Community grid infrastructures for geosciences and materials modelling

1
Community grid infrastructures for geosciences
and materials modelling Martin Dove University
of Cambridge and National Institute for
Environmental eScience
2
Hello and thank you NIEeS
National Institute for Environmental eScience
Partnership between

• aiming to support the development of escience
  work within the UK environmental science
  community
• centre for national capability for escience
  within the UK environmental sciences community

3
Hello and thank you NIEeS
National Institute for Environmental eScience

• Stuart Ballard
• Ian Frame
• Gen-Tao Chiang
• Therese Williams

4
Hello and thank you NIEeS
National Institute for Environmental eScience

• Training events (eg Google maps, XML from
  Fortran)
• Source of expertise, capability information
• Grid-enabling geoscience and environmental
  sciences applications

5
New initiatives eGY
We are one of the UK representatives for the
Electronic Geophysical Year 20078
We can achieve a major step forward in
geoscience capability, knowledge, and usage
throughout the world for the benefit of humanity
by accelerating the adoption of modern and
visionary practices for managing and sharing data
and information.
6
Hello and thank you eMinerals
eMinerals project
NERC-funded escience project involving Cambridge,
Bath, UCL, Royal Institution, CCLRC Daresbury,
Birkbeck

• Developing the concept of community grids to
  support collaborative working in molecular-scale
  simulations
• Emphasis on grid computing, data management,
  information delivery and collaborative working
• Core applications in understanding pollutants in
  the environment at the atomic scale

7
Hello and thank you eMinerals

• Toby White, Kat Austen, Andrew Walker, Emilio
  Artacho, Peter Murray-Rust (Cambridge)
• Rik Tyer, Kerstin Kleese (CCLRC)
• Steve Parker, Arnaud Marmier, Corinne Arrouvel
  (Bath)
• Ismael Bhana (Reading)

8
Scales of geosciences and environmental sciences
9
Our view of eScience

• eScience refers to new science opportunities that
  require distributed collaborations and are
  enabled by emerging internet technologies.
• These technologies include grid computing,
  distributed data management and collaborative
  tools.
• Many tools are still in the process of rapid
  development, and in some cases standards are not
  yet established.

10
Our view of eScience
11
Implicit in the grid area
12
Example adsorption of dioxin molecules on clay
surfaces

Combinatorial study requiring grid computing,
data management and collaborative tools
13
eScience Science beyond the lab book

• Management of many tasks
• Management of the resultant data deluge
• Sharing the information content with collaborators

Stretching science beyond human limitations
whilst maintaining accuracy and accountability
14
eMinerals science Molecular-scale environmental
issues
Radioactive waste disposal
Pollution molecules and atoms on mineral surfaces
Crystal dissolution and weathering
Crystal growth and scale inhibition
15
eMinerals science Molecular-scale environmental
issues
Radioactive waste disposal
Pollution molecules and atoms on mineral surfaces
Crystal dissolution and weathering
Crystal growth and scale inhibition
16
Using the Virtual Organisation model in
environmental science

• A comprehensive assault on the issue of transport
  of pollutants in the environment
• Heavy metal poisonous waste
• Toxic organic molecules
• Nuclear waste encapsulation

17
Collaborative science and the eMinerals Virtual
Organisation
18
Collaborative science and the eMinerals Virtual
Organisation
19
So what does our Virtual Organisation need?

• Computing grid infrastructure for running
  large-scale simulation studies
• Easy-to-use tools for managing large-scale
  combinatorial computational studies
• Ability to share data between collaborators
• Ability to extract and share information content
• Grid-enabling of simulation codes
• Means to communicate effectively (NOT email!)

20
Components of the compute grid

• Authentication authorisation, and job
  submission, handled by Globus

21
Researcher
22
Data grid the San Diego Storage Resource Broker
Distributed file management
Distributed data vaults
23
SRB client tools

• Unix Scommands (eg Sput, Sls, Scd, Sget)
• Web interface
• GUI for MS Windows (InQ)

24
Running jobs on our minigrid Job workflow

• The scientist places input files and application
  executables into the SRB
• The scientist submits the job to a grid resource
• The job downloads the files from the SRB
• The calculations are performed
• The job places all output files into the SRB
• Metadata and core information are collected from
  the output files
• The scientist retrieves data files from the SRB
  and/or core information from the metadata store

25
Researcher
4. Job runs on grid compute resources
Application server
26
User interface my_condor_submit tool
Executable ossia2004 pathToExe
/home/rty.eminerals/OSSIA2004preferredMachin
eList lv1.nw-grid.ac.uk-serial
dl1.nw-grid.ac.uk-serial jobType
performance numOfProcs 1 Output
trans.out Sdir
/home/mdv.eminerals/RMCSdemo Sget
Sput GetEnvMetadata
true RDesc Test sweep of temperature
using ossia RDatasetID 263 AgentX
Temperature,trans.xml/ParameterListtitle'Init
ial System'/Parametername'Temperature' AgentX
Energy,trans.xml/PropertyListlast/Pr
opertytitle'Energy' AgentX
OrderParameter,trans.xml/Modulelast/Propertyti
tle'Order parameter' AgentX
HeatCapacity,trans.xml/Modulelast/Propertytitl
e'Heat capacity' AgentX
Susceptibility,trans.xml/Modulelast/Propertyti
tle'Susceptibility'
27
User profile

• Users do not want portals portals are for tools,
  not for the working environment
• Users do not want their applications pre-wrapped
  as services they want to have complete control
  over their applications, e.g. to add capability
• Users do not want a provider/consumer model that
  does not provide the freedom they need

28
Data sharing the need for information delivery
tools
Classical molecular dynamics methods
Quantum mechanical methods
29
Collaborative grid Data and information sharing
?
30
Data and information sharing XML data
representation
lt?xml version"1.0" encoding"UTF-8"?gt ltcml
convention"FoX_wcml-2.0" fileId"cis1.cml"
version"2.4" xmlns"http//www.xml-cml.org/schema
"gt ltmetadataList name"Metadata"gt ltmetadata
name"LeadProgramAuthor" content"Martin
Dove"/gt ltmetadata name"Code name"
content"ossia"/gt ... lt/metadataListgt ltmo
dule title"Initial System" dictRef"emininitialM
odule"gt ltparameterListgt ltparameter
dictRef"ossiatemperature" name"Temperature"gt lts
calar dataType"xsddouble units"cmlUnitseV"gt1.
000000000000e-1lt/scalargt lt/parametergt ltparam
eter dictRef"ossiaNumberOfSteps" name"Number
of steps"gt ltscalar dataType"xsdinteger"
units"unitscountable"gt10000000lt/scalargt lt/parame
tergt ... lt/parameterListgt lt/modulegt
... ltmodule title"Finalization"
dictRef"eminfinalModule"gt ltpropertyListgt ltproper
ty dictRef"ossiaEnergy" title"Energy"gt ltscalar
dataType"xsddouble" units"cmlUnitseV"gt2.052516
362912e-1lt/scalargt lt/propertygt
... lt/propertyListgt lt/modulegt lt/cmlgt
Chemical Markup Language
Capturing metadata
Capturing initial parameters
Capturing computed properties
31
What XML gives us

• Simulation code output that is self-describing
  (no more mere lists of numbers!)
• XML files can be transformed to give user-centric
  and information-centric representations of data,
  including plotted data
• XML files can have key information extracted
  easily, essential for large combinatorial studies
• XML enables automatic capture of metadata, and
  metadata is essential for managing data

32
XML ? metadata

• Our job submission tools automatically harvest
  metadata from our output XML files
• We have developed a new set of tools to access
  the metadata database (RCommands)

• We use metadata for locating data and datasets
  created by our colleagues
• We also use metadata for extracting core
  information from data  useful for analysing
  combinatorial studies

33
Collaborative grids
Classical molecular dynamics methods
Quantum mechanical methods
34
Researcher A
Web 2.0
Researcher B
35
Example Compressibility of amorphous silica

• Density is not quite linear note that the
  gradient is larger in the middle of the plot than
  at either end.
• Bulk modulus (BM)
• BM has minimum around 2 GPa compressibility
  1/B has maximum

36
Example Compressibility of amorphous silica
Molecular dynamics simulations of
pressure-dependence of amorphous silica Volume
curve shows that silica gets softer around 2
GPa Negative derivative defines the
compressibility
37
The message from this example

• We had to run over 600 sets of simulations and
  analyses ...
• ... each generating around 2 GB data
• We used grid computing to run the simulations
  using our tools, and our data management system
  to extract and share the key data
• And part of this was run by a third-year project
  student, Lucy

38
Example Dioxin molecules on layer silicate
neutral surfaces

39
Range of molecule/substrate systems
Calculating how the binding energy varies across
the molecular series from all chlorine to no
chlorine
40
Summary

• The tools I have discussed can be used for many
  different application domains within the
  geosciences and environmental sciences

• Emphasis on all core areas compute, data and
  collaborative grids