ischia, italy - 9-21 july 20061 session 1 second part day 1 monday 10 th july malcolm atkinson
Post on 19-Dec-2015
214 views
TRANSCRIPT
Ischia, Italy - 9-21 July 2006 3
Introduction to GridsIntroduction to Grids
History of Distributed Computing
Challenges of Distributed Computing
Driving Forces & Motivation for Grids Collaboration
Data Deluge
Computational Roadmap
A history of Grids
Ischia, Italy - 9-21 July 2006 4
What is e-Science?What is e-Science?
• Goal: to enable better research in all disciplines• Method: Develop collaboration supported by
advanced distributed computation– to generate, curate and analyse rich data resources
• From experiments, observations and simulations• Quality management, preservation and reliable evidence
– to develop and explore models and simulations• Computation and data at all scales• Trustworthy, economic, timely and relevant results
– to enable dynamic distributed collaboration• Facilitating collaboration with information and resource sharing• Security, trust, reliability, accountability, manageability and agility
Our challenge is to develop an integrated approach to all three
Ischia, Italy - 9-21 July 2006 5
Distributed Systems HistoryDistributed Systems History
1960 1970 1980 1990 2000
Bespoke pioneering distributed systems, e.g. SABRE & SAGE
IBM CICS
Linklider proposes shared multi-site computing
Dozens of academic networks Many incompatible protocols
ARPA net
CORBA & DCOM
IP-based InternetAcademic & Research
WWW
Ethernet
TCP
Ischia, Italy - 9-21 July 2006 6
Distributed Systems to GridsDistributed Systems to Grids
1960 1970 1980 1990 2000
Bespoke pioneering distributed systems
IBM CICS
Linklater proposes shared multi-site computing
Dozens of academic networks Two dominant protocols
ARPA net
CORBA & DCOM
IP-based InternetAcademic & Research
WWWCondor
I-way
GlobusStarts
Unicore
Web Services
Many research gridsUsing many protocols & M/W stacks
D-Gridstarts
Collaboration via shared bio/chem/medical “DBs”
EDG EGEEEGEE II
Ischia, Italy - 9-21 July 2006 8
A Grid Computing Timeline
I-Way
: Sup
erCom
putin
g ‘95
1995 ’96 ’97 ’98 ’99 2000 ’01 ’02 ’03 ’04 ’05 2006US G
rid F
orum
form
s at S
C ‘98
Grid F
orum
s mer
ge, fo
rm G
GF
OGSA-WG fo
rmed
OGSA v1.0
“Ana
tomy”
pape
r
“Phy
siolog
y” pa
per
Europ
ean
& AP G
rid F
orum
s
• UK e-Science programme starts• UK e-Science programme starts
• DARPA funds Globus Toolkit & Legion• EU funds UNICORE project• US DoE pioneers grids for scientific research• NSF funds National Technology Grid• NASA starts Information Power Grid
• DARPA funds Globus Toolkit & Legion• EU funds UNICORE project• US DoE pioneers grids for scientific research• NSF funds National Technology Grid• NASA starts Information Power Grid
Today:• Grid solutions are common for HPC• Grid-based business solutions are
becoming common• Required technologies & standards are
evolving
Today:• Grid solutions are common for HPC• Grid-based business solutions are
becoming common• Required technologies & standards are
evolving
Japan government funds:• Business Grid project • NAREGI project
Japan government funds:• Business Grid project • NAREGI project
Source: Hiro Kishimoto GGF17 Keynote May 2006
GGF & E
GA
form
OGF
Ischia, Italy - 9-21 July 2006 9
What is a Grid?
LicenseLicense
PrinterPrinter
A grid is a system consisting of− Distributed but connected resources and − Software and/or hardware that provides and manages logically
seamless access to those resources to meet desired objectives
A grid is a system consisting of− Distributed but connected resources and − Software and/or hardware that provides and manages logically
seamless access to those resources to meet desired objectives
R2AD
DatabaseDatabase
Webserver
Webserver
Data CenterCluster
Handheld Supercomputer
Workstation
Server
Source: Hiro Kishimoto GGF17 Keynote May 2006
Ischia, Italy - 9-21 July 2006 10Source: Hiro Kishimoto GGF17 Keynote May 2006
Grid & Related Paradigms
Utility Computing• Computing “services”• No knowledge of provider• Enabled by grid technology
Utility Computing• Computing “services”• No knowledge of provider• Enabled by grid technology
Distributed Computing• Loosely coupled• Heterogeneous• Single Administration
Distributed Computing• Loosely coupled• Heterogeneous• Single Administration
Cluster• Tightly coupled• Homogeneous• Cooperative working
Cluster• Tightly coupled• Homogeneous• Cooperative working
Grid Computing• Large scale• Cross-organizational• Geographical distribution• Distributed Management
Grid Computing• Large scale• Cross-organizational• Geographical distribution• Distributed Management
Ischia, Italy - 9-21 July 2006 11
How Are Grids Used?
High-performance computing
Collaborative data-sharing
Collaborative design
Drug discovery
Financial modeling
Data center automation
High-energy physics
Life sciences
E-Business
E-Science
Source: Hiro Kishimoto GGF17 Keynote May 2006
Ischia, Italy - 9-21 July 2006 12
Grids In Use: E-Science Examples
• Data sharing and integration− Life sciences, sharing standard data-sets,
combining collaborative data-sets− Medical informatics, integrating hospital information
systems for better care and better science− Sciences, high-energy physics
• Data sharing and integration− Life sciences, sharing standard data-sets,
combining collaborative data-sets− Medical informatics, integrating hospital information
systems for better care and better science− Sciences, high-energy physics
• Capability computing− Life sciences, molecular modeling, tomography− Engineering, materials science− Sciences, astronomy, physics
• Capability computing− Life sciences, molecular modeling, tomography− Engineering, materials science− Sciences, astronomy, physics
• High-throughput, capacity computing for − Life sciences: BLAST, CHARMM, drug screening− Engineering: aircraft design, materials, biomedical− Sciences: high-energy physics, economic modeling
• High-throughput, capacity computing for − Life sciences: BLAST, CHARMM, drug screening− Engineering: aircraft design, materials, biomedical− Sciences: high-energy physics, economic modeling
• Simulation-based science and engineering− Earthquake simulation
• Simulation-based science and engineering− Earthquake simulation
Source: Hiro Kishimoto GGF17 Keynote May 2006
Ischia, Italy - 9-21 July 2006 14
Grids: integrating & providing homogeneityGrids: integrating & providing homogeneity
• (Potentially) Generic & Industry Supported• Mechanism for combining many heterogeneous
resources so they can be used remotely– Data resources– Computation resources– Instruments– Research processes & procedures
• Restricting choices as to how it may be done– Harder for provider to make localised decisions– Deployment can be challenging
• Providing more homogeneity through virtualisation– Should be easier to compose services– More opportunity to amortise costs
• A component of e-Infrastructure
Deliberately choosing consistent interfaces, protocols
& management controls across a set of compatible
services
Ischia, Italy - 9-21 July 2006 15
Grids as a Foundation for SolutionsGrids as a Foundation for Solutions
• The grid per se doesn’t provide– Supported e-Science methods– Supported data & information resources– Computations – Convenient access
• Grids help providers of these– International & national secure e-Infrastructure– Standards for interoperation– Standard APIs to promote re-use
• But Research Support must be built– What is needed?– Who should do it?
Much to be done by
developers of
applications & services
and by resource providers
Ischia, Italy - 9-21 July 2006 16
Grids as a Foundation for SolutionsGrids as a Foundation for Solutions
• The grid per se doesn’t provide– Supported e-Science methods– Supported data & information resources– Computations – Convenient access
• Grids help providers of these– International & national secure e-Infrastructure– Standards for interoperation– Standard APIs to promote re-use
• But Research Support must be built– What is needed?– Who should do it?
Much to be done by
developers of
applications & services
and by resource providers
Therefore must support many categories of user:
•Application & Service Developers
•Deployers & Operations teams
•Higher-level abstraction builders & gateway providers
Ischia, Italy - 9-21 July 2006 18
Why use / build Grids?Why use / build Grids?
• Research Arguments– Enables new ways of working– New distributed & collaborative research– Unprecedented scale and resources
• Economic Arguments– Reduced system management costs– Shared resources better utilisation– Pooled resources increased capacity– Load sharing & utility computing – Cheaper disaster recovery
Ischia, Italy - 9-21 July 2006 19
Why use / build Grids?Why use / build Grids?
• Computer Science Arguments– New attempt at an old hard problem– Frustrating ignorance existing results– New scale, new dynamics, new scope
• Engineering Arguments– Enable autonomous organisations to
• Write complementary software components• Set up run & use complementary services• Share operational responsibility• General & consistent environment for
Abstraction, Automation, Optimisation & Tools• Generally available code mobility
Ischia, Italy - 9-21 July 2006 20
Why use / build Grids?Why use / build Grids?
• Political & Management Arguments– Stimulate innovation– Promote intra-organisation collaboration– Promote inter-enterprise collaboration
Ischia, Italy - 9-21 July 2006 22
Drivers for GridsDrivers for Grids
• Collaboration– Expertise is distributed– Necessary to achieve critical mass of effort– Necessary to raise sufficient resources
• Computational Power– Rapid growth in number of processors– Powered by Moore’s law + device roadmap– Challenge to transform models to exploit this
• Deluge of Data– Growth in scale: Number and Size of resources– Growth in complexity– Policy drives greater availability
Biomedical Research Informatics Delivered by Grid Enabled Services
Glasgow Edinburgh
Leicester Oxford
London
Netherlands
Publically Curated Data
Private data
Private data
Private data
Private data
Private data
Private data
CFG Virtual Organisation Ensembl
MGI
HUGO
OMIM
SWISS-PROT
… DATA HUB
RGD
SyntenyGrid
Service
blast
+
Portal
http://www.brc.dcs.gla.ac.uk/projects/bridges/
Overview of e-Infrastructure 25
eDiaMoND: Screening for Breast Cancer
1 Trust Many TrustsCollaborative WorkingAudit capabilityEpidemiology
Other Modalities-MRI-PET-Ultrasound
Better access toCase informationAnd digital tools
Supplement MentoringWith access to digitalTraining cases and sharingOf information acrossclinics
LettersRadiology reportingsystems
eDiaMoNDGrid
2ndary CaptureOr FFD
Case Information
X-Rays andCase Information
DigitalReading
SMF
Case andReading Information
CAD Temporal Comparison
Screening
ElectronicPatient Records
Assessment/ SymptomaticBiopsy
Case andReading Information
Symptomatic/AssessmentInformation
Training
Manage Training Cases
Perform Training
SMF CAD 3D Images
Patients
Provided by eDiamond project: Prof. Sir Mike Brady et al.
climateprediction.net and GENIE
• Largest climate model ensemble
• >45,000 users, >1,000,000 model years
10K2K
Response of Atlantic circulation to freshwater forcing
27Courtesy of David Gavaghan & IB Team
Integrative Biology
Tackling two Grand Challenge research questions:
• What causes heart disease?• How does a cancer form and grow?
Together these diseases cause 61% of all UK deaths
Will build a powerful, fault-tolerant Grid infrastructure for biomedical science
Enabling biomedical researchers to use distributed resources such as high-performance computers, databases and visualisation tools to develop complex models of how these killer diseases develop.
Ischia, Italy - 9-21 July 2006 29
Foundations of CollaborationFoundations of Collaboration
• Strong commitment by individuals– To work together– To take on communication challenges– Mutual respect & mutual trust
• Distributed technology– To support information interchange– To support resource sharing– To support data integration– To support trust building
• Sufficient time• Common goals• Complementary knowledge, skills & data
Can we predictwhen it will work?Can we findremedies when itdoesn’t?
Ischia, Italy - 9-21 July 2006 30
Grid & Collaboration QuestionsGrid & Collaboration Questions
• Without collaboration little is achievable• But must collaboration precede successful
grid applications?• Or will persistently and pervasively available
grids stimulate collaborations?
• If we deliver support for collaborative teams, will we also support the individual researcher?
• Can we use grids to democratise computation?
Slide 31
CARMEN - Scales of Integration
resolving the ‘neural code’ from the timing of action potential activity
determining ion channel contribution to the timing of action potentials
examining integration within networks of differing dimensions
Understanding the brain may be the greatest
informatics challenge of the 21st century
Slide 32
CARMEN Consortium
Leadership & Infrastructure
Colin Ingram
Paul Watson
Leslie Smith Jim Austin
Slide 33
CARMEN Consortium
International Partners
Daniel Gardner(Cornell)
Lead for the US NIH, Neuroscience Information Framework and Brain ML
Shiro Usui(RIKEN Brain Science Institute)
Lead for the Japan Node of the International Neuroinformatics Coordinating Facility
Sten Grillner(Karolinska Institute)
Chairman of the OECD,International Neuroinformatics Coordinating Facility
Ad Aertsen(Freiburg)
Neural network modelling and large-scale simulations
George Gerstein(Pennsylvania)
Analysis of spike pattern trains
Slide 34
CARMEN Consortium
Commercial Partners
- applications in the pharmaceutical sector
- interfacing of data acquisition software
- application of infrastructure
- commercialisation of tools
16th March 2006
NanoCMOSgriD Meeting the Design Challenges of Nano-CMOS Electronics
The ChallengeThe Challenge
Ischia, Italy - 9-21 July 2006
International Tech nology Roadmap for Semiconductors Year 2005 2010 2015 2020
MPU Half Pitch (nm) 90 45 25 14
MPU Gate Length (nm) 32 18 10 6
2005 edition Toshiba 04
Device diversification
90nm: HP, LOP, LSTP
45nm: UTB SOI
32nm: Double gate
25 nm
Bulk MOSFET
FD SOI
UTB SOI
FinFET
HP(MPU)
LOP
LSTP
Stat.Sets
230 nm
Bulk MOSFET
Standard
SingleSet
16th March 2006
NanoCMOSgriD Meeting the Design Challenges of Nano-CMOS Electronics
Advanced Processor Technologies Group (APTGUM)
Device Modelling Group (DMGUG)
Electronic Systems Design Group (ESDGUS)
Intelligent Systems Group (ISGUY)
National e-Science Centre (NeSC)
Microsystems Technology Group (MSTGUG)
Mixed-Mode Design Group in IMNS (MMDGUE)
e-Science NorthWest Centre (eSNW)
University PartnersUniversity Partners
Ischia, Italy - 9-21 July 2006
16th March 2006
NanoCMOSgriD Meeting the Design Challenges of Nano-CMOS Electronics
Industrial PartnersIndustrial PartnersGlobal EDS vendor and world TCAD leader600 licences of grid implementation, model implementation
UK fabless design company and world microprocessor leaderCore IP, simulation tools, staff time
UK fabless design company and world mixed mode leaderAdditional PhD studentship for mixed mode design
Global semiconductor player with strong UK presenceAccess to technology, device data, processing
Global semiconductor player with strong UK presenceAccess to technology, device data, processing
Global semiconductor player with UK presenceCASE studentship, interconnects
Trade association of the microelectronics industry in the UKRecruiting new industrial partners and dissemination
Ischia, Italy - 9-21 July 2006 39
Grids in contextGrids in context
• Part of a long-term drive for distributed computing– A new and ambitious form
• Search for trade-offs & multiple uses– Leads to many varieties– Multiple stake holders
• Many good reasons for building & using grids• Question
– Will we have many grids or a consistent general purpose “foundation” grid?
• What are the minimum standards across the grids
• Collaboration is a key driver & enabler
Ischia, Italy - 9-21 July 2006 40
Minimum Grid FunctionalitiesMinimum Grid Functionalities
• Supports distributed computation– Data and computation– Over a variety of
• hardware components (servers, data stores, …)• Software components (services: resource managers,
computation and data services)
– With regularity that can be exploited• By applications• By other middleware & tools• By providers and operations
– It will normally have security mechanisms • To develop and sustain trust regimes
Ischia, Italy - 9-21 July 2006 41
After the breakAfter the break
• Distributed system challenges
• Service Oriented Architectures
• Examples of the basic architecture