Enabling, facilitating and delivering quality training in the UK and Internationally

Introduction to e-science concepts

Mike Mineter TiiTraining O Otutreach and dEdti Education National e-Science Centre

[email protected] 2 Overview The vision –what is e -science?! The pioneers StfSome aspects of e-science The week ahead

All slides will be uploaded to the agendapage 3 4

“Enger lunRound d” 1: Odd one outGeese

Is the effect > Σ parts ??

FthRilForth Rail bridge Orchestra 5

“Enger lunRound d” 1: Odd one outGeese

Go further , faster together Is the effect > Σ parts ?? Coor din ated use of resources

Infrastructure: communication & use of Railwaydistributed resourcesbridge Orchestra 6

Go further , faster together

Coor din ated use of resources Infrastructure: communication & use of distributed resources 7 Enhancing science Go further, Researchers faster together

Enabled by

Coordinated use ThTechnol ogi es and servi ces of resources Perm itting access to Infrastructure Resources (Data, Storage, Internet Computers, Instruments,…) connected by the Internet 8 Enhanced (e-) science is….

‘…about global collaboration in key areas ofif science, and dtht the next generation of infrastructure that will enable it.’ John Taylor Director General of Research Councils Office of Science and Technology 2001 Computing intensive science Enabling Grids for E-sciencE • Many vital challenges require community effort – Fundamental ppproperties of matter – Genomics – Climate change – Medical diagnostics • Research is increasingly digital, with increasing amounts of data • Computation ever more demanding e.g.: experimental science uses ever more sophisticated sensors – HtfdtHuge amounts of data – Serves user communities around the world – International collaborations

EGEE-II INFSO-RI-031688 9 10 Overview The vision –what is e -science?! The pioneers - 3 early adopters StfSome aspects of e-science The week ahead The

Enabling Grids for E-sciencE Experiments

ATLAS CMS

~10-15 PetaBytes /year ~108 events/year ~103 batch and interactive users

LHCb

EGEE-II INFSO-RI-031688 11 The LHC Data Challenge Enabling Grids for E-sciencE Starting from this event

Looking for this “signature”

Æ Selectivity: 1 in 1013 (Like looking for a needle in 20 million haystacks)

EGEE-II INFSO-RI-031688 12 UK Grid for Particle Physics

GridPP www.gridpp.ac.uk ATLAS detectors, 2/3/06 VirtualVir tual ObseObservatoriesr vatori es

Observations made across entire electromagnetic spectrum

μ μ μ ROSAT k V ~keV DSS O O li ptical 2MASS 2 2 IRAS 25 25 IRAS 100 GB 6cm NVSS 20 20cm WENSSENSS 92 92cm ⇒e.g. different views of a local galaxy Nee d a ll o f them to un ders tan d p hys ics fu lly Databases are located throughout the world

Peter Clarke 14 Biomedical Research Informatics Delivered by Grid Enabled Services

VO Authorisation CFG Virtual Publically Curated Data Organisation Ensembl OMIM Glasgow Private SWISS-PROT Edinburgh MGI Information data HUGO RGD Private Integrator … data DATA HUB Leicester Oxford Private data Netherlands

Private PiP rivate data Synteny data London Grid Private Service data

+

http://www.brc.dcs.gla.ac.uk/projects/bridges/ 16 Overview The vision –what is e -science?! The pioneers - 3 early adopters StfSome aspects of e-science The week ahead 17 What is fundamental to e-Science? 3 m ost evi den t aspects Data, Computation, Collaboration Underppyinned by mechanisms for crossin g administrative domains with security Communicating identity – of resources as well as users Communicating role and membership in collaborations – basis of authorisation Single sign-on Encryption / integrity-checking Virtual comppguting across administrative domains… “” for short. Typical current grid

Enabling Grids for E-sciencE • User communities negotiate with sites to agree access to resources

• Grid middl eware runs on each shared resource to provide INTERNET – DiData services – Computation services – Single sign-on

• Distributed services (both people and middleware) enable the grid

INFSO-RI-508833 18 Grid Middleware Enabling Grids for E-sciencE

• When using a PC or • When using a Grid you workstation you – Login with digital – Login with a username credentials – single sign- and password on (“Authentication”) (“Authentication ” ) – Use rights given you – Use rights given to you (“Authorisation”) (“Authorisation”) – Run jobs – Run jobs – Manage files: create – Manage files: create them, read/write, list them, read/write, list directories directories • Components are • Services are linked by link ed b y a b us the I nt ernet • Operating system • Middleware • One admin . domain • MdidiMany admin. domains

EGEE-II INFSO-RI-031688 19 20 Scales of grids

International collaboration International gg(rid (EGEE) (and DEISA for HPC) neity ion ee National datacentres & tt HPC centres,… UK’s National Grid bora

terog Service

Nationwide aa collaboration ee

r coll Regional grid (NWGrid, Regional collaboration ter h ee aa White R ose Grid) University-wide Wid Gre collaboration Campus grid

Desktop 21 One story of e-science evolution

Ecosystem of technologies

Grid computing Web services

World-wide web High-end computing and Meta-computing Massively parallel INTERNET computing 22 Empowering collaboration – Route 1 “Service-oriented Sharing and orchestrating research” components

Collaborative Accessing data, computers across administrative domains ….Enabled “virtual computing” by Grids

Imppprovised cooperation Email File exchange ssh access to run programs Enabled by networks: national, regional and People with shared goals International: GEANT 23 Empowering collaboration – Route 2

Community that already shares and orchestrates components / services

Add grid-enabled services and components

Effect – expand horizons of collaboration and richness of resources accessible 24

Workflow, Portals…. Services seen by researcher Applications, data …

Ecosystem of e-IfInfrast ruct ure e-science services technologies

Networks Resources (Storage, Compp,uters, Instruments, ,)…) 25 Roles Workflow, Portals…. researcher Applications, data …

Service provider e-IfInfrast ruct ure services

Networks Resources Resource provider (Storage, Compp,uters, Instruments, ,)…) 26 The challenge for informaticians

Where computer science AliiApplication meets the application Application communities toolkits, standards Building upon middleware Middleware: “higher level services ” Basic services: AA,,j job submission, info, … Grids provide these services.

(AA: authorisation, authentication) 27 e-Science science that is made possible by the sharing across the Internet of(f resources (data, instruments, computation, people’s expertise...) Resources within a collaboration Resources shared between collaborations Enabled by “e-infrastructures” at campus, UK, international scales Effect: “better, bigger, faster” research Faster from concept to doing research Scale and heterogeneity of computers, data,…. Foundation for orchestration of service-oriented research Distinct roles: researcher can do research, not also managgging resources…, infrastructure provider, resource provider Informaticians are the vital link 28 This week

Taverna, MyGrid Workflow, Portals…. P-Grade

UK’s e-Infrastructure Services International grids Resources (Storage, Compp,uters, Instruments, ,)…) This week 29

Taverna, Monday: Taverna workflow MyGrid Tuesday:advanced Taverna, web P-Grade services, MyGrid

UK’s National Wed./ Thursday: National Grid Grid Service Service (with P-Grade) International grids Friday: New NGS technologies, International grids 30

Agendapage – go via httppg://www.nesc.ac.uk/training