Grid based molecular simulators
This presentation is the property of its rightful owner.
Sponsored Links
1 / 37

Grid based Molecular Simulators PowerPoint PPT Presentation


  • 62 Views
  • Uploaded on
  • Presentation posted in: General

EGEE Generic Applications Advisory Panel first meeting. Grid based Molecular Simulators. CERN, Geneva, June 14 th , 2004. Osvaldo Gervasi Dept. of Mathematics and Computer Science University of Perugia [email protected] Antonio Laganà Dept. of Chemistry University of Perugia

Download Presentation

Grid based Molecular Simulators

An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.


- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -

Presentation Transcript


Grid based molecular simulators

EGEE Generic Applications Advisory Panelfirst meeting

Grid based Molecular Simulators

CERN, Geneva, June 14th, 2004

Osvaldo Gervasi

Dept. of Mathematics and

Computer Science

University of Perugia

[email protected]

Antonio Laganà

Dept. of Chemistry

University of Perugia

[email protected]


Grid based molecular simulators

Summary

  • COST Chemistry and METACHEM

  • Ongoing metalaboratories in Chemistry

  • Architecture of the Simulator

  • Conclusions


Cost countries 2002 participating institutions

Participation of Non-COST Institutions (Total: 78)

  • Ukraine (8)

  • USA (16)

  • Russia (22)

  • Algeria (1)

  • Armenia (2)

  • Argentina (1)

  • Australia (3)

  • NGO (3)

  • Canada (13)

  • Japan (6)

  • China (2)

  • India (1)

  • Eritrea (1)

COST- Countries (2002) Participating Institutions

  • EU COST States (2078)

  • Turkey (20)

  • Non COST (78)

  • Switzerland (143)

  • Israel (27)

  • EU Framework Programme

  • FYROM (4)

associated states (902)

  • F.R.Y. (6)

  • Austria (152)

  • Slovenia (100)

  • Belgium (165)

  • Other COST States (211)

  • Croatia (38)

  • Denmark (133)

  • Slovakia (48)

  • Cooperating States (27)

  • Romania (67)

  • Finland (137)

  • Poland (105)

  • Participation of non-COST institutions

    (Details on following Chart)

  • Norway (124)

  • France (174)

  • Malta (12)

  • Lithuania (42)

  • Germany (179)

  • Latvia (37)

  • Iceland (18)

  • Greece (113)

  • Hungary (134)

  • Ireland (102)

  • Estonia (10)

  • Czech Republic (104)

  • Italy (173)

  • Cyprus (52)

  • Luxembourg (6)

  • Bulgaria (49)

  • Netherlands (150)

  • Spain (184)

  • United Kingdom (180)

  • Portugal (96)

  • Sweden (134)


Grid based molecular simulators

COST Member States

  • The twentyfive EU Member States

  • EFTA Member States

    • Iceland

    • Norway

    • Switzerland*

  • Candidate Counties

    • Bulgaria

    • Romania

    • Turkey *

  • Other Countries

    • Federal Republic of Yugoslavia*

    • Former Yugoslav Republic of Macedonia*

    • Croatia *

  • Co-operating State

    • Israel

* Not Associated to FP


The cost chemistry domain

The COST Chemistry domain

  • Created in 1992

  • 34 Actions launched since 1992 (12 completed with final evaluation)

  • 22 running (10 had the mid term evaluation)

  • Research Chemists from 31 COST Countries

  • Associated Institutions from Australia, Japan, Russia, Ukraine and the USA


Grid based molecular simulators

Second batch of Actions

  • Health and Therapy

    - D13 (New Molecules for Human Health),

    - D18 (Lanthanide Chemistry for Diagnosis and Therapy),

    - D20 (Metal Compounds in the Treatment of Cancer and Viral Diseases)

  • Biology and Prebiotic

    - D21 (Metalloenzymes and Chemical Biomimethics),

    - D22 (Protein Lipid Interaction),

    - D27 (Prebiotic Chemistry and Early Evolution),

    - D28 (Natural Products as a Source for Discovery, Synthesis and Application of Pharmaceuticals)


Second batch of actions

Second batch of Actions

  • Nanostructures and new Functions and Materials

    - D14 (Functional Molecular Material),

    - D17 (Oligomers, Polymers and Copolymers via Metal Catalysis),

    - D31 (Organising Non-Covalent Chemical Systems with Selected Functions),

    - D34 (Nanoscale Electrochemical and Bioprocesses at solid aqueous interfaces of industrial materials)

  • New Molecules and Clean Processes

    - D15 (Interfacial Chemistry and Catalysis),

    - D24 (Sustainable Chemical Processes Stereoselective Transition Metal Catalysed Reactions),

    - D25 (Applied Biocatalysis: Stereoselective and Environmentally Friendly Reactions catalysed by Enzymes)

    - D29 (Sustainable Chemical Processes: Stereoselective Transition Metal-Catalysed Reactions)


Second batch of actions1

Second batch of Actions

  • High Pressure and Energy

    - D30 (High Pressure Tuning of Chemical and Biochemical Processes),

    - D32 (Chemistry in High Energy Microenvironments)

  • Theory and computing

    - D16 (Combinatorial Chemistry),

    - D23 (Metachem: Metalaboratories for Complex Computational Applications in Chemistry),

    - D26 (Integrative Computational Chemistry)


Metachem matalaboratories for complex applications in chemistry

Metachem: Matalaboratories for Complex Applications in Chemistry

  • MURQM: Multireference Quantum Chemical Methods

  • DIRAC: Four Component Relativistic Quantum Chemical Calculations

  • SIMBEX: Simulation of Molecular Beam Experiments

  • DYSTS: Dynamics and Spectroscopy of Systems : Relevant to Environment and Applied Chemistry

  • ELCHEM: E-learning Technologies for Chemistry

  • ICAB: Integration of Codes for Ab Initio Methods

  • COMOVIT: Collaborative Molecular and Electronic Structure Visualization tools


Eu partners of d23 wgs

EU partners of D23 WGs

Simbex

Murqm

Dirac

Elchem

Icab

Dysts

Comovit


Labs per nationality

Labs per nationality

1 Isr,Pl,Sk,Nl

2 Cz,Ch, Fr, Dk, A, Sw, No

3 Hu

4 Gr

5 E

6 D, Uk,

9 I


Murqm p carsky cz

MURQM (P. CARSKY, CZ)

  • Ab initio codes are designed in a cooperative way. They deal with large matrices and linear algebra operations. Diagonalizations and minimizations. Construction of hypersurfaces of potential energy values bypoints

  • 9 Laboratories (Petr Cársky - J. Heyrovsky Institute - Czech Republic, Jiri Pittner - J. Heyrovsky Institute - Czech Republic, Ivan Hubac - Comenius University – Slovakia, Stephen Wilson - Rutherford Appleton Laboratory –UK, Wolfgang Wenzel Universität Dortmund, Germany, Leszek Meissner - Nicholas Copernicus University – Poland, Volker Staemmler - Ruhr Universität – Germany, Constantinos Tsipis - Aristotle University of Thessaloniki – Greece, Aristides Mavridis - National and Kapodistrian Univ. of Athens - Greece)


Dirac k faegri no

DIRAC (K. FAEGRI, NO)

  • Cooperative development of MC-SCF, gradient minimization, DFT capabilities, integral algorithms for relativistic accurate concurrent calculations

  • 6 Laboratories (Knut Faegri - University of Oslo – Norway, Hans J. Aagaard Jensen University of Southern Denmark – Denmark, Uzi Kaldor -Tel Aviv University – Israel, Patrick Norman - Linköping University – Sweden, Trond Saue - Université Louis Pasteur – France, Lucas Visscher - Vrije Universiteit Amsterdam – Netherlands, Timo FleigHeinrich - Heine University - Germany)


Simbex o gervasi i

SIMBEX (O. GERVASI, I)

  • Coordinated implementation of a workflow management environment to simulate molecular beam experiments and molecular processes. Ab initio, dynamics, kinetics and statistics codes are assembled

  • 10 Laboratories (Osvaldo Gervasi - Università di Perugia – Italy, Ernesto Garcia - Universidad del Pais Vasco – Spain, Gabriel Balint-Kurti - University of Bristol – UK, Gunnar Nyman - University of Goteborg – Sweden, Peter Kacsuk - MTA SZTAKI – Hungary, Jaroslaw Nabrzyski Inst. Bioorganic Chem., Poland, Francisco Tirado - Universidad Complutense – Spain, Ranieri Baraglia – CNR – Italy, Robert J. Allan – Daresbury Laboratory – UK, Gyorgy Lendvay – HAS -Hungary)


Dysts a aguilar e

DYSTS (A. AGUILAR, E)

  • Environment and application driven dynamics and spectroscopy calculations distributed among the participating laboratories

  • 4 Laboratories (Antonio Aguilar-Navarro - Universidad de Barcelona – Spain, Vincenzo Aquilanti - Università di Perugia – Italy, Florent Xavier Gadea - Université Paul Sabatier – France, Paolo Palmieri - Universita di Bologna - Italy)


Elchem a lagana i

ELCHEM (A. LAGANA, I)

  • Coordinated development of ubiquitous learning technologies and virtual laboratories. Design of distributed virtual reality tools at human and molecular level

  • 10 Laboratories (Antonio Laganà - Università di Perugia – Italy, Ernst Bratz - TU Muenchen – Germany, Gustavo Avitabile - Università Federico II – Italy, Dimitra Kovala-Demertzi - University of Ioannina – Greece, Jens Josephsen - Roskilde University – Denmark, Paul Yates - Keele University – UK, Anna Croft - University of Wales – UK, Evangelia Varella - University of Thessaloniki – Greece)


Icab e rossi i

ICAB (E. ROSSI, I)

  • Coordinated development of linear scaling methods for ab initio calculations with particular emphasis on chemical data transfer and handling

  • 6 Laboratories (Elda Rossi – CINECA –Italy, Renzo Cimiraglia - Università di Ferrara – Italy, Daniel Maynau – IRSAMC – France, José Sanchez-Marin - Universitat de Valencia – Spain, Peter Szalay - Eötvös Loránd University – Hungary, Rosa Caballol - Universitat Rovira i Virgili Spain)


Comovit h luethi ch

COMOVIT (H. LUETHI, CH)

  • Design and development of multimedia distributed software to handle 3D molecular information of ab initio origin

  • 6 Laboratories (Hans Peter Luethi, ETH - Zurich - CH)


The architecture of the simulator

SURFACE

Construction of the

Potential Energy Surface

PROPERTIES

DYNAMICS

Dynamical properties

Calculation

Calculation of

Averaged quantities

Good

Results?

no

yes

end

The architecture of the Simulator


Electronic structure calculation and representation

Electronic Structure calculation and representation

  • Isolated (gas phase) small molecules

  • Isolated (gas phase) large molecules

  • Condensed phase and solid state calculations

  • Topological studies

  • Modeling and functional representations of the potential energy surfaces


Molecular dynamics calculations

Molecular Dynamics calculations

  • Exact quantum dynamics for small systems

  • Semiclassical and mixed classical-quantum for intermediate systems

  • QM/MM and Car Parrinello

  • Classical dynamics


Observable properties

Observable Properties

  • Structure and stability calculations for aggregates of various sizes

  • Kinetics and fluid dynamics calculations

  • Thermodynamics properties

  • Direct Monte Carlo calculations

  • Condensed phase and liquid crystals

  • Cross sections and rate coefficients


Electronic structure programs

Electronic Structure Programs

  • Small molecules (GAMESS-UK, GAMESS US, MOLPRO)

  • Large molecules (GAUSSIAN03)

  • Topological analysis of the interactions (AIMPAC, TOPOND)

  • Modelling and fitting of the potential energy surfaces (FITTING)


Molecular dynamics programs

Molecular Dynamics Programs

  • Car-Parrinello (CPMD)

  • Classical dynamics (ABCtraj, VENUS96; DL_POLY)

  • Quantum dynamics (TD)

  • Semiclassical dynamics (ABCsem)


Observable properties programs

Observable Properties Programs

  • Direct Monte Carlo (DSMC)

  • Energy and angular distributions


Critical features of the individual programs

Critical Features of the Individual Programs

  • AB INITIO METHODS (molpro, gamess, adc, gaussian, ) resource requests are proportional to N3(N is the number of electrons) and to MD (M is the number of grid points per dimension D) for CPU and disc demand.

  • EMPIRICAL FORCE FIELDS (Venus, dl_poly, …) resource requests are proportional to P! (P is the number of atoms)


Grid based molecular simulators

Critical Features of the Individual Programs (ii)

  • DYNAMICS (APH3D, TIMEDEP, …) these programs use as input the output of the previous module most critical dependence is on the total angular momentumJ value that can increase up to several hundred units and the size of the matrices depend on 2J+1

  • KINETICS PROGRAMS use dynamics results for integrating relevant time dependent applications


The model architecture of a simulator

SimulatorPrograms

If VMs

involved

Java Applets

SimGate

Grid infrastructure

Grid middleware

The model architecture of a Simulator

Ab-initio techniques,

PES fitting,

Molecular Dynamics

Net Sockets

Net Sockets

Client side

Polling

HTTP Server side


Local platforms

Local Platforms

  • Several workstations (IBM, Sun, SG, HP, ..)

  • Clusters of PC

  • Parallel machines (IBM SP, Origin, Sun multiprocessor, …)

  • Supercomputer centers (CINECA, EPCC, CESCA, ..)


Metalaboratories components

Metalaboratories components

  • Molecular expertise centers

  • Specialized software fabric (programs, interfaces, ..)

  • Grid middleware

  • Grid infrastructure

  • Users


Grid based molecular simulators

Virtual Monitors of theAngular Distribution of H + ICl

ABCtraj:

QCT for 3 body systems


Oh hcl trajectory

OH-HCl trajectory

Venus:

QCT for many body systems


Grid based molecular simulators

Virtual Monitor of Ar+C6H6

DL-POLY:

QCT for complex systems


Conclusions

Conclusions

  • A proposal to deploy a Chemistry Grid Application aimed at building molecular simulators is presented

  • It is based on the activities of some already established Metalaboratories in D23 COST Action, some of them already involved in EGEE activities.

  • It will impact all research activities in which complex simulations need to be carried out to govern advanced experimentsand a priori rationalizations of real systems.


Surface

START

Are

Ab initio

computations

feasible?

Application taking

empirical data

from Databases

to generate a PES

NO

NO

SURFACE

YES

YES

Application using

Ab Initio programs

for electronic

structure

Application using

Fitting programs

to generate a PES

DYNAMICS

SURFACE

Construction of the interaction

Are

Ab initio

computations

available?


Dynamics

integration of scattering equations

SURFACE

Are

Quantum

Dynamics

Calculations

unfeasible?

Application using Time-Independent Quantum Techniques

NO

NO

Single Initial

State?

DYNAMICS

YES

YES

Application using

Time-Dependent

Quantum

Techniques

Application using

Classical Dynamics

(trajectory)

Techniques

PROPERTIES

DYNAMICS


Properties

Return

to

START

DYNAMICS

NO

NO

Rate

Coefficients

Virtual Monitor

YES

YES

Vibrational, Rotational, Angular Distribution Virtual Monitors

Cross Section

Virtual Monitor

Do

measured

and calculated

properties

agree ?

NO

YES

Save the data of

Virtual Monitors

STOP

PROPERTIES

Reconstruction of reaction properties

State to State

observable

quantities?

State Specific

observable

quantities?

PROPERTIES


  • Login