Title: EU DataGrid Project Status and perspective for EU FP6
1EU DataGrid ProjectStatus and perspective for EU
FP6
- Fabrizio Gagliardi
- Project Leader, CERN
- Fabrizio.Gagliardi_at_cern.ch
2Outline
- Experience of first year and main issues
- Results of first project review and follow up
- Testbed status
- Relationships with other Grid projects
- Future directions and EU FP6
3Highlights (from First Project Review on March
1st)
- The project is up and running!
- All 21 partners are now contributing at
contractual level - Project administrative and managerial structure
established with minimum resources - All EU deliverables (40, gt2000 pages) submitted
in time for the review according to the contract
technical annex - First test bed delivered with real production
demos
4Major achievements (from First Project Review on
March 1st)
- Core middleware group developed innovative S/W
also exported to US (GDMP and resource broker) - Excellent collaboration with US Globus and Condor
developments - Close collaboration with similar US activities
(PPDG, GriPhyN/iVDGL) - Large community of enthusiastic, dedicated
scientists (mostly unfunded) - End users involved in all stages of the project
(requirements definition, architecture, S/W
integration, deployment, validation and tests) - Unfunded staff effort about twice the EU funded
(voluntary participation from Portugal, Ireland,
Russia Denmark both in M/W and in the test bed)
5more achievements, continued (from First
Project Review on March 1st)
- Good relations to industry (IR Forum and WP11)
- Seed funds for national Grid projects,
coordinator and initiator of other projects
(DataTAG, CrossGrid, GridSTART, security
proposal) - Initiator and active participant in GGF,
Intergrid, EIROForum Grid WG, OCDE interest to
start a WG, exploratory work in Asian Pacific and
South America - PM in GGF GFAC, 2 members in GGF WGSG, 2 GGF WG
chairs - 2 GGF WGs and 1 GGF research group being proposed
on networking by WP7 - Pioneering role (EU Grid flagship project) first
opportunity to work on Grid for ESA with
fostering effect of internal Grid activity - Prototype use of national RNs for Grid deployment
(building Grids of Grids)
6more achievements, continued(from First Project
Review on March 1st)
- Collaboration with Geant/Dante (WP7 and DataTAG)
- Collaboration agreement established in June 2001
- Technical Proposal
- Test equipment located in three GEANT PoPs
- Analysis of TCP performance over a wide area
high-speed network - Demonstration of high-throughput data flows over
long time scales - Pilot of the IP Premium service including
end-to-end tests with biology use-case
applications (WP10) - Exploration of use of reliable multicast for file
replication - Better understanding of how high-speed
infrastructures react to traffic load - Support in the definition of TCP-based services
- Integration and sharing of the existing GEANT
monitoring with end-to-end DataGRID network
monitoring data
7Issues (from First Project Review on March 1st)
- Overall under funded project relative to the
size, goals and number of partners - New technology, 21 partners and diversity of
applications and communities (CS, PP, EO and Bio) - CERN (and some other partners) original funding
assumptions no longer valid - Project office under staffed and below minimum
operating budget one project secretary, one
technical coordinator and one quality engineer
marginally funded (0 travel budget and personnel
cost higher than expected) - Overall the project within budget, but ½ of
partners overspent travel and 2 largely overspent
first year budget - Almost completely unfunded resources for test bed
equipment, networking and security
8Experience of first year and main issues
- The project is still up and running but,
- Needs quickly to implement EU recommendations
- Needs to understand how to progress with the test
beds by mastering complexity, emerging standards
and support issues - Needs to adapt and connect to proliferation of
Grid projects world-wide - Needs to consolidate user communities (VOs)
- Needs to keep cost under control
- Needs to develop an after EDG strategy (LCG, FP6?)
9First project review follow up
- Mobil agents explained why we dont need them
- Security features and cross-authentication
explained our plans and EoI for a security
project I FP6 submitted by a subset of partners - Integration of new resources in the testbed
explained WP6 approach and WP4 plans for a light
weight installation procedure - More flexible fabric installation procedures and
requirements WP4 is following this - Interface to Globus and participation towards a
common evolution main issue under active
discussion (also in the context of GGF, Glue etc.)
10First project review follow up
- Scalability of resource scheduling architecture
needs test-plan implemented and test group
activated - Improve data replica scheme being address for TB
2 - Improve test use cases being discussed also in
collaboration with DataTAG - Involve more EO applications being addressed by
WP9 also as part of their review - Expand the use of Grid in EO being addressed by
WP9 also as part of their review - Improve project visibility and publish the
content of EU deliverables being addressed by
WP11 and WP12 (Information officer hired)
11First project review follow up
- ESA review to address their role in the project
and their expenditure profile carried out
successfully last week in Frascati - PO short-coming addressed by replacing second
project admin. - ESTEDI project contacted by WP2, publication of
content of deliverables reviewed weekly - GGF participation and GridStart
- Involvment of industrial partners to be better
documented for next review - EDG license agreed
- WP11 should improve quantitative measurement of
dissemination, web content etc.
12Grid aspects covered by EDG testbed 1
13Testbeds
- Development
- Small set of sites (approx 5) with small clusters
of PCs for development purposes to test new
versions of the EDG middleware, configurations
etc. - Not stable constantly changing
- Currently testing EDG release 1.2
- Production
- More stable environment for use by application
groups - more sites
- more nodes per site (grow to meaningful size at
major centres) - more users per VO
- Currently running EDG release 1.1
- Hope to install EDG 1.2 during June
14Software Release Procedure
- Coordination meeting
- Gather feedback on previous release
- Review plan for next release
- WP meeting
- Take basic plan and clarify effort/people/dependen
cies - Sw development
- Performed by WPs in dispersed institutes and run
unit tests - Software integration
- Performed by WP6 on frozen sw
- Integration tests run
- Acceptance tests
- Performed by Loose Cannons et al.
- Roll-out
- Present sw to application groups
- Deploy on testbed
themed tech. meets
testbed 1 Dec 11 2001 100 participants
Software release Plan http//edms.cern.ch/document
/333297
Roll-out. meeting
15EDG platforms
- EDG 1.1 (now) supports Linux RedHat 6.2
- Globus2beta21
- EDG 1.4 (July) will support Linux RedHat 7.2
6.2 - Globus2.x
- EDG 2.0 (October) will support Linux RedHat 7.2
6.2 - Globus2.x
- In 2002 we will start port to Solaris as 3 steps
- WP1 UI tools
- Other middleware
- WP4 fabric mgmt tools
16Release Plan details
- Current release EDG 1.1.4
- Deployed on testbed under RedHat 6.2
- Finalising testing of EDG 1.2
- WP1 APIs (see Massimos talk)
- WP2 GDMP 3.0 (see Heinzs talk)
- WP5 secure access to MSS (Castor)
- EDG 1.3 (end June)
- Build using autobuild tools to ease future
porting - INTERNAL RELEASE will not be deployed on
production testbed
- EDG 1.4 (end July)
- Support RH 6.2 7.2
- Basic support for interactive jobs
- Use MDS 2.2 with first GLUE schema
- EDG 2.0 (Oct)
- Still based on Globus 2.x (pre-OGSA)
- Use updated GLUE schema
- Job partitioning check-pointing
- Advanced reservation/co-allocation
- Integration of Condor DAGman
See http//edms.cern.ch/document/333297 for
further details
17John Gordons Testbed Trident
iVDGL Datagrid
LCG
Les Robertson, CERN
- Grid projects -
- responsible for testbeds on which developers
can build and test their software - testbeds for demos, Beta testing, scheduled
data challenges
LCG - responsible for operating a production
Grid - 24 X 7 service - as easy to use
as LXBATCH
18Issues when adding new sites to the testbed
- EDG is currently setting-up procedures explaining
how to add new sites - Variations already tested with Taiwan and Romania
- Step-by-step instructions produced which we
expect to become simpler over time - Need to clarify the minimum requirements for a
site to become a member of the testbed - A number of regular tasks have to be performed by
the sites administrators - A maximum delay needs to be defined for
responding to requests/problems if the testbed is
to run efficiently - Sites from new countries have to identify/create
a supporting CA - Since CAs need mutual trust this could lead to an
explosion of inspection activities - Some tasks will fall on the people responsible
for managing the VOs - HEP experiment secretariats already perform some
level of authentication of their institutes and
members. How an we get some leverage from this?
19GLUE
Antonia Ghiselli, INFN
- http//www.hicb.org/glue/GLUE-v0.04.doc
- Grid Laboratory Uniform Environment (GLUE) focus
is on interoperability between Grid software
projects. - GLUE is a collaboration of grid projects with
effort and management through iVDGL and DataTAG. - GLUE will report to and get guidance from HICB
and JTB. - GLUE components are finalized to HEP experiments.
- To date participating experiments are Alice,
Atlas, BaBAR, CDF, CMS, D0, (LHCb?) - and projectsApGrid, CrossGrid,DataTAG, DataGrid,
GriPhyN, IN2P3, INFN-Grid, iVDGL, LCG, GridPP,
PPDG
20Interaction with PPDG/GriPhyN/iVDGL
- Work with dataTAG via the InterGrid (GLUE) to
investigate inter-operability of US and EU grids - Authentication infrastructure
- perform cross organizational authentication
- DOE/EDG CAs mutual trust for Certs
- Unified service discovery and information
infrastructure - discover the existence and configuration of
services offered by the testbeds - Mature draft of information schema developed
- Data movement infrastructure
- move data from storage services operated by one
organization to another - Authorization services
- perform some level of cross organization,
community based authorization - Computational services
- coordinate computation across organizations to
allow submission of jobs in EU to run on US sites
and vice versa - These steps have to matched with EDG software
release plan for deployment
Done
Started
21The LHC Computing Grid Project
Goal Prepare and deploy the LHC computing
environment
- applications - tools, frameworks, environment,
persistency - computing system ? services
- cluster ? automated fabric
- collaborating computer centres ? grid
- CERN-centric analysis ? global analysis
environment - foster collaboration, coherence of LHC
regional computing centres - central role of data challenges
This is not yet another grid technology project
it is a grid deployment project
Les Robertson, CERN
22The futureOpen Grid Services Architecture
- Increasingly popular standards-based framework
for accessing network applications - W3C standardization Microsoft, IBM, Sun, others
- WSDL Web Services Description Language
- Interface Definition Language for Web services
- SOAP Simple Object Access Protocol
- XML-based RPC protocol common WSDL target
- WS-Inspection
- Conventions for locating service descriptions
- UDDI Universal Desc., Discovery, Integration
- Directory for Web services
- Service orientation to virtualize resources
- Everything is a service
- From Web services
- Standard interface definition mechanisms
multiple protocol bindings, local/remote
transparency - From Grids
- Service semantics, reliability and security
models - Lifecycle management, discovery, other services
- Multiple hosting environments
- C, J2EE, .NET,
Globus 3.0 (OGSA compliant) end 2002 Industry
involvement IBM, MicroSoft, Sun
23Future Plans
- Collaborate with closely related activities
(Glues, iVDGL, EDT, X) to deliver the basis of a
production testbed to LCG - Extend collaboration with US based LHC activity
via EDT/iVDGL - Prepare for second test bed in autumn 2002
- Promote early standards adoption with
participation to GGF and other international
bodies
24Conclusion
- Is there life after EDG?
- EoI submitted for a large III (Integrated
Infrastructure Initiative) in coordination with
European RNs (RN Geant follow on) - 300 M Euros total budget, 100 M Euros EU funding
- Overwhelming interest from European Grid
community and industry - Will our community be able to deliver up to these
very high expectations? - What is LCG role into this? and the other HEP
national Grid activities? - Interesting discussions today and in the year
ahead
25Learn more about Grids DataGrid
Programme includes Grid Lectures by Ian
Foster Carl Kesselman Hands-on tutorial DataGrid
Apply now via web http//csc.web.cern.ch/CSC/ Plac
es are limited
CERN School of Computing 2002
Vico Equense, Italy, 15-28 September 2002 The
2002 CERN School of Computing is organised by
CERN, with the Institute of Composite and
Biomedical Materials, National Research Council,
Naples, Italy.