DataTAG project presentation -Internet2 Spring meeting (Arlington)

1
The DataTAG Project 5th EU DataGrid
Conference Pilicsaba, Hungary Olivier H. Martin
http//www.datatag.org
2
The project

• CERN is the project coordinator
• Contacts Olivier Martin
• European partners INFN (IT), PPARC (UK),
  University of Amsterdam (NL) and, since July
  2002, INRIA (FR)
• Budget 3.98 MEUR
• Start date January, 1, 2002
• Duration 2 years (aligned on DataGrid)
• Funded manpower 15 people/year

3
Goals

• End to end Gigabit Ethernet performance using
  innovative high performance transport protocols.
• Assess experiment inter-domain QoS and
  bandwidth reservation techniques.
• Interoperability between major GRID projects
  middleware/testbeds in Europe and North America.
• DataGrid, possibly other EU funded Grid projects
• PPDG, GriPhyN, iVDGL, Teragrid (USA)
• LCG

4
In a nutshell

• Two main focus
• Grid related network research (WP2, WP3)
• 2.5 Gbps transatlantic lambda between CERN
  (Geneva) and StarLight (Chicago) (WP1)
• Dedicated to research (no production traffic)
• Very unique multi-vendor testbed with layer 2 and
  layer 3 capabilities
• in effect, a distributed transatlantic Internet
  Exchange Point.
• Open to other European Grid projects for
  demonstrations
• Interoperability between European and US Grids
  (WP4)
• Middleware integration and coexistency -
  Applications
• GLUE integration/standardization
• GLUE testbed and demo

5
The 2.5 Gbps transatlantic circuit

• Why yet another 2.5 Gbps transatlantic circuit?
• Most existing or planned 2.5 Gbps transatlantic
  circuits are for production, i.e. not suitable
  for advanced networking experiments requiring
  operational flexibility, e.g.
• deploying new equipment (routers, G-MPLS capable
  multiplexers),
• activating new functionality (QoS, MPLS,
  distributed VLAN)
• The only known exception to date is the Surfnet
  circuit between Amsterdam Chicago (Starlight)

6
DataTAG connectivity
NewYork
Abilene
Canarie
32.5G
STAR-LIGHT
ESNET
CERN
MREN
Major 2.5 Gbps circuits between Europe USA
7
What we have achieved

• The 2.5Gbps transatlantic circuit
• Operational since August 20 (delivered by
  Deutsche Telekom/T-Systems )
• Routing plan developed (access across GEANT)
• As expected, the full DataTAG testbed could not
  be readily deployed because of late ordering but
  also delivery delays, therefore the circuit will
  initially be connected to Cisco 7600 routers
• Full testbed to be deployed by the end of October
• Six high performance PCs will be deployed on each
  side of the high performance circuit in the
  coming week(s). The full layer2/layer3 testbed
  should be ready by the end of October

8
DataTAG/WP4 framework and
relationships
HEP applications, Other experiments
Integration
HICB/HIJTB
Interoperability standardization
GLUE
9
Status of GLUE activities in DataTAG

• Resource Discovery and GLUE schema
• Authorization services from VO LDAP to VOMS
• Common software deployment procedures
• IST2002 and SC2002 joint EU-US demos

Interoperability between Grid domains for core
grid services, coexistence of
optimization/collective services

• Data Movement and Description
• Job Submission Services

10
Resource Discovery and GLUE schema
Computing Resources Structure Description
Entry point into queueing system
Container grouping subcluster or nodes
Homogeneous collection of nodes
Physical computing nodes
11
GIIS giis.ivdgl.org mds-vo-nameglue
Gatekeeper Padova-site
GIIS edt004.cnaf.infn.it Mds-vo-nameDatatag
LSF
Resource Broker
Gatekeeper US-CMS
GIIS giis.ivdgl.org mds-vo-nameivdgl-glue
Gatekeeper grid006f.cnaf.infn.it
Gatekeeper edt004.cnaf.infn.it
Condor
Gatekeeper US-ATLAS
WN1 edt001.cnaf.infn.it WN2 edt002.cnaf.infn.it
Computing Element-1 PBS
Computing Element -2 Fork/pbs
Gatekeeper
LSF
dc-user.isi.edu
hamachi.cs.uchicago.edu
rod.mcs.anl.gov
DataTAG
Job manager Fork
iVDGL
12
Common software deployment procedures
13
IST2002 and SC2002 joint EU-US demos
14
Conclusions

• Personnel recruited 21FTE.
• Network testbed in place by October 2002
• Active collaboration with GEANT
• Active collaboration with US/iVDGL for Grid
  Integration
• First version of GLUE schema for resource
  discovery in place
• Working on common authorization solutions
• First evaluation of software deployment tools
• First interoperability tests on heterogeneous
  transatlantic testbeds
• Demos organization
• Questions to EDT manager Olivier.Martin_at_cern.ch