APAC Initiatives for Large-Scale Data Sets and Grid Computing - PowerPoint PPT Presentation

1 / 16
About This Presentation
Title:

APAC Initiatives for Large-Scale Data Sets and Grid Computing

Description:

Universities of Sydney and ... IAU demo of Data Grid and Visualisation testbed ... TPAC (Uni of Tasmania..), Bureau of Meteorology, University of Reading ... – PowerPoint PPT presentation

Number of Views:38
Avg rating:3.0/5.0
Slides: 17
Provided by: johno163
Category:

less

Transcript and Presenter's Notes

Title: APAC Initiatives for Large-Scale Data Sets and Grid Computing


1
APAC InitiativesforLarge-Scale Data Sets and
Grid Computing
  • Robin Stanton
  • Bernard Pailthorpe
  • Australian Partnership for Advanced Computing

Presentation to NeSC 28 May 2003
2
Topics
  • Infrastructure for eResearch
  • Australian Partnership for Advanced Computing
  • APAC
  • GrangeNet
  • Grid projects at ANU supported by APAC
  • APAC Initiatives

3
Changing How Science is Done
  • Collect data from digital libraries, laboratories
    and observation
  • Analyze data with models run on the Grid
  • Visualize and share data over the Web
  • Publish results in a Digital Library

From Sid Karin, SDSC/NPACI
4
Grid Services for eResearch
User Communities
Bio-informatics
Astronomy
- - - - - - - - - - - - - - - - - - - - - - - -
- - - - - - - - - - - - - - - - - - - - - - - - -
- - - - - - -
Physics
Environment
Collaborative Visualisation
Distributed Computing
Information Access
Cooperative Environments
On-Line Instruments
Web Services Advanced Communications Services
5
APAC Achievements
  • The APAC partnership formed June 2000
  • A partner in each State as well as ANU and CSIRO
  • The APAC National Facility operational April
    2001.
  • APAC and partner facilities serviced over 1,100
    users.
  • Over 110 projects supporting users and developing
    expertise in 13 computational science and
    engineering themes.
  • Over 50 university courses prepared and delivered
    in computational science and engineering.

6
APAC National Facility
  • Computing Systems
  • HP AlphaServer SC ES45 (127 nodes)
  • ranked number 63 in latest TOP500 list
  • Dell Linux cluster, HP Marvel
  • Mass Storage
  • Storagetek (robotic silo) tape library
  • Capable of a petabyte (1015 bytes) of storage
  • Visualisation
  • visualisation virtual reality systems
  • Staff
  • staff at the Australian National University (ANU)

http//nf.apac.edu.au
7
GrangeNetA GRid And Next GEneration
Networkwww.grangenet.net
Supported by the Federal Governments BITS
Advanced Networks Program
8
APAC Partners and Backbone Networks
Darwin
Brisbane QPSF
USA
Canberra ANU
Perth IVEC CSIRO
Sydney ac3
Adelaide SAPAC
APAC National Facility
Melbourne VPAC CSIRO
GrangeNet Backbone AARNet Links
Hobart TPAC CSIRO
9
Gravitational Wave Astronomy
  • GWA involves exchange and simultaneous data
    processing between multiple detectors
  • Gravity wave detectors environmental monitoring
  • ACIGA Australia
  • LIGO USA VIRGO,GEO Europe TAMA Japan
  • Technical collaborations with GriPhyN and iVDGL
  • Operational data-pipeline centred on APAC MDSS
  • Upgrading to Lightweight Data Replicator (LDR)

Australian Consortium for Interferometric
Gravitational Astronomy
10
ACIGA Data Grid
Australian Consortium for Interferometric
Gravitational Astronomy
Rsync/LDR
GridFTP
ACIGA APAC resources
Environmental Monitors
11
High-Energy Particle Physics
  • Belle Physics Collaboration
  • K.E.K. B-factory detector, Tsukuba, Japan
  • Matter/Anti-matter investigations, Atlas
    test-run
  • 45 Institutions, 400 users worldwide
  • 10 TB data currently
  • Universities of Sydney and Melbourne active
    participants
  • Australian collaborators leading Grid adoption
  • Australian Data-grid centred on APAC MDSS
  • Exploiting Globus 2.x, Gfarm
  • Atlas Experiment
  • Large Hadron Collider (LHC) at CERN
  • Collaboration 2000 people, 150 institutes
    internationally, 34 countries
  • 3.5 PB data per year
  • operational in 2007

12
Virtual Observatories
  • MACHO Project Data
  • Largest online astrophysical data set in
    Australia
  • 10TB Data collected over 10 years
  • Hosted on APAC MDSS
  • Web interface at wwwmacho.anu.edu.au
  • Currently using Z39.50 metadata standard
  • Mapping metadata to VOTable 1.0 standard
  • Emerging IVO metadata standard
  • International Virtual Observatory
  • MACHO data being incorporated into SDSC SRB
    system
  • www.ivoa.net
  • Australian Virtual Observatory
  • IAU demo of Data Grid and Visualisation testbed
  • Distributed data-sets, Tomcat rendering software
  • www.atnf.csiro.au/projects/avo/

13
Bioinformatics
  • Many initiatives to support bio-community
  • Bio-mirror supported by AARNet and ANU
  • www.bio-mirror.net
  • Bio-database search by VPAC and Ausbiotech
  • www.ausbioinfo.com
  • Australian National Genomic Information Service
    (ANGIS)
  • www.angis.org
  • ARC Centre for Bioinformatics (M Ragan)
    (www.imb.uq.edu.au)
  • Plan to coordinate infrastructure
  • Replicate access mechanisms to data sets
  • Provide common Web interfaces to applications
  • Provide specialised systems (Gaussian, Blast..)

14
Earth Observation
  • GADS Grid Access Data Service
  • for oceanographic and climate data
  • World Ocean Circulation Experiment (WOCE)
  • project funded by APAC
  • TPAC (Uni of Tasmania..), Bureau of Meteorology,
    University of Reading
  • Grid access via Web services
  • Interface to DODS/OPeNDAP
  • Used by Earth Systems Grid and NERC DataGrid

15
Cultural Language Archives
  • PARADISEC
  • Pacific and Regional Archive for Digital Sources
    in Endangered Culture
  • Uni of Sydney, Uni of Melbourne, ANU
  • Digitised oral and music recordings from
    Asia-Pacific region
  • Integrate with sociological data sets
  • International archival standard for digital audio
  • 24bit 96KHz Stereo metadata
  • APAC MDSS to host 10,000 hours
  • 2GB/Hr gt 20TB total

16
APAC Initiatives
  • Provide more support for data-intensive
    computing
  • APAC support for large-scale data sets
  • ask research organisations for proposals to have
    large-scale data sets managed by APAC and its
    partners
  • concentrate on national and international data
    access
  • develop plans for managing these data sets
  • Install and operate an APAC Grid
  • Consider a pilot project for eResearch
  • support a research community through data
    management
Write a Comment
User Comments (0)
About PowerShow.com