Quantum Chemistry in Dataflow: Density-Fitting MP2

Bridgette Cooper; Stephen Girdlestone; Pavel Burovskiy; Georgi Gaydadjiev; Vitali Averbukh; Peter J. Knowles; Wayne Luk

doi:10.1021/acs.jctc.7b00649

Quantum Chemistry in Dataflow: Density-Fitting MP2

Bridgette Cooper^*, Stephen Girdlestone, Pavel Burovskiy, Georgi Gaydadjiev, Vitali Averbukh, Peter J. Knowles, Wayne Luk

^*Corresponding author for this work

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

8 Citations (Scopus)

Abstract

We demonstrate the use of dataflow technology in the computation of the correlation energy in molecules at the Møller-Plesset perturbation theory (MP2) level. Specifically, we benchmark density fitting (DF)-MP2 for as many as 168 atoms (in valinomycin) and show that speed-ups between 3 and 3.8 times can be achieved when compared to the MOLPRO package run on a single CPU. Acceleration is achieved by offloading the matrix multiplications steps in DF-MP2 to Dataflow Engines (DFEs). We project that the acceleration factor could be as much as 24 with the next generation of DFEs.

Original language	English
Pages (from-to)	5265-5272
Number of pages	8
Journal	Journal of chemical theory and computation
Volume	13
Issue number	11
DOIs	https://doi.org/10.1021/acs.jctc.7b00649
Publication status	Published - 14 Nov 2017
Externally published	Yes

Access to Document

10.1021/acs.jctc.7b00649

Cite this

@article{eb3788db53fe48c993f7a9e7aef6291c,

title = "Quantum Chemistry in Dataflow: Density-Fitting MP2",

abstract = "We demonstrate the use of dataflow technology in the computation of the correlation energy in molecules at the M{\o}ller-Plesset perturbation theory (MP2) level. Specifically, we benchmark density fitting (DF)-MP2 for as many as 168 atoms (in valinomycin) and show that speed-ups between 3 and 3.8 times can be achieved when compared to the MOLPRO package run on a single CPU. Acceleration is achieved by offloading the matrix multiplications steps in DF-MP2 to Dataflow Engines (DFEs). We project that the acceleration factor could be as much as 24 with the next generation of DFEs.",

author = "Bridgette Cooper and Stephen Girdlestone and Pavel Burovskiy and Georgi Gaydadjiev and Vitali Averbukh and Knowles, {Peter J.} and Wayne Luk",

year = "2017",

month = nov,

day = "14",

doi = "10.1021/acs.jctc.7b00649",

language = "English",

volume = "13",

pages = "5265--5272",

journal = "Journal of chemical theory and computation",

issn = "1549-9618",

publisher = "American Chemical Society (ACS)",

number = "11",

}

TY - JOUR

T1 - Quantum Chemistry in Dataflow

T2 - Density-Fitting MP2

AU - Cooper, Bridgette

AU - Girdlestone, Stephen

AU - Burovskiy, Pavel

AU - Gaydadjiev, Georgi

AU - Averbukh, Vitali

AU - Knowles, Peter J.

AU - Luk, Wayne

PY - 2017/11/14

Y1 - 2017/11/14

N2 - We demonstrate the use of dataflow technology in the computation of the correlation energy in molecules at the Møller-Plesset perturbation theory (MP2) level. Specifically, we benchmark density fitting (DF)-MP2 for as many as 168 atoms (in valinomycin) and show that speed-ups between 3 and 3.8 times can be achieved when compared to the MOLPRO package run on a single CPU. Acceleration is achieved by offloading the matrix multiplications steps in DF-MP2 to Dataflow Engines (DFEs). We project that the acceleration factor could be as much as 24 with the next generation of DFEs.

AB - We demonstrate the use of dataflow technology in the computation of the correlation energy in molecules at the Møller-Plesset perturbation theory (MP2) level. Specifically, we benchmark density fitting (DF)-MP2 for as many as 168 atoms (in valinomycin) and show that speed-ups between 3 and 3.8 times can be achieved when compared to the MOLPRO package run on a single CPU. Acceleration is achieved by offloading the matrix multiplications steps in DF-MP2 to Dataflow Engines (DFEs). We project that the acceleration factor could be as much as 24 with the next generation of DFEs.

UR - http://www.scopus.com/inward/record.url?scp=85034256199&partnerID=8YFLogxK

U2 - 10.1021/acs.jctc.7b00649

DO - 10.1021/acs.jctc.7b00649

M3 - Article

AN - SCOPUS:85034256199

SN - 1549-9618

VL - 13

SP - 5265

EP - 5272

JO - Journal of chemical theory and computation

JF - Journal of chemical theory and computation

IS - 11

ER -

Quantum Chemistry in Dataflow: Density-Fitting MP2

Abstract

Access to Document

Other files and links

Fingerprint

Cite this