Introduction to the EPIKH/CHAIN projects Riccardo Rotondo - - PowerPoint PPT Presentation

introduction to the epikh chain projects
SMART_READER_LITE
LIVE PREVIEW

Introduction to the EPIKH/CHAIN projects Riccardo Rotondo - - PowerPoint PPT Presentation

The EPIKH Project (Exchange Programme to advance e-Infrastructure Know-How) Introduction to the EPIKH/CHAIN projects Riccardo Rotondo (riccardo.rotondo@garr.it) Consortium GARR Joint CHAIN/EPIKH School for Application Porting to Science


slide-1
SLIDE 1

www.epikh.eu

The EPIKH Project

(Exchange Programme to advance e-Infrastructure Know-How)

Introduction to the EPIKH/CHAIN projects

Riccardo Rotondo (riccardo.rotondo@garr.it) Consortium GARR Joint CHAIN/EPIKH School for Application Porting to Science Gateways Beijing, 11.04.2012

slide-2
SLIDE 2

Beijing, Asia 4, 11-04-2012 2

Outline

  • Grids and Infrastructures
  • The EPIKH project
  • The CHAIN project
slide-3
SLIDE 3

Computationally Intensive research

Beijing, Asia 4, 11-04-2012 3

slide-4
SLIDE 4

Today’ research

Beijing, Asia 4, 11-04-2012

slide-5
SLIDE 5

The Grid

  • A ¡GRID ¡is ¡a ¡distributed ¡compu4ng ¡and ¡storage ¡

infrastructure ¡– ¡spanning ¡several ¡administra4ve ¡ domains ¡-­‑ ¡allowing ¡sharing ¡of ¡resources ¡in ¡a ¡ coordinated ¡manner ¡by ¡a ¡set ¡of ¡homogeneous ¡users ¡

  • rganized ¡within ¡Virtual ¡Organiza4ons ¡
  • A ¡GRID ¡provides ¡single ¡sign ¡on ¡to ¡access ¡a ¡large ¡variety ¡
  • f ¡resources ¡and ¡an ¡added ¡value ¡with ¡respect ¡to ¡the ¡

bare ¡sum ¡of ¡its ¡components ¡

  • GRIDS ¡are ¡the ¡key ¡enabler ¡of ¡e-­‑Science ¡

Beijing, Asia 4, 11-04-2012 5

slide-6
SLIDE 6

e-Science

– e-Science is about global collaboration in key areas of science, and the next generation of infrastructure that will enable it e-Infrastructure = Networks + Grids ..

+ Operations, Support, Training… + Data centres, archives, instruments…

  • Networks interconnect resources
  • Grids enable flexible usage of interconnected resources:
  • Distributed computing across different administrative domains

Beijing, Asia 4, 11-04-2012 6

slide-7
SLIDE 7

Overview

Beijing, Asia 4, 11-04-2012 7

  • Virtual ¡organisa-ons ¡

e-­‑Science ¡

7

Applications Data Instruments

e-Infrastructure

  • Network
  • Computational and

Storage resources

slide-8
SLIDE 8

e-Infrastructure

Beijing, Asia 4, 11-04-2012 8

National Research and Education Networks National Grid Initiatives International Grid Initiatives

slide-9
SLIDE 9

World’ Grids

Beijing, Asia 4, 11-04-2012 9

The ¡Global ¡Grid ¡

EELA OSG XSEDE

NAREGI EUMedGrid BalticGrid SEE-Grid EUIndiaGrid EUAsiaGrid EUChinaGrid DEISA

EGI EU fundend Non EU

slide-10
SLIDE 10

Beijing, Asia 4, 11-04-2012 10

slide-11
SLIDE 11

Beijing, Asia 4, 11-04-2012 11

The Triangle of Knowledge

Research & Development Innovation Education & Training Building e-Infrastructures is a waste if we don’t “build”, at the same time, their users. Along with e-Infrastructures, t-Infrastructures and training programmes are also needed.

t-Infrastructure

slide-12
SLIDE 12

Beijing, Asia 4, 11-04-2012 12

The EPIKH Project

(www.epikh.eu)

  • “Exchange Programme to advance e-Infrastructure

Know-How” (EPIKH)

– EU FP7- Marie Curie Actions – People - International Research Staff Exchange Scheme (IRSES)

  • Consortium “numbers”:

– 23 partners; – 18 countries; – 4 continents (Africa, Latin America, Asia, Europe); – 115 persons involved; – >650 researcher-months; – >500 secondments;

  • Duration: March, 1, 2009 – February, 28, 2013 (48

months)

  • EC contribution: 1,188,000 €
slide-13
SLIDE 13

Beijing, Asia 4, 11-04-2012 13

The EPIKH Partners

Chinese Partner

slide-14
SLIDE 14

Beijing, Asia 4, 11-04-2012 14

The EPIKH goals

  • Strategic aims:

– Reinforce the impact of e-Infrastructures in scientific research defining and delivering stimulating programme of educational events, including Grid Schools and High Performance Computing courses; – Broaden the engagement in e-Science activities and collaborations both geographically and across disciplines.

  • Specific actions:

– Spreading the knowledge about the “Grid Paradigm” to all potential users: both system administrators and application developers through an extensive training programme; – Easing the access of the trained people to the e-Infrastructures existing in the areas of action of the project; – Fostering the establishment of scientific collaborations among the countries/continents involved in the project.

slide-15
SLIDE 15

Beijing, Asia 4, 11-04-2012 15

EPIKH secondments’ programme

1st Phase: Non-EU → EU

(selected)

  • Res. Programs

2nd Phase: EU → Non-EU Grid Schools

(with 1st phase tutors)

Workshops

slide-16
SLIDE 16

Beijing, Asia 4, 11-04-2012 16

EPIKH Quality & Certification Programme

  • All seconded people must be
  • fficially selected against their

CVs;

  • All Non-EU→EU secondments

must have a Research Program agreed between “sending” and “accepting” organisations;

  • All secondments must have final

reports both from the seconded person and the “accepting”

  • rganisation

– If both reports will be successfull the seconded person will be provided with an official certificate

slide-17
SLIDE 17

Beijing, Asia 4, 11-04-2012 17

EPIKH on the web

www.epikh.eu wiki.epikh.eu documents.epikh.eu agenda.epikh.eu

slide-18
SLIDE 18

Beijing, Asia 4, 11-04-2012 18

Grants and applications

(wiki.epikh.eu)

The call for applications is still open !

slide-19
SLIDE 19

GEOMAP Views - Partners

Beijing, Asia 4, 11-04-2012 19

slide-20
SLIDE 20

GEOMAP Views – Events

Beijing, Asia 4, 11-04-2012 20

slide-21
SLIDE 21

GEOMAP Views – Events

Beijing, Asia 4, 11-04-2012 21

slide-22
SLIDE 22

EPIKH Applications

Beijing, Asia 4, 11-04-2012 22

slide-23
SLIDE 23

Beijing, Asia 4, 11-04-2012 23

GILDA: the “de facto” standard EPIKH t-Infrastructure

  • GILDA: Grid INFN Laboratory for Dissemination Activities

https://gilda.ct.infn.it

  • GILDA is a complete Grid production quality infrastructure

fully devoted to demonstrate/disseminate Grid computing

  • GILDA is an international effort funded by multiple sources
  • GILDA is adopted by several Grid projects worldwide
  • It runs the latest Grid middleware version supporting many

middleware such as Condor, Globus, OurGrid, UNICORE, etc.

  • Using GILDA is free of charge
slide-24
SLIDE 24

Beijing, Asia 4, 11-04-2012 24

The GILDA formats and services

(https://gilda.ct.infn.it)

  • Training on Grid

– Tutorial for Users – Tutorial for Site Admin – Grid Schools – Tutorial for Trainers – Customized events – University courses (Degree, Ms.C, Ph.D, etc.) – Training materials: Wiki pages, Printable Documentation, Demos, Videos, etc.

  • Grid Application development

– Support to Grid application development – Deployment of the applications into the GENIUS and P-GRADE grid portals – GILDA Ticketing support system

  • Infrastructure

– GILDA CA – GILDA VOMS – Creation of pilot test-beds: EELA, EUMED, EUCHINA, EUINDIA, SAGRID, etc. – Testing and development on new available gLite services – Dedicated Software repository/Virtual services (INFN Grid) – Monitoring (GStat, SAM, GridMap, etc. )

Science Gateways Grid School

slide-25
SLIDE 25

Beijing, Asia 4, 11-04-2012 25

Conclusions

  • Research communities can greatly benefit from EPIKH

and use GILDA to “gridify” and run applications of local interest;

  • EPIKH can represents for young brilliant scientists and

technologists a unique occasion to learn and adopt the Grid paradigm through stimulating research programmes;

  • Thanks to EPIKH activities, EUMED Organisations can

setup “clones” of GILDA locally to implement campus grids or even nation-wide e-infrastructures (good examples are already existing both in South Africa and Cuba);

  • EPIKH/GILDA can offer free technical support for anyone

willing to organize local training events on Grid computing:

– It can be undergraduate courses, M.Sc. courses, Ph.D. curricula, tutorials, Grid Schools, etc.

slide-26
SLIDE 26

Beijing, Asia 4, 11-04-2012 26

slide-27
SLIDE 27

Outline

} General information and CHAIN vision } Workplan } State of the art analysis } Data analysis and recommendations } Virtual Research Communities } Interoperation & Interoperability } Conclusions

27 Beijing, Asia 4, 11-04-2012

slide-28
SLIDE 28

Regional Grid infrastructures

CNGrid

NKN & Garuda

EUAsiaGrid

SAGrid & SANREN

GISELA

28 Beijing, Asia 4, 11-04-2012

slide-29
SLIDE 29

CHAIN: global coverage

29 Beijing, Asia 4, 11-04-2012

slide-30
SLIDE 30

Project objectives

  • Define a strategy and a model for external

collaboration, in close collaboration with EGI.eu which will enable operational and organisation interfacing of EGI and external eInfrastructures

  • Validate this model, as a proof-of-principle, by

supporting the extension and consolidation of worldwide Virtual Research Communities

  • Explore and propose concrete steps forward towards

the coordination with other projects and initiatives (e.g. EGI.eu, EUMEDGRID-Support, EUIndiaGrid2, LinkSCEEM2, NKN & Garuda, etc.)

30 Beijing, Asia 4, 11-04-2012

slide-31
SLIDE 31

Project information

  • Grant Agreement for a total EC contribution of 1.1 M€
  • Total cost: about 1.9 M€
  • Start Date: 1st December 2010 - Duration 24 Months
  • Partners:

1)

INFN (Italy - Coordinator)

2)

CESNET (Czech Rep.)

3)

CIEMAT (Spain)

4)

GRNET (Greece)

5)

IHEP (China)

6)

UBUNTUNET (Africa)

7)

CLARA (Latin America)

8)

PSA (India)

9)

ASREN (Med./Middle East/Gulf) Since 1 August 2011

31 Beijing, Asia 4, 11-04-2012

slide-32
SLIDE 32

Project strategic vision

} A world-wide Distributed Computing Infrastructure can address big scientific challenges that are not manageable with departmental computing systems } Virtual Research Communities can transparently access different kind of resources: scientific applications and tools, Data Repositories, down to CPUs and Disks. The vision is that of VRCs sharing resources ubiquitously across different administrative domains } Regional e-Infrastructures should be made interoperable among each other. CHAIN is committed to promote and validate a proof-of-concept that addresses this

32 Beijing, Asia 4, 11-04-2012

slide-33
SLIDE 33

Disseminate (WP5)

Project workplan

State of the Art Assessment (WP2) Analyse the different Regional Approaches (WP2, WP4) Make Recommendatio ns (WP2, WP3, WP4)

Involve the VRCs (WP3) Propose a Road- Map and Intermediate solutions (WP4, WP3) Demonstrate the usefulness of interoperation (WP3, WP4)

33

Beijing, Asia 4, 11-04-2012

slide-34
SLIDE 34

State of the art analysis (WP2)

} Analysis of existing NGI literature and related questionnaires } Creation of the regional and NGI questionnaires } Questionnaires being implemented and published online

¡

} Collection of contact points from all

continents

} Questionnaire is kept open and

collection of contact points from all continents is continued

} Questionnaire data provided

through the CHAIN Knowledge Base

34 Beijing, Asia 4, 11-04-2012

slide-35
SLIDE 35

Knowledge base (WP2, WP5)

35

Beijing, Asia 4, 11-04-2012

www.chain-project/knowledge-base

slide-36
SLIDE 36

Country view

36

Beijing, Asia 4, 11-04-2012

slide-37
SLIDE 37

Data analysis (WP4, WP2)

Nuber of sites and number of CPU - Mediterranean

1 2 3 4 5 6 7 8 A l g e r i a E g y p t T u n i s i a M

  • r
  • c

c

  • I

r a n U n i t e d A r a b E m i r a t e s J

  • r

d a n Sites 100 200 300 400 500 600 700 CPU cores

Number of Grid sites / clusters: Number of CPU cores:

Number of sites and number of CPUs

37

Beijing, Asia 4, 11-04-2012

Nuber of sites and number of CPU - Latin America

10 20 30 40 50 60 Brazil Ecuador Colombia Argentina Panamá Paraguay Mexico Cuba Venezuela Costa Rica Guatemala Peru Sites 1000 2000 3000 4000 5000 6000 CPU cores

Number of Grid sites / clusters: Number of CPU cores:

Nuber of sites and number of CPU - Asia Pacific

1 2 3 4 5 6 7 8 Indonesia Singapore Vietnam New Zealand Taiwan Sites 1000 2000 3000 4000 5000 6000 7000 CPU cores

Number of Grid sites / clusters: Number of CPU cores:

slide-38
SLIDE 38

WP2 Recommendations

  • 74 Detailed recommendations classified by:

– Short (1 year), Medium (3 years) and Long term (5 years) – High, Medium, Low priority – National Grid Initiatives (9):

§ General (5); Regional (4)

– Interoperations (14):

§ General (1), ROC (3), User Support (1), Monitoring (3), Security (2), Core Services (2); Middleware (2)

– Interoperability (2): General (1), Input/Output (1) – Virtual Research Communities’ perspective: General (2) – Regional planning (47):

§ Africa (9), Asia Pacific (6), Central Asia (5), China (7), India (4), Latin America (5), Mediterranean & Arab Countries (11)

38 Beijing, Asia 4, 11-04-2012

slide-39
SLIDE 39

Identified VRC (1)

} Coordinator: Dr. Alexandre M.J.J. Bonvin – Universiteit Utrecht } Structural biology, making use of NMR (Nuclear Magnetic Resonance),

SAXS (Small Angle X-ray Scattering), computational modelling or other related techniques to study biomolecules

} We-NMR partners in red, collaborations in orange

} MoU signed on 21/09/2011 } New contacts (WP2 survey)

} Burundi } Costa Rica

http://www.wenmr.eu

Beijing, Asia 4, 11-04-2012 39

slide-40
SLIDE 40

Identified VRC (2)

} Coordinator: Dr. Antonio Cofiño – Universidad de Cantabria } Meteorology and other Earth Science related areas willing to enhance

their forecasting techniques on Idealized Simulations, Regional and Global Applications, Parameterization and Data Assimilation or Forecast and Hurricane Research

} Dissemination within CORDEX project } WRF4G partners in red, collaborations in orange

} MoU signed on 19/09/2011 } New contacts (WP2 survey)

} Burundi } China } Costa Rica } Cuba

http://www.meteo.unican.es/es/software/ wrf4g

Beijing, Asia 4, 11-04-2012 40

slide-41
SLIDE 41

Identified VRC (3)

} Coordinator: Dr. David Posada – Universidad de Vigo } Evolutionary biology for the statistical selection of best-fit models of

nucleotide substitution and amino acid replacement for a given set of aligned sequences (molecular systematics, phylogenetics, phylogenomics, molecular evolution and/or bioinformatics)

} jModelTest & ProtTest registered users

} Modeltest ~ 30.000 } jModelTest ~ 11.000 } Prottest ~ 5.000 } MoU signed on 21/02/2012 } New sequential and distributed computing versions available on SG

and being tested

} New contacts (WP2 survey): Burundi, Costa Rica, Democratic Republic

  • f Congo, Ethiopia, Nigeria, Panama, Sudan, Taiwan

http://darwin.uvigo.es/

Beijing, Asia 4, 11-04-2012 41

slide-42
SLIDE 42

Identified VRC (4)

} LSGC

} MoU signed during EGI CF 2012 } One of the biggest Grid users, HEP excluded. According to EGI

accounting portal, during last year, there were jobs running in these Regions:

} Asia Pacific } Europe } Canada } Latin America (IGALC & ROC_LA) } Russia

Beijing, Asia 4, 11-04-2012 42

slide-43
SLIDE 43

Identified VRC (5)

} Coordinator: Dr. Rossella Caffo – ICCU/Ministero per i Beni e le Attività

Culturali

} These initiatives are working on coordination of policy and best practice

regarding the use of e-Infrastructures for digital cultural heritage

} INDICATE / DC-NET partners in red,collaborations in orange } MoU signed during EGI CF 2012 } This VRC offers much

different requirements than those from traditional R&D communities

} New contacts to integrate

a digital repository in Panama by means of the INDICATE e-Culture Science Gateway

} http://www.indicate-

project.org/ - http://www.dc-net.org/

Beijing, Asia 4, 11-04-2012 43

slide-44
SLIDE 44

Identified VRC (6)

} Coordinator: Dr. Fulvio Galeazzi - GARR } Design, implement, and validate a GRID-based e-Infrastructure

building upon neuGRID and relying on the Pan-European backbone GEANT and the NRENs. Over this e-Infrastructure, a service will be provided for the computer-aided extraction of diagnostic markers for Alzheimer's disease and schizophrenia from medical images

} DECIDE partners in red } (http://www.eu-decide.eu)

Beijing, Asia 4, 11-04-2012 44

slide-45
SLIDE 45

Identified VRC (7)

} Climate Change

} Conf. Role of e-Infrastructures in Climate Change } List of interested people and projects

} 4 projects identified } 20 contacts } WRF4G is being promoted inside the VRC } New development: CAM4G à Strategy to be also used in

  • ther models

} Adaptation of GridWay to SAGA (near future) } Seismology groups identified in LA, India and Italy

Beijing, Asia 4, 11-04-2012 45

slide-46
SLIDE 46

Identified VRC (8)

} SuperB

} SuperB will be a heavy flavour accelerator that will provide

complementary information to LHC, looking at rare decays with a very high luminosity electron-positron asymmetric collider

} Potential contacts world-wide by means of LHC collaborations

} Continuous contacts with them

Beijing, Asia 4, 11-04-2012 46

slide-47
SLIDE 47

.......

Science Gateway

  • App. 1
  • App. 2
  • App. N

Embedded ¡Applica-ons ¡ Administrator ¡ Power ¡User ¡ Basic ¡User ¡ Users from different

  • rganisations

having different roles and privileges

Access: the Science Gateway model (WP3, WP4)

Standard-based middleware-independent

Grid Engine

47 Beijing, Asia 4, 11-04-2012

slide-48
SLIDE 48

Conclusions

} CHAIN project has successfully agreed with other regional projects on the SG approach } The first test using SG to access different infrastructures has been very encouraging } A final test/demo is foreseen at the EGI TF 2012 in Prague in September with possibly other middlewares.

48 Beijing, Asia 4, 11-04-2012

slide-49
SLIDE 49

Beijing, Asia 4, 11-04-2012 49

Thank you for your kind attention !

Any questions ?