Grids and eScience

1
Grids and eScience

• Mark Hayes
• Technical Director - Cambridge eScience Centre

CeSC Computer Officers Forum - 12th January 2004
2
eScience - a definition
eScience is about global collaboration in key
areas of science and the next generation of
infrastructure that will enable it.
Dr.John Taylor, Director General of the Research
Councils
3
Grids are not about...

• the next generation Internet
• a brand-new high bandwidth network
• free-for-all supercomputing-on-tap
• (c.f. Clay Shirky http//shirky.com/writings/g
  rids.html )
• It might be... the next step in the long
  history of distributed computing.
• (some well-known examples NFS, X11, WWW)

4
Save the world while you sleep
1.2 million CPU years so far...
Computational drug design
Protein folding
5
Its not just compute cycles...
An exponential growth in data from many areas of
science.
6

The data explosion - some big numbers

• CFD turbulence simulations - 100TB
• BaBar particle physics experiment - 1TB/day
• CERN LHC will generate 1GB/s or 10PB/year
• VLBA radio telescope generates 1GB/s today
• NCBI/EMBL database is only 0.5TB but doubling
  each year
• brain imaging - 4TB/brain at full colour, 10mm
  resolution
• (4PB/brain at 1mm i.e. cellular resolution)
• Pixar - 100TB/movie

FTP and GREP are not adequate (Jim Gray)
7

• A typical Grid application

BAE
Cambridge
Portal
HPCF
Reflection data
Visualisation
CAD Design
8

• The Virtual Organization

My definition a group of computers, data and
people distributed across institutional
boundaries, who wish to share their resources in
pursuit of a common goal (curing cancer, finding
the Higgs boson, designing a better
aircraft) For the official definition, see the
papers by Foster et alThe Anatomy of the
Grid The Physiology of the Grid http//www.glob
us.org/research/papers.html For this you need
cross-institutional trust and security
mechanisms.
9

• The Grid in Cambridge

• CeSC - one of 10 regional eScience centres
  around the country.
• Were here to provide advice support to the
  University (and
• its neighbouring institutions) in the use of
  Grid technology.
• NIEeS - national centre for Grid training and
  research in
• the environmental sciences - http//www.niees.ac
  .uk/
• GridPP - particle physicists getting ready for
  the Large
• Hadron Collider (2007?) - presence at the
  Cavendish
• AstroGrid - the virtual observatory - presence
  at IoA
• eMinerals - modelling the atomistic processes
  involved in
• environmental issues - presence at Earth
  Sciences

10

• The Grid in Cambridge

• Cam-Grid linking departmental clusters
• initially Physics, CeSC Earth Sciences
• spare CPU cycles on PWF managed clusters?
• Cambridge Computational Biology Institute
• will foster collaborative research across
  medicine, biology, mathematics
• the physical sciences. CeSC is helping to
  assess how this will effect the
• use of IT across participating departments.
• Biologists now need the Grid and HPC too!

11
The Grid in the UK
Pilot projects in particle physics, astronomy,
medicine, bioinformatics, environmental
sciences...
Contributing to international Grid software
development efforts
10 regional eScience Centres
12
Some UK Grid resources

• Daresbury - loki - 64 proc Alpha cluster
• Manchester - green - 512 proc SGI Origin 3800
• Imperial - saturn - large SMP Sun
• Southampton - iridis - 400 proc.Intel Linux
  cluster
• Rutherford Appleton Lab - hrothgar - 32 proc
  Intel Linux
• Cambridge - herschel - 32 proc Intel Linux
  cluster
• ...
• coming soon 4x gt64 CPU JISC clusters, HPC(X)

13
Applications on the UK Grid
Ion diffusion through radiation damaged crystal
structures (Mark Calleja, Earth Sciences,
Cambridge)

• Monte Carlo simulation lots of independent runs
• small input output
• more CPU -gt higher temperatures, better stats
• access to 100 CPUs on the UK Grid
• Condor-G client tool for farming out jobs

14
Applications on the UK Grid
Reality Grid (Peter Coveney, UCL)

• Fluid dynamics of complex mixtures, e.g
• oil, water and solid particles (mud)
• Used CPU at Manchester, London, Cambridge...
• Remote visualisation using SGI Onyx in
  Manchester
• (from a laptop in Sheffield)
• Recently won the award for the most innovative
  
• data-intensive application at SC2003

15
International Grid activities
LCG - The European particle physics
Grid TeraGrid - US Grid linking major HPC
centres GGF - the Grid standards body (c.f.
W3C, IETF)
16
What does it take to build a Grid?

• Resources - CPU, network, storage
• People - sysadmins, application developers, Grid
  experts
• Grid middleware - Globus, Condor, Unicore
• Security infrastructure
• Maintenance - ongoing monitoring, upgrades and
• co-ordination of this between multiple sites
• Applications and users!

17

My co-ordinates
mah1002_at_cam.ac.uk Centre for Mathematical
Sciences http//www.escience.cam.ac.uk/mark/