 100 (1996) 19,824], which is in turn a variant of the model proposed by Schaeffer and Froemmel [J. Mol. Biol. 216 (1990) 1045]. The performance of this implementation on CRAY SV1 vector machines is discussed with reference to several proteins. Loop vectorization has shown improvements by as much as a factor of 10. Comparison of the timings of the GB method against simulations using fully hydrated proteins favors the GB method.

AB - Vectorization and performance of the generalized Born solvation model (GB) as it is implemented in the AMBER program is presented in this study. Nonbonded interactions computed within the generalized Born model use the pairwise approximation of Hawkins et al. [J. Phys. Chem. 100 (1996) 19,824], which is in turn a variant of the model proposed by Schaeffer and Froemmel [J. Mol. Biol. 216 (1990) 1045]. The performance of this implementation on CRAY SV1 vector machines is discussed with reference to several proteins. Loop vectorization has shown improvements by as much as a factor of 10. Comparison of the timings of the GB method against simulations using fully hydrated proteins favors the GB method.

KW - AMBER

KW - CRAY SV1

KW - Generalized Born

KW - Molecular dynamics

KW - Vector

UR - http://www.scopus.com/inward/record.url?scp=0035817214&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035817214&partnerID=8YFLogxK

U2 - 10.1016/S0166-1280(01)00497-3

DO - 10.1016/S0166-1280(01)00497-3

M3 - Article

AN - SCOPUS:0035817214

SN - 2210-271X

VL - 549

SP - 193

EP - 201

JO - Computational and Theoretical Chemistry

JF - Computational and Theoretical Chemistry

IS - 1-2

ER -

Vectorization of the generalized Born model for molecular dynamics on shared-memory computers

Abstract

All Science Journal Classification (ASJC) codes

Keywords

Access to Document

Other files and links

Fingerprint

Cite this