DOCSLIB.ORG

Compact Orbitals Enable Low-Cost Linear-Scaling Ab Initio Molecular Dynamics for Weakly-Interacting Systems Hayden Scheiber,1, A) Yifei Shi,1 and Rustam Z

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Compact Orbitals Enable Low-Cost Linear-Scaling Ab Initio Molecular Dynamics for Weakly-Interacting Systems Hayden Scheiber,1, A) Yifei Shi,1 and Rustam Z Compact orbitals enable low-cost linear-scaling ab initio molecular dynamics for weakly-interacting systems Hayden Scheiber,1, a) Yifei Shi,1 and Rustam Z. Khaliullin1, b) Department of Chemistry, McGill University, 801 Sherbrooke St. West, Montreal, QC H3A 0B8, Canada Today, ab initio molecular dynamics (AIMD) relies on the locality of one-electron density matrices to achieve linear growth of computation time with systems size, crucial in large-scale simulations. While Kohn-Sham orbitals strictly localized within predefined radii can offer substantial computational advantages over density matrices, such compact orbitals are not used in AIMD because a compact representation of the electronic ground state is difficult to find. Here, a robust method for maintaining compact orbitals close to the ground state is coupled with a modified Langevin integrator to produce stable nuclear dynamics for molecular and ionic systems. This eliminates a density matrix optimization and enables first orbital-only linear-scaling AIMD. An application to liquid water demonstrates that low computational overhead of the new method makes it ideal for routine medium-scale simulations while its linear-scaling complexity allows to extend first- principle studies of molecular systems to completely new physical phenomena on previously inaccessible length scales. Since the unification of molecular dynamics and den- LS methods restrict their use in dynamical simulations sity functional theory (DFT)1, ab initio molecular dy- to very short time scales, systems of low dimensions, namics (AIMD) has become an important tool to study and low-quality minimal basis sets6,18–20. On typical processes in molecules and materials. Unfortunately, the length and time scales required in practical and accurate computational cost of the conventional Kohn-Sham (KS) AIMD simulations, LS DFT still cannot compete with DFT grows cubically with the number of atoms, which the straightforward low-cost cubically-scaling KS DFT. severely limits the length scales accessible by AIMD. To In this work, we present an AIMD method that over- address this issue, substantial efforts have been directed comes difficulties of DFT based on compact orbitals to to the development of linear-scaling (LS) DFT. achieve LS with extremely low computational overhead. In all LS DFT methods, the delocalized eigenstates To demonstrate advantages of the new method we ap- of the effective KS Hamiltonian must be replaced with plied it here to systems of weakly-interacting molecules. an alternative set of local electronic descriptors. Most However, the same approach is readily applicable to LS methods2–5 explore the natural locality of the one- systems of strongly-interacting fragments that do not electron density matrix (DM). However, the DM DFT be- form strong covalent bonds such as ionic materials— comes advantageous only for impractically large systems salts, liquids, and semiconductors. A generalization of when accurate multifunction basis sets are used3,6–8. the method to all finite-gap systems, including covalently This issue is rectified in optimal-basis DM methods9–11 bonded atoms, will be reported later. that contract large basis sets into a small number of new localized functions and then optimize the DM in The new AIMD method utilizes a recently developed the contracted basis. Despite becoming the most popu- LS DFT8 based on absolutely localized molecular orbitals lar approach to LS DFT, the efficiency of these methods (ALMOs) – compact orbitals first described in Ref. 21. is hampered by the costly optimization of both the con- Unlike delocalized KS orbitals, each ALMO has its own tracted orbitals and the DM12. From this point of view, a localization center and a predefined localization radius 8,21 direct variation of compact molecular orbitals—orbitals Rc that typically includes nearby atoms or molecules . that are strictly localized within predefined regions—is In the current implementation, a localization center is preferable because LS can be achieved with significantly defined as a set of all Gaussian atomic orbitals of one fewer variables. Advantages of the orbitals-only LS DFT molecule. However, the approach can use other local and are especially pronounced in accurate calculations that nonlocal basis sets22,23. The key feature of ALMO DFT require many basis functions per atom. Unfortunately, is that its one-electron wavefunctions are constructed in the development of promising orbital-based LS methods a two-stage self-consistent-field (SCF) procedure8 to cir- has been all but abandoned13,14 because of the inherently cumvent the problem of the sluggish variational optimiza- difficult optimization of localized orbitals2,13–17. tion emphasized above. In the first stage, ALMOs are Thus, despite impressive progress of the LS descrip- constrained to their localization centers24 whereas, in the tion of the electronic and atomic structure of large static second stage, ALMOs are relaxed to allow delocalization systems6,18, the high computational overhead of existing onto the neighbor molecules within their localization ra- dius Rc. To achieve a robust optimization in the prob- lematic second stage, it is important to keep the delocal- ization component of the trial wavefunction orthogonal a)Electronic mail: [email protected] to the fixed orbitals obtained in the first stage. For math- b)Electronic mail: [email protected] ematical details, see the ALMO SCF method in Ref. 8. 2 ALMO constraints imposed by Rc prohibit electron zero systematically by increasing Rc and decreasing ǫSCF. density transfer between distant molecules, but retain Our approach to compensate for the missing δfiα(t) all other types of interaction such as long-range elec- term is inspired by the methodology introduced into trostatic, exchange, polarization, and—if the exchange- AIMD by Krajewski et al.27, formalized by K¨uhne et correlation (XC) functional includes them—dispersion al.28 and rationalized by Dai et al.29 before becoming in- interactions25. Since the importance of electron transfer formally known as the second generation Car-Parrinello decays exponentially with distance in finite-gap materi- molecular dynamics30. Adopting the principle of Refs. 27 als2, the ALMO approximation is expected to provide and 28, ALMO AIMD is chosen to be governed by the a natural and accurate representation of the electronic Langevin equation of motion that can be written in terms structure of molecular systems. Because of the greatly of the unknown reference forces reduced number of electronic descriptors and the robust KS γ optimization, the computational complexity of ALMO mir¨iα = fiα (t) − γmir˙iα + Riα(t), (2) DFT grows linearly with the number of molecules while where mi is the mass of atom i, riα is its position along its computational overhead remains very low. These fea- γ dimension α, γ is the Langevin scaling factor, and Riα(t) tures make ALMO DFT a promising method for accurate is the stochastic force represented by a zero-mean white AIMD simulations of large molecular systems. Gaussian noise The challenge of adopting ALMO DFT for dynamical γ simulations arises from the slightly nonvariational char- hRiα(t)i =0, (3) acter of the localized orbitals. While ALMOs are varia- γ γ ′ ′ hR (t)R (t )i =2kBTγmiδij δαβδ(t − t ). (4) tionally optimized in both SCF stages, the occupied sub- iα jβ space defined in the first stage must remain fixed dur- The last relation means that, for any value of γ, the ing the second stage to ensure convergence. In addition, damping and stochastic terms are in perfect balance and electron transfer effects can suddenly become inactive in trajectories generated with Eq. (2) will sample the canon- the course of a dynamical simulation when a neighboring ical ensemble at a specified temperature T 31. In the limit molecule crosses the localization threshold Rc. Futher- γ → 0, the Newton equation is recovered and the micro- more, the variational optimization in any AIMD method canonical ensemble is sampled. is never complete in practice and interrupted once the The main assumption of ALMO AIMD is that the error maximum norm of the gradient of the energy with respect in the ALMO forces is well approximated by Gaussian ∆ to the electronic descriptors drops below small but nev- noise Riα(t): ertheless finite convergence threshold ǫSCF. These errors ∆ do not affect the accuracy of static ALMO DFT calcula- δfiα(t)= Riα(t) (5) tions, geometry optimization, and Monte-Carlo simula- tions. Unfortunately they tend to accumulate in molec- that obeys ular dynamics trajectories leading to non-physical sam- ∆ hRiα(t)i =0, (6) pling and eventual failure. Traditional strategies to cope ∆ ∆ ′ ′ with these problems are computationally expensive and hRiα(t)Rjβ (t )i =2kBT ∆miδij δαβδ(t − t ). (7) include computing the nonvariational contribution to the 4,26 This assumption, shown to be well justified, allows us to forces via a variational coupled-perturbed procedure , rewrite the Langevin equation using the ALMO forces increasing Rc, and decreasing ǫSCF. In this work, we propose another approach that ob- ALMO γ+∆ mir¨iα = fiα (t) − γmir˙iα + Riα (t), (8) viates the need in a coupled-perturbed solver, relaxes tight constraints on Rc and ǫSCF, and thus enables us where the two stochastic terms are combined into one γ+∆ γ ∆ to maintain stable dynamics and to keep the algorithmic Riα = Riα + Riα. The only missing piece in the complexity and cost of simulations low. In our approach, modified Langevin equation is the value of ∆, which de- the forces on atoms are calculated approximately after scribes the strength of the newly introduced stochastic the two-stage ALMO SCF using a straightforward pro- term. This term compensates for imperfections in ALMO cedure that computes only the Hellmann-Feynman and forces and must be adjusted to re-balance the damping Pulay components and neglects the computationally in- and stochastic components in ALMO AIMD. tense nonvariational component of the forces. The differ- In principle, ∆ can be calculated using the integral of ence between these approximate ALMO forces and the Eq.
Load more

Recommended publications
  • Introducing ONETEP: Linear-Scaling Density Functional Simulations on Parallel Computers Chris-Kriton Skylaris,A) Peter D
    THE JOURNAL OF CHEMICAL PHYSICS 122, 084119 ͑2005͒ Introducing ONETEP: Linear-scaling density functional simulations on parallel computers Chris-Kriton Skylaris,a) Peter D. Haynes, Arash A. Mostofi, and Mike C. Payne Theory of Condensed Matter, Cavendish Laboratory, Madingley Road, Cambridge CB3 0HE, United Kingdom ͑Received 29 September 2004; accepted 4 November 2004; published online 23 February 2005͒ We present ONETEP ͑order-N electronic total energy package͒, a density functional program for parallel computers whose computational cost scales linearly with the number of atoms and the number of processors. ONETEP is based on our reformulation of the plane wave pseudopotential method which exploits the electronic localization that is inherent in systems with a nonvanishing band gap. We summarize the theoretical developments that enable the direct optimization of strictly localized quantities expressed in terms of a delocalized plane wave basis. These same localized quantities lead us to a physical way of dividing the computational effort among many processors to allow calculations to be performed efficiently on parallel supercomputers. We show with examples that ONETEP achieves excellent speedups with increasing numbers of processors and confirm that the time taken by ONETEP as a function of increasing number of atoms for a given number of processors is indeed linear. What distinguishes our approach is that the localization is achieved in a controlled and mathematically consistent manner so that ONETEP obtains the same accuracy as conventional cubic-scaling plane wave approaches and offers fast and stable convergence. We expect that calculations with ONETEP have the potential to provide quantitative theoretical predictions for problems involving thousands of atoms such as those often encountered in nanoscience and biophysics.
    [Show full text]
  • Natural Bond Orbital Analysis in the ONETEP Code: Applications to Large Protein Systems Louis P
    WWW.C-CHEM.ORG FULL PAPER Natural Bond Orbital Analysis in the ONETEP Code: Applications to Large Protein Systems Louis P. Lee,*[a] Daniel J. Cole,[a] Mike C. Payne,[a] and Chris-Kriton Skylaris[b] First principles electronic structure calculations are typically Generalized Wannier Functions of ONETEP to natural atomic performed in terms of molecular orbitals (or bands), providing a orbitals, NBO analysis can be performed within a localized straightforward theoretical avenue for approximations of region in such a way that ensures the results are identical to an increasing sophistication, but do not usually provide any analysis on the full system. We demonstrate the capabilities of qualitative chemical information about the system. We can this approach by performing illustrative studies of large derive such information via post-processing using natural bond proteins—namely, investigating changes in charge transfer orbital (NBO) analysis, which produces a chemical picture of between the heme group of myoglobin and its ligands with bonding in terms of localized Lewis-type bond and lone pair increasing system size and between a protein and its explicit orbitals that we can use to understand molecular structure and solvent, estimating the contribution of electronic delocalization interactions. We present NBO analysis of large-scale calculations to the stabilization of hydrogen bonds in the binding pocket of with the ONETEP linear-scaling density functional theory package, a drug-receptor complex, and observing, in situ, the n ! p* which we have interfaced with the NBO 5 analysis program. In hyperconjugative interactions between carbonyl groups that ONETEP calculations involving thousands of atoms, one is typically stabilize protein backbones.
    [Show full text]
  • Density Functional Theory (DFT)
    Herramientas mecano-cuánticas basadas en DFT para el estudio de moléculas y materiales en Materials Studio 7.0 Javier Ramos Biophysics of Macromolecular Systems group (BIOPHYM) Departamento de Física Macromolecular Instituto de Estructura de la Materia – CSIC [email protected] Webinar, 26 de Junio 2014 Anteriores webinars Como conseguir los videos y las presentaciones de anteriores webminars: Linkedin: Grupo de Química Computacional http://www.linkedin.com/groups/Química-computacional-7487634 Índice Density Functional Theory (DFT) The Jacob’s ladder DFT modules in Maretials Studio DMOL3, CASTEP and ONETEP XC functionals Basis functions Interfaces in Materials Studio Tasks Properties Example: n-butane conformations Density Functional Theory (DFT) DFT is built around the premise that the energy of an electronic system can be defined in terms of its electron probability density (ρ). (Hohenberg-Kohn Theorem) E 0 [ 0 ] Te [ 0 ] E ne [ 0 ] E ee [ 0 ] (easy) Kinetic Energy for ????? noninteracting (r )v (r ) dr electrons(easy) 1 E[]()()[]1 r r d r d r E e e2 1 2 1 2 X C r12 Classic Term(Coulomb) Non-classic Kohn-Sham orbitals Exchange & By minimizing the total energy functional applying the variational principle it is Correlation possible to get the SCF equations (Kohn-Sham) The Jacob’s Ladder Accurate form of XC potential Meta GGA Empirical (Fitting to Non-Empirical Generalized Gradient Approx. atomic properties) (physics rules) Local Density Approximation DFT modules in Materials Studio DMol3: Combine computational speed with the accuracy of quantum mechanical methods to predict materials properties reliably and quickly CASTEP: CASTEP offers simulation capabilities not found elsewhere, such as accurate prediction of phonon spectra, dielectric constants, and optical properties.
    [Show full text]
  • Quantum Chemistry (QC) on Gpus Feb
    Quantum Chemistry (QC) on GPUs Feb. 2, 2017 Overview of Life & Material Accelerated Apps MD: All key codes are GPU-accelerated QC: All key codes are ported or optimizing Great multi-GPU performance Focus on using GPU-accelerated math libraries, OpenACC directives Focus on dense (up to 16) GPU nodes &/or large # of GPU nodes GPU-accelerated and available today: ACEMD*, AMBER (PMEMD)*, BAND, CHARMM, DESMOND, ESPResso, ABINIT, ACES III, ADF, BigDFT, CP2K, GAMESS, GAMESS- Folding@Home, GPUgrid.net, GROMACS, HALMD, HOOMD-Blue*, UK, GPAW, LATTE, LSDalton, LSMS, MOLCAS, MOPAC2012, LAMMPS, Lattice Microbes*, mdcore, MELD, miniMD, NAMD, NWChem, OCTOPUS*, PEtot, QUICK, Q-Chem, QMCPack, OpenMM, PolyFTS, SOP-GPU* & more Quantum Espresso/PWscf, QUICK, TeraChem* Active GPU acceleration projects: CASTEP, GAMESS, Gaussian, ONETEP, Quantum Supercharger Library*, VASP & more green* = application where >90% of the workload is on GPU 2 MD vs. QC on GPUs “Classical” Molecular Dynamics Quantum Chemistry (MO, PW, DFT, Semi-Emp) Simulates positions of atoms over time; Calculates electronic properties; chemical-biological or ground state, excited states, spectral properties, chemical-material behaviors making/breaking bonds, physical properties Forces calculated from simple empirical formulas Forces derived from electron wave function (bond rearrangement generally forbidden) (bond rearrangement OK, e.g., bond energies) Up to millions of atoms Up to a few thousand atoms Solvent included without difficulty Generally in a vacuum but if needed, solvent treated classically
    [Show full text]
  • Introduction to DFT and the Plane-Wave Pseudopotential Method
    Introduction to DFT and the plane-wave pseudopotential method Keith Refson STFC Rutherford Appleton Laboratory Chilton, Didcot, OXON OX11 0QX 23 Apr 2014 Parallel Materials Modelling Packages @ EPCC 1 / 55 Introduction Synopsis Motivation Some ab initio codes Quantum-mechanical approaches Density Functional Theory Electronic Structure of Condensed Phases Total-energy calculations Introduction Basis sets Plane-waves and Pseudopotentials How to solve the equations Parallel Materials Modelling Packages @ EPCC 2 / 55 Synopsis Introduction A guided tour inside the “black box” of ab-initio simulation. Synopsis • Motivation • The rise of quantum-mechanical simulations. Some ab initio codes Wavefunction-based theory • Density-functional theory (DFT) Quantum-mechanical • approaches Quantum theory in periodic boundaries • Plane-wave and other basis sets Density Functional • Theory SCF solvers • Molecular Dynamics Electronic Structure of Condensed Phases Recommended Reading and Further Study Total-energy calculations • Basis sets Jorge Kohanoff Electronic Structure Calculations for Solids and Molecules, Plane-waves and Theory and Computational Methods, Cambridge, ISBN-13: 9780521815918 Pseudopotentials • Dominik Marx, J¨urg Hutter Ab Initio Molecular Dynamics: Basic Theory and How to solve the Advanced Methods Cambridge University Press, ISBN: 0521898633 equations • Richard M. Martin Electronic Structure: Basic Theory and Practical Methods: Basic Theory and Practical Density Functional Approaches Vol 1 Cambridge University Press, ISBN: 0521782856
    [Show full text]
  • What's New in Biovia Materials Studio 2020
    WHAT’S NEW IN BIOVIA MATERIALS STUDIO 2020 Datasheet BIOVIA Materials Studio 2020 is the latest release of BIOVIA’s predictive science tools for chemistry and materials science research. Materials Studio empowers researchers to understand the relationships between a material’s mo- lecular or crystal structure and its properties in order to make more informed decisions about materials research and development. More often than not materials performance is influenced by phenomena at multiple scales. Scientists using Materials Studio 2020 have an extensive suite of world class solvers and parameter sets operating from atoms to microscale for simulating more materials and more properties than ever before. Furthermore, the predicted properties can now be used in multi-physics modeling and in systems modeling software such as SIMULIA Abaqus and CATIA Dymola to predict macroscopic behaviors. In this way multiscale simulations can be used to solve some of the toughest pro- blems in materials design and product optimization. BETTER MATERIALS - BETTER BATTERIES Safe, fast charging batteries with high energy density and long life are urgently needed for a host of applications - not least for the electrification of all modes of transportation as an alternative to fossil fuel energy sources. Battery design relies on a complex interplay between thermal, mechanical and chemical processes from the smallest scales of the material (electronic structure) through to the geometry of the battery cell and pack design. Improvements to the component materials used in batteries and capacitors are fundamental to providing the advances in performance needed. Materials Studio provides new functionality to enable the simula- tion of key materials parameters for both liquid electrolytes and electrode components.
    [Show full text]
  • O (N) Methods in Electronic Structure Calculations
    O(N) Methods in electronic structure calculations D R Bowler1;2;3 and T Miyazaki4 1London Centre for Nanotechnology, UCL, 17-19 Gordon St, London WC1H 0AH, UK 2Department of Physics & Astronomy, UCL, Gower St, London WC1E 6BT, UK 3Thomas Young Centre, UCL, Gower St, London WC1E 6BT, UK 4National Institute for Materials Science, 1-2-1 Sengen, Tsukuba, Ibaraki 305-0047, JAPAN E-mail: [email protected] E-mail: [email protected] Abstract. Linear scaling methods, or O(N) methods, have computational and memory requirements which scale linearly with the number of atoms in the system, N, in contrast to standard approaches which scale with the cube of the number of atoms. These methods, which rely on the short-ranged nature of electronic structure, will allow accurate, ab initio simulations of systems of unprecedented size. The theory behind the locality of electronic structure is described and related to physical properties of systems to be modelled, along with a survey of recent developments in real-space methods which are important for efficient use of high performance computers. The linear scaling methods proposed to date can be divided into seven different areas, and the applicability, efficiency and advantages of the methods proposed in these areas is then discussed. The applications of linear scaling methods, as well as the implementations available as computer programs, are considered. Finally, the prospects for and the challenges facing linear scaling methods are discussed. Submitted to: Rep. Prog. Phys. arXiv:1108.5976v5 [cond-mat.mtrl-sci] 3 Nov 2011 O(N) Methods 2 1.
    [Show full text]
  • Porting the DFT Code CASTEP to Gpgpus
    Porting the DFT code CASTEP to GPGPUs Toni Collis [email protected] EPCC, University of Edinburgh CASTEP and GPGPUs Outline • Why are we interested in CASTEP and Density Functional Theory codes. • Brief introduction to CASTEP underlying computational problems. • The OpenACC implementation http://www.nu-fuse.com CASTEP: a DFT code • CASTEP is a commercial and academic software package • Capable of Density Functional Theory (DFT) and plane wave basis set calculations. • Calculates the structure and motions of materials by the use of electronic structure (atom positions are dictated by their electrons). • Modern CASTEP is a re-write of the original serial code, developed by Universities of York, Durham, St. Andrews, Cambridge and Rutherford Labs http://www.nu-fuse.com CASTEP: a DFT code • DFT/ab initio software packages are one of the largest users of HECToR (UK national supercomputing service, based at University of Edinburgh). • Codes such as CASTEP, VASP and CP2K. All involve solving a Hamiltonian to explain the electronic structure. • DFT codes are becoming more complex and with more functionality. http://www.nu-fuse.com HECToR • UK National HPC Service • Currently 30- cabinet Cray XE6 system – 90,112 cores • Each node has – 2×16-core AMD Opterons (2.3GHz Interlagos) – 32 GB memory • Peak of over 800 TF and 90 TB of memory http://www.nu-fuse.com HECToR usage statistics Phase 3 statistics (Nov 2011 - Apr 2013) Ab initio codes (VASP, CP2K, CASTEP, ONETEP, NWChem, Quantum Espresso, GAMESS-US, SIESTA, GAMESS-UK, MOLPRO) GS2NEMO ChemShell 2%2% SENGA2% 3% UM Others 4% 34% MITgcm 4% CASTEP 4% GROMACS 6% DL_POLY CP2K VASP 5% 8% 19% http://www.nu-fuse.com HECToR usage statistics Phase 3 statistics (Nov 2011 - Apr 2013) 35% of the Chemistry software on HECToR is using DFT methods.
    [Show full text]
  • HPC Issues for DFT Calculations
    HPC Issues for DFT Calculations Adrian Jackson EPCC Scientific Simulation • Simulation fast becoming 4 th pillar of science – Observation, Theory, Experimentation, Simulation • Explore universe through simulation rather than experimentation – Test theories – Predict or validate experiments – Simulate “untestable” science • Reproduce “real world” in computers – Generally simplified – Dimensions and timescales restricted – Simulation of scientific problem or environment – Input of real data – Output of simulated data – Parameter space studies – Wide range of approaches http://www.nu-fuse.com Reduce runtime • Serial code optimisations – Reduce runtime through efficiencies – Unlikely to produce required savings • Upgrade hardware – 1965: Moore’s law predicts growth in complexity of processors – Doubling of CPU performance – Performance often improved through on chip parallelism http://www.nu-fuse.com Parallel Background • Why not just make a faster chip? – Theoretical • Physical limitations to size and speed of a single chip • Capacitance increases with complexity • Speed of light, size of atoms, dissipation of heat • The power used by a CPU core is proportional to Clock Frequency x Voltage 2 • Voltage reduction vs Clock speed for power requirements – Voltages become too small for “digital” differences – Practical • Developing new chips is incredibly expensive • Must make maximum use of existing technology http://www.nu-fuse.com Parallel Systems • Different types of parallel systems P M – Shared memory P M P M – Distributed memory P M Interconnect
    [Show full text]
  • Lawrence Berkeley National Laboratory Recent Work
    Lawrence Berkeley National Laboratory Recent Work Title From NWChem to NWChemEx: Evolving with the Computational Chemistry Landscape. Permalink https://escholarship.org/uc/item/4sm897jh Journal Chemical reviews, 121(8) ISSN 0009-2665 Authors Kowalski, Karol Bair, Raymond Bauman, Nicholas P et al. Publication Date 2021-04-01 DOI 10.1021/acs.chemrev.0c00998 Peer reviewed eScholarship.org Powered by the California Digital Library University of California From NWChem to NWChemEx: Evolving with the computational chemistry landscape Karol Kowalski,y Raymond Bair,z Nicholas P. Bauman,y Jeffery S. Boschen,{ Eric J. Bylaska,y Jeff Daily,y Wibe A. de Jong,x Thom Dunning, Jr,y Niranjan Govind,y Robert J. Harrison,k Murat Keçeli,z Kristopher Keipert,? Sriram Krishnamoorthy,y Suraj Kumar,y Erdal Mutlu,y Bruce Palmer,y Ajay Panyala,y Bo Peng,y Ryan M. Richard,{ T. P. Straatsma,# Peter Sushko,y Edward F. Valeev,@ Marat Valiev,y Hubertus J. J. van Dam,4 Jonathan M. Waldrop,{ David B. Williams-Young,x Chao Yang,x Marcin Zalewski,y and Theresa L. Windus*,r yPacific Northwest National Laboratory, Richland, WA 99352 zArgonne National Laboratory, Lemont, IL 60439 {Ames Laboratory, Ames, IA 50011 xLawrence Berkeley National Laboratory, Berkeley, 94720 kInstitute for Advanced Computational Science, Stony Brook University, Stony Brook, NY 11794 ?NVIDIA Inc, previously Argonne National Laboratory, Lemont, IL 60439 #National Center for Computational Sciences, Oak Ridge National Laboratory, Oak Ridge, TN 37831-6373 @Department of Chemistry, Virginia Tech, Blacksburg, VA 24061 4Brookhaven National Laboratory, Upton, NY 11973 rDepartment of Chemistry, Iowa State University and Ames Laboratory, Ames, IA 50011 E-mail: [email protected] 1 Abstract Since the advent of the first computers, chemists have been at the forefront of using computers to understand and solve complex chemical problems.
    [Show full text]
  • Virt&L-Comm.3.2012.1
    Virt&l-Comm.3.2012.1 A MODERN APPROACH TO AB INITIO COMPUTING IN CHEMISTRY, MOLECULAR AND MATERIALS SCIENCE AND TECHNOLOGIES ANTONIO LAGANA’, DEPARTMENT OF CHEMISTRY, UNIVERSITY OF PERUGIA, PERUGIA (IT)* ABSTRACT In this document we examine the present situation of Ab initio computing in Chemistry and Molecular and Materials Science and Technologies applications. To this end we give a short survey of the most popular quantum chemistry and quantum (as well as classical and semiclassical) molecular dynamics programs and packages. We then examine the need to move to higher complexity multiscale computational applications and the related need to adopt for them on the platform side cloud and grid computing. On this ground we examine also the need for reorganizing. The design of a possible roadmap to establishing a Chemistry Virtual Research Community is then sketched and some examples of Chemistry and Molecular and Materials Science and Technologies prototype applications exploiting the synergy between competences and distributed platforms are illustrated for these applications the middleware and work habits into cooperative schemes and virtual research communities (part of the first draft of this paper has been incorporated in the white paper issued by the Computational Chemistry Division of EUCHEMS in August 2012) INTRODUCTION The computational chemistry (CC) community is made of individuals (academics, affiliated to research institutions and operators of chemistry related companies) carrying out computational research in Chemistry, Molecular and Materials Science and Technology (CMMST). It is to a large extent registered into the CC division (DCC) of the European Chemistry and Molecular Science (EUCHEMS) Society and is connected to other chemistry related organizations operating in Chemical Engineering, Biochemistry, Chemometrics, Omics-sciences, Medicinal chemistry, Forensic chemistry, Food chemistry, etc.
    [Show full text]
  • Ab Initio Molecular Dynamics
    AB INITIO MOLECULAR DYNAMICS Iain Bethune ([email protected]) Ab Initio Molecular Dynamics • Background • Review of Classical MD • Essential Quantum Mechanics • Born-Oppenheimer Molecular Dynamics • Basics of Density Functional Theory • Performance Implications Background • Code Usage on ARCHER (2014-15) by CPU Time: Rank Code Node hours Method 1 VASP 5,443,924 DFT 3 CP2K 2,121,237 DFT 4 CASTEP 1,564,080 DFT 9 LAMMPS 887,031 Classical 10 ONETEP 805,014 DFT 12 NAMD 516,851 Classical 20 DL_POLY 245,322 Classical • 52% of all CPU time used by Chemistry / Materials Science / Biomolecular Simulation Image from Karlsruhe Institute of Technology (http://www.hiu.kit.edu/english/104.php) Classical Atomistic Simulation • The main elements of the simulation model are: • Particles v0 r1, m1, q1 • Force field r0, m0, q0 v • Pair potentials 1 r2, m2, q2 v2 • Three-body • Four-body • e.g. CHARMM, GROMOS, AMBER, AMOEBA, ReaxFF … Classical Atomistic Simulation • Molecular Dynamics • Newton’s 2nd Law F = ma • Integrate using e.g. Velocity Verlet algorithm r(t),r(t) → r(t +δt),r(t +δt) • Structural/Geometry Optimisation • Minimise total energy w.r.t. atomic positions Classical Atomistic Simulation • Successes: • Computationally cheap and parallelises well ( > 1,000,000,000 atoms on 10,000 cores) • Able to predict mechanical properties • Density, elasticity, compressability, heat capacity (of insulators) • Can predict structure • RDF of crystals, local ordering in liquids, protein folding … • Failures: • Anything involving electron transfer (i.e. all of Chemistry!) • Bonding, electrochemistry • Heat capacity of metals • Electronic structure/conductivity • Magnetic properties • etc. Essential Quantum Mechanics • We need a model which can describe electrons… • … so turn to Quantum Mechanics – the Physics of the very small.
    [Show full text]