VREs: addressing the Needs of e-Research

1
VREs addressing the Needs of e-Research

• Rob Allan
• CCLRC e-Science Centre
• Daresbury Laboratory

2
The Research Process

• Create or access Data
• Data exists in many forms text, numbers,
  pictures, sound, printed, digital
• Simulations and experimentation or observation
  create new data
• Data is archived/ curated and has associated
  metadata
• Analyse to produce Information
• Definition of information?
• Trends and relationships between data (if I do
  A then B will happen)
• Metadata important
• Synthesize existing and new Knowledge
• Knowledge links concepts
• Based on information (C is related to D in
  some way)
• Knowledge base must be consistent new knowledge
  introduced carefully!
• Science relates knowledge to real world events

3
What if ?

• You could automatically access all of the
  Archived Data Sets and those used in every social
  research publication and decide on the most
  appropriate data for your research needs, without
  having to spend days reading through coding
  schedules and questionnaires
• You could automatically re-estimate all the
  models others have used on these data sets, and
  see what happens if you drop or add new variables
  to the analysis
• You could quickly formulate (check the
  identification etc.) and estimate any new models
  or combinations of existing models you thought
  might be relevant
• You could do this across multiple datasets
• You could match your research questions to
  information held in existing digital resources.
  Search for new explanations
• Integrate multiple sources of data and text to
  help to fill in missing data and ideas.

Example from Social Sciences (Rob Crouchley)
4
Another Use Case

• User logs in to his desktop toolset using single
  sign-on
• Desktop tools must share identity and
  authentication information
• Link into Grid
• Link into digital archives
• Based on survey of literature has an idea that
  an observed phenomenon happens for a certain
  range of a physical parameter
• Need literature cross search facilities google
  style using scientific metadata
• How do we make it easy to ask the right question?
• Chooses a simulation model capable of testing
  this behaviour
• Needs metadata about models accuracy vs.
  parameter range, accuracy vs. performance,
  complexity of model
• Runs a set of simulations over appropriate
  parameter space
• Analyses results Real time? Steering?
  Visualisation?
• Compare with previous work and observation
• XML or other representation to bring result sets
  together

5
Virtual Research Environment

• How can a VRE help e-Research?
• Access everything from an integrated set of
  desktop tools
• At the present time will probably only address a
  subset of the steps identified
• Many tools are currently used in both research
  and support, e.g. MicroSoft Office (Outlook,
  Excel, Word, PowerPoint)
• Many tools used to collaborate and search and
  share data, e.g. Web browsers and servers need
  to aggregate data and information (dont just
  cut and paste)
• e-Science has introduced more collaboration tools
  such as Access Grid and PIG giving audio/ visual
  collaboration for groups
• Peer-to-Peer tools growing in popularity
• Portals popular for CLEs and e-Research (Grid)
  CHEF/ Sakai.
• VRE enables single sign-on and re-use of key
  services.

6
Service Oriented Architecture

• Re-usable distributed services
• Sites can host their own content aggregation/
  watermarking/ protect IPR
• Share metadata and formats cross searching
• Publish services through a variety of interfaces
  as a VRE
• Registries
• Portals
• GUIs
• Programming APIs
• Common services can be used differently depending
  upon application
• Research
• Teaching
• Dissemination
• But, this is only useful if the services are rich
  and access all resources required in the research
  process.

7
Example 1 EGEE and ARDA

• EGEE Enabling Grids for e-Science in Europe.
• ARDA Architectural Roadmap towards Distributed
  Analysis.
• LCH Computing Grid Report CERN-LCG-2003-033
• Builds on existing software - e.g. AliEn portal
  and services
• Assesses future user requirements for LCG
  application area
• Build and extend Grid/ database services
• Provide application frameworks, shells, APIs,
  interactive GUIS, portals etc.
• Includes an object-oriented programming API
• Proposed as an example component of the EGEE work
  programme for the EU Grid.

8
Key Services for Distributed Analysis
9
APIs and User Interfaces
10
Some UK Research Resources

• Existing resources need to be service enabled
• Access Grid Nodes (e-Science Centres)
• Course Content (University and Training
  Institutions)
• Condor pools of workstations (University and
  Teaching institutions)
• Resource Discovery Network resources (JCIE)
• AHDS (AHRB) and e-SS (ESRC) and related training
  and awareness material, e.g. REDRESS
• Directories Z-Directory (UKOLN), Z39-50 target
  directory (Index Data), RSS-express (UKOLN), OAI
  Data providers (OAI), IESR (JISC)
• Text mining service (BBSRC), Data Curation Centre
  and any other specific research resources funded
  in partnership with Research Councils
• Resources referenced in the JISC subject
  resources guides
• Subject gateways Data services Learning and
  teaching
• Support services.
• Tools referenced in JISC Collections publications
  list
• National Grid Service nodes (JCSR).
  Supercomputing facilities such as HPCx, CSAR
  (managed by EPSRC)
• Data Archive and MIMAS (ESRC)
• Protein Data Bank (Hosted by Wellcome Foundation
  at EBI)
• Large-scale facilities such as SRS, ISIS, Diamond
  (hosted at CCLRC) and associated scientific data
  collections
• LHC Data Grid (PPARC)
• NERC Data Centres and CEH

11
Service Classification

• There are many services in common between the
  three JISC pillars. Based on studies in
  previous work (JISC/ CETIS/ ETF) we proposed the
  following classification
• E-Collaboration
• E-Research
• E-Learning
• Digital Information
• Common
• Some of these will be outlined.
• Where some services cross between these classes
  there may still be differences in name and exact
  meaning which have to be resolved.
• We address the e-Research areas by considering
  data, information, knowledge and support.

Click on link
http//www.grids.ac.uk/ETF/public/WebServices/clas
ses.html
12
Data the what

• Archiving
• Cataloguing
• Data Access and Integration
• Data Virtualisation
• Data Replication
• Data Management
• Deposition
• Markup
• Resource Discovery
• Transformation
• Validation

Data and Metadata Services
Metadata is crucial to discovery and quality Who
will curate raw scientific results?
13
Scientific Data Life Cycle
14
Information the how

• Information is also available in many forms,
  often textual, but also audio/ video
• Semantics are important to exchange and compare
• Need agreed terminologies
• How is it derived? Can it be trusted? bias,
  hype
• Privacy and security

• Access
• Aggregation
• Content Registration
• Creation
• Query
• Metadata
• Presentation
• Notification
• Update

Information Services
15
Knowledge the why

• Knowledge is the end result of a long process, so
  far around 4000 years!
• It is highly valued and protected (IPR)
• Passed on in many forms oral, written,
  behavioural
• May be subject to cultural and other
  interpretations
• Different views
• Education process is fundamental to acquiring and
  using knowledge
• Knowledge representation and digital processing
  is relatively new
• Ethical considerations

• Knowledge discovery
• Workflow management
• Knowledge management
• Syndication (join)
• Dictionaries and Ontologies
• Terminology

Knowledge Services
16
The Research Support Processes

• We no longer have research supported solely by
  philanthropists, it's a job for most scientists
  both in academe and industry. It is also
  competitive. There are related and important
  processes
• Write up ideas based on background, reading
  journals, attending conferences
• Submit funding proposal wait for success/
  failure
• Set up project plan risks, tasks, deadlines,
  metrics...
• Link into university finance and people system
• Carry out research based on plan
• Write up results for peers via journals or
  conferences
• Write up results for funding and monitoring
  agencies (e.g. RAE review)
• Publish and disseminate - electronic, printed,
  Web, other?

17
Other Research Services

• Application Management
• Business Workflow
• Deployment
• Distribution
• Fabric Management
• Grid Information
• Job Management
• Process Building
• Proposal Writing
• Resource Discovery
• Resource Management
• Scheduling
• Security
• Validation and Verification
• Visualisation

This area includes active services for control
of computation, observation and experimental
resources (Grid).
18
Service Framework for e-Research
From S. Wilson, K. Blinco and D. Rehak Service
Oriented Frameworks DEST (Australia), JISC-CETIS
(UK) and Industry Canada http//www.jisc.ac.uk/upl
oaded_documents/AltilabServiceOrientedFrameworks.p
df
19
Example 2 Integrative Biology

• Large project funded by EPSRC with international
  partners. Actively defining and seeking
  middleware to implement shared services.
• Multi length/ time scale modelling of heart and
  cancer growth
• Linking multiple components across 3 continents.
• Developing services in the following areas
• User management
• Executable building
• Executable management
• Data management
• Job management
• Workflow management
• Visualisation and interactive services
• Steering
• Collaborative working

20
IB Architecture
21
Services and Aggregation
Web is ubiquitous to both research and leisure.
Has Google replaced the traditional library? I
dont need to leave my desk and can download
plenty of material.
22
VRE access to many Services
23
(No Transcript)