Title: Mirco%20Mazzucato%20Infn%20-Padova%201
1The International Grid activities in Italy
-
- and the eBusiness
- eIndustry
- eGovernment
- EScience and
- Technology
- (BIGEST) Italian Grid Initiative
- Rome MIUR March 2003
- Mirco Mazzucato
- INFN-Padova
- mirco.mazzucato_at_pd.infn.it
2Summary
- The general background
- The national grid strategy
- The national projects
- INFN Grid, FIRB, PON
- The contribution to the European Projects
- DataGrid, DataTAG...
- The international collaborations
- HICB, GLUE, World Grid, Magic....
- The new IG-BIGEST initiative for FP6
- Areas of activities
- Participants
- New EU projects for FP6
- Conclusions
3eScience and the Grid
- Italy started to develop the Grid technology and
related infrastructures in the second half of
1999 - Main motivations
- Modern fundamental Science in all domains
requires more and more global collaboration
(eScience) improve efficiency, avoid effort
duplication, combine distributed expertise .. - The scientific world is flowed by an enormous
amount of distributed raw data - HEP 10 PBytes/year in 2007
- Earth Observation Envisat 500 TBytes/year now
- Mammography screening of a country like It
200TBytes/year - Genome databases now 200TBytes. quickly growing
- ..they need to be transformed in precious piece
of knowledge - Laws of nature, new discoveries, innovative
technologies. - This challenging task need new advanced support
infrastructures - Modern Virtual Organizations,i.e.eScience
distributed teams of scientists with a common
scientific goal, should be able to share
resources, data, instruments and knowledge in
real time at national, regional (Europe) and
international level to make scientific progress
more efficient and faster - At the end of 1999 in Italy, after a carefull
evaluation, we decided that Grids were the
promising enabling technology for eScience -
4eIndustry, eGovernment and the Grid
- Modern Industries and also Governments rely more
and more on innovative choices which are
increasingly based on the cycle - Theoretical description of the problem
- Simulation/Calculations of different options
- Selection of best solution
- Realization
- and require access to large computing resources,
distributed data and collaborative methods - The Grid middleware can provide common services
that all e-Science, e-Industry and e-Government
Grid applications can use as TCP/IP provide a
common internet access protocol - Italy has fostered and aims to remain a main
player in the development of this new e-World (or
ERA) and believe that Grids will provide the
enabling technology and infrastructure - Italy spend 1 of GNP in research. Grids should
open the possibility to profit of the much larger
international scientific efforts - Insisting on pursuing the generality of any
solution for the Grid middleware should allow
large scale economies and large synergies between
different sectors - Industry and Governments can re-use most of the
services developed in the favourable e-Science
environment that are generally made available as
open source implementations enhancing in this way
the European competitiveness
5The Grid National Strategy
- The development of the Grid middleware and of the
national Grid infrastructure is being actively
pursued, since end of 1999, through a series of
national and international projects - In general Grid middleware has been whenever
possible developed through European funded
projects - DataGrid, DataTAG, GridLab....
- ..and international collaborations (with US
Globus, Condor, iVDGL, PPDG, GGF...) to allow
worldwide interoperability - GLUE, World Grid, MAGIC....
- ...but also by INFN Grid and ISUFI at national
level - ...and now by FIRB....
- The national Grid infrastructure is being
developed by national projects - Garr for the underline broadband Research
Networking - INFN Grid, FIRB WP3, S-PACI PON for Grids
- ....and integrated at international level through
EU DataGrid, DataTAG - The next step is the new HEP LHC Computing Grid
projects that will provide a common grid
environment for the collaborative work of
thousands of physicists distributed world-wide - EO, Astronomy Virtual Observatory, Gravitational
Waves are following
6Early Grid RD in Italy The INFN-GRID Project
- First national Grid project approved in Europe
beg. 2000 - Focused on the preparation of the INFN LHC comp.
infrastructure - The size of the project 20 Italian Sites, 100
people, 50 FTEs - Budget devoted to the development of the LHC
Regional Computing Centers and related
collaborative Grid infrastructure - ..but since the beginning the development of the
middleware in INFN Grid was conceived as being of
general use and has taken into account the
requirements of other sciences - Biology (PD) and Earth Observation(Esrin-ESA-Frasc
ati) - It is a successful example of collaboration
between physicists, sw engineers, computer
professionals and computer scientists (CS Dep. of
Universities of VE, PD, BO, CT, TO,) and
Italian Industries - DatamatSPA and Nice have been major contributors
in the developments of the DataGrid middleware - They have now top level expertise in Globus based
Grid middleware at European and International
Level - Comparable only with IBM
- INFN Grid has been and is the national container
for INFN to coordinate the contribution to all EU
and International Grid projects and to the GGF
standardization
7INFN Grid activities (traditional)
- INFN Grid has developed activities on all those
components of a distributed computing
infrastructure which allow in principle common
solutions . - These tradionally included
- The development of the National testbed
(including LHC Computing Centers prototypes
integrated and related services ) and the access
to the Garr Italian Research Network - The development of the grid middleware and of the
European testbed in the EU DataGrid project - The study and solution of Grid M/W
interoperability with US and the creation of the
first World Grid testbed based on de facto
standard components in the EU DataTAG project - Grid integration in HEP applications
- The INFN participation to internationl
coordination bodies - HEP Joint Technical Board (JTB)
- Common EU-US ineroperable solutions (GLUE)
- HEP Intergrid Collaboration Board (HICB)
EU-US-AP grid projects managers - The contributions to the Grid standardization
activities in the Global Grid Forum (GGF) - INFN M/W specific developments Monitoring
(Nagios), Genius portal, LFCG installation kit,
test suites....
8DataGrid Italian activitieswww.edg.org
- Financed by EU for Middleware development and EU
testbed (9.8 M) - INFN 1 M, CNR 0.3 M, Datamat 0.5 M,
IRST(Tn) 0.3 M - Italy 2.1 M for personnel for 3 years
- The main Italian contributions are in
- WP1 (Workload Management, resource discovery,
grid scheduling) - WP2(Data Management)
- WP4( Fabric Management, automatic installation
kit) - WP8 (HEP and Biology applications)
- Particular care and efforts have been devoted to
guarantee support to WP1 managed by INFN - The INFN testbed connected to the European EDG
infrastructure has constantly grown and now
includes all 20 major INFN sites located in all
major Italian Universities - EDG will provide basic middleware, prototype
testbed and services for HEP, Biology and EO and
input to the new FP6 Integrated Infrastructure
Initiative (EGEE)
9To Russia/Japan
To USA
Cern
Milano
Padova/LNL
Bologna
Torino
Roma
Cagliari
The major sites of the INFN DataGrid testbed
connected to the EU EDG testbed US and ASIA
Catania
10Main Partners CERN, INFN,UvA(NL) PPARC(UK),
INRIA(FR)
- Main focus of INFN EDT activities
- Interoperability between Grids in EU and US (WP4
managed by INFN ) - US partnershipGlobus, Condor, PPDG, iVDGL
projects and HEP experiments
11DataTAG and GLUE activitieswww.cern.ch/datatag
- Financed by EU with 0.6 M for personnel
- INFN is managing WP4 (Grid interoperability
EU-US) - WP4 has set up with the US project iVDGL, the
GLUE ( Global Laboratory Uniform Environment)
http//128.3.182.66/mtgs/ivdgl/Interoperability.ht
m activities that have now solved the
basic Grid interoperability issues between EU and
US - Middleware
- Services
- Application Grid integration support
- INFN has also a small participation in WP2
(Network measurements) - Record in transatlantic transfer speed have been
recently achieved - The development of common solutions (or at least
agreed and introperable) is a natural interest of
European Applications, in particular those
concerned by EU-US collaboration. - Most of the HEP experiments have strong US
components together with EU ones - and many other sciences now can profit of these
solutions EO, AVO... - INFN DataTAG with iVDGL(US) have provided the
first interoperable EDG(EU) and VDT(US)
middleware - Demo at IST2002 an SC2002
12Grid aspects covered by EDG (in red INFN main
developments)
VO servers and VOMS LDAP directory for mapping users (with certificates) to correct VO Storage Element Grid-aware storage area, situated close to a CE
User Interface Submit monitor jobs, retrieve output Replica Manager Replicates data to one or more CEs
Job Submission Service Manages submission of jobs to Res. Broker Replica Catalog Keeps track of multiple data files replicated on different CEs
Information index Provides info about grid resources via GIIS/GRIS hierarchy Information Monitoring Provides info on resource utilization performance
Resource Broker Uses Info Index to discover select resources based on job requirements Grid Fabric Mgmt Configure, installs maintains grid sw packages and environ.
Logging and Bookkeeping Collects resource usage job status Network performance, security and monitoring Provides efficient network transport, security bandwidth monitoring
Computing Element Gatekeeper to a grid computing resource Testbed admin. Certificate auth.,user reg., usage policy etc.
13The INFN development of middleware
www.infn.it/grid
- Progress on middleware developed by INFN Grid was
unexpected - Genius Portal is becoming almost a standard
- It is used in all EDG demos and is being adopted
by several application (also industrial) and in
US - The Grid Monitoring system is now adopted by the
LHC Computing Grid project - The installation configuration kit is now a
baseline - The Virtual Organization Mapping System is
adopted by LCG and by several US projects in
alternative to Globus CAS - Their development is continuing and their usage
by applications in a wider eScience context is
expanding - Tutorials and contacts with Industries are
increasing - Last Grid tutorial organized in Torino December
5th and 6th(50 people) - Next Grid tutorial in Catania
14The Grid FIRB Proposal(see talk of M.Vanneschi)
- In FIRB INFN will study and prototype a national
Grid Operation Center (GOC) in Italy to support
the creation of a national grid infrastructure - The generalization of the infrastructure support
is a model successfully established with the
research network (GARR, GEANT..) - GOC will support several Italian Sciences
applications and eventually the operation of this
Italian infrastructure in the context of the new
European Infrastructure project EGEE - EU eScience will count on national GOCs (like the
FIRB or UK ones) to guarantee 24x24x7 operations
of the Grid infrastructure as counts on GEANT and
GARR for the network operation - The Italian eScience Grid infrastructure
currently support - Astrophysics
- Biology
- Computational Chemistry
- Geophysics
- Earth Observation
- Integration with international Grid
infrastructures occurs via EDG,EDT,US proj.
15Some issues
- Grid M/W is being developed by several projects
- in US Globus, Condor, PPDG, GriPhyN, iVDGL.....
- in EU DataGrid, DataTAG, Cross Grid, Gridlab...
- Grid M/W is now being adopted by a variety of
Science sectors - Current M/W software is typical of prototype
development. It is working but it is far from
having developed robust solutions and well
engineered production quality components assuring
long term survival and easy evolution - See e.g. Globus issues
- The Grid middleware is deployed in several
testbeds. This requires large manpower,
duplications of efforts etc and all national Grid
programs are looking for sinergies and common
deployment and operation support - Policies for resource sharing within comunities
having common objectives (VO), security issues,
accounting etc start to be generally addressed
only now - FP6 need to allow a definite step forward and
increase European competitiveness - Grids are now seen by many governments as a
Science and Industry development enabling factor
and large funds are made available also at
national level
16Million
Now US proposal for research cyber-infrastructur
e 1020 M Japan Grid 120M EU FP6 300 M for
Geant and Grids
Thanks to D. Laforenza
17The Italian Grid for Business, Industry,
Government, EScienceTechnology (IG-BIGEST)
- It is a new national initiative leveraging from
INFN, FIRB and PON projects and aiming at
coordinating all Italian efforts towards a
national grid program and a coordinated Italian
effective participation to FP6 and international
grid projects - IG-BIGEST includes all Italian Sciences
Institutions, major computing centers and many
Industries ready to invest in early tests - The IG-BIGEST main objectives
- Create a general EU grid infrastructure for
eScience integrating all available EU national
infrastructures and open to industry early test .
Make current Grid middleware robust and fault
tolerant for this production infrastructure to
allow to start EU eScience - -gtResearch Infrastructure
- Promote RD on open Grid issues
- Support development of specific components and
usage of Grids by major applications in
e-Science, eIndustry and eGovernment - -gtIST(Complex Problem Solving) or specific
programs (e.g. eHealth)
18The 1rst IG-BIGEST FP6 ProposalEGEE
- Consortium Partners, coordinated by CERN, cover
20 European countries - It is an Integrated Infrastructure Initiative
for the RI call - 45-47 Millions available
- Initial EGEE request 35 M for 2 years
- But it is a 4 year project aiming at 100M
- Preliminar cost sharing
- Overall management 7
- Quality assurance 3
- Interface to applications 10 for requirements
and validation - Middleware re-engineering 30 to consolidate
current M/W - Grid deployment and operation centers 50
- 1rst EGEE grid infrastructure size gt50 major EU
sites,6-8 operation and call centers, POP on
reference sites in each minor country - Strong cooperation with Geant and successor
- but EGEE will not allow to make M/W RD and
developments in many area where this is badly
needed - Neither to adress the development of higher level
specific Industrial, Governments or even science
and general application components - -gtIST
19What after EGEE in FP6?
- IG-BIGEST need to address with the EU partners
major issues present in current grid middleware
and more long term developments and can provide
expertise in all areas - Programming Tools and Environment
- Security AAAA in particular Audit and Accounting
- Aim at the development of services that will
bring current Grid security suitable for
Business, Industries and Government Virtual
Organizations - Data Intensive Services
- Extension of current Data Grid Services to cover
requirements of extended classes of Science e.g.
genomic databases, medical application etc - Knowledge discovery Services
- Grid Portals
- Resource Management (Brokering, Scheduling and
Monitoring) - IG-BIGEST will support the development of the
CORE Grid proposal for the next IST call,
coordinated by INRIA(T.Priol) - Issues participation of Industries,
applications.
20The HEAlthgrid VENture(HEAVEN) Network of
Excellence
- Develop GRID compliant e-Health application on
the European scale - On 5 levels Molecular, Cellular, Organ,
Patient, Population - Integrate them, whenever possible, into a
European Grid for Health - Partners CEA, CNRS, EPCC,IG-BIGEST/INFN,
- SIEMENS, IBM,... Italian Industry (in
negotiation) - Timescale 5 years
- Researchers 70 FTE
- Budget 7 MEuros for eHealth call closing April
25
21How will HEAVEN Work?
Feedback
22The GPCALMA Pilot Application
telediagnosis for breast cancer screening
Data Collection Centre
Diagnostic Centre
Data MetaData Catalogue
1 - Data Collection
4 - Transfer Selected Data
3 - Run CAD remotely
5 - Interactive Diagnosis
2 - Data Registration
CAD selection to minimize data transfers for
quick diagnosis
23Additional FP6 proposals for eScience
- Many IG-BIGEST applications are part of large
European consortia who need to develop their
specific Grid higher level services. These
include - Astrophysics
- Biology
- Computational Chemistry
- Geophysics
- Climate
- Earth Observation
- eLearning
- Digital Libraries
- Need to find appropriate EU calls to extend the
European Research Area to cover these
applications to allow them to make a critical
mass at EU level to compete in the world
24The IG-BIGEST Industrial activities
- IG-BIGEST collaboration with Industries is
actively ongoing along 3 lines - Participation of solution providers in FP6
proposal - E.g. Datamat, Nice , a couple more under
discussion in EGEE and HEAVEN - Develop a specific higher level of services on
top of the basic Grid layer to increase the
productivity of specific sectors - Exploit current INFN and IG-BIGEST expertise on
Grids - Involve Industries in early tests using existing
Grid infrastructure available in Italy by INFN,
FIRB and PON and in future by EGEE (we hope...)
25Current case of studyGrids in Industrial
Engineering
- Involve major Italian Industries dealing with the
engineering in the Automotive sectors and some
civil engineering companies - We have started regular meetings and mail
exchanges to identify potential interest of Grids
- Automotive
- grid components to allow collaborative work of
geographically separate groups working in product
development with interconnected data - Identify integration issues of currently most
used tools in CAE, CAD, Virtual Reality and
Rendering - Identify issue in standardization of data
exchange - Identify data transfer requirements
- Identify interface and specific requirement
towards basic grid services - Engineering
- Grid components to allow the outsourcing of most
intensive computational activities alowing them
to pay what they really use - Simulation, finite element analysis, rendering
and virtual reality - This is particularly interesting for SMEs whose
market results are increasingly dependent on the
capability to develop simulations and Virtual
Reality demos - All these are very good candidates for IP in
IST-CPS - In near future we will start to address issues
related to the Farmaceutical industries of
IG-BIGESt and SMEs and Real time grids for
apparatus control and catastrophy management, NMI
26Conclusions
- Italian Science and Industries are making a
coordinated effort to be active part in the
construction of the European Research Area - IG-BIGEST is the national initiative aiming at
supporting the development of eScience,
eGovernment and eBusiness in Italy and Europe
aiming at giving Europe the edge on Grid Research
and Industrial outreach - IG-BIGEST will contribute to all aspects of Grids
development RD, S/W Engineering, deployment of
infrastructure, operation, development of
specific components for applications for
eScience, eBusiness, eIndustryGovernment,
eHealth... - We count on IST-CPS Unit support to align
activities and to develop adequate partnerships
and identify possibilities and most convenient
Instruments