The EU DataTAG Project

The EU DataTAG Project

Presented at the 4Presented at the 4thth DataGrid DataGrid Conference Conference

88thth March 2002, PARIS March 2002, PARIS

Olivier H. MartinOlivier H. MartinCERN - IT DivisionCERN - IT Division

http://www.pparc.ac.uk/

http://www.infn.it/

http://www.cern.ch/

04/22/23 The EU DataTAG Project Olivier H. Martin (2)

The project The project European partners: INFN (IT), PPARC (UK), European partners: INFN (IT), PPARC (UK),

University of Amsterdam (NL) and CERN, as University of Amsterdam (NL) and CERN, as project coordinator. project coordinator.

INRIA (FR) expected to join during 1H2002. INRIA (FR) expected to join during 1H2002. ESA/ESRIN (IT) will provide Earth ESA/ESRIN (IT) will provide Earth

Observation demos together with NASA.Observation demos together with NASA. Budget: 3.98 MEURBudget: 3.98 MEUR Start date: January, 1, 2002 Start date: January, 1, 2002 Duration: 2 years (aligned on DataGrid)Duration: 2 years (aligned on DataGrid) Funded manpower: ~ 15 persons/yearFunded manpower: ~ 15 persons/year


US Funding & US Funding & collaborationscollaborations

US NSF support through the existing US NSF support through the existing collaborative agreement with CERN (Eurolink collaborative agreement with CERN (Eurolink award).award).

US DoE support through the CERN-USA line US DoE support through the CERN-USA line consortium.consortium.

Significant contributions to the DataTAG workplan have been made by Jason Leigh (EVL@University of Illinois), Joel Mambretti (Northwestern University), Brian Tierney (LBNL).

Strong collaborations already in place with ANL, Caltech, FNAL, SLAC, University of Michigan, as well as Internet2 and ESnet.


In a nutshellIn a nutshell

Two main focus:Two main focus: Grid related network research (WP2, WP3) Interoperability between European and US

Grids (WP4) 2.5 Gbps transatlantic lambda between 2.5 Gbps transatlantic lambda between

CERN (Geneva) and StarLight (Chicago) CERN (Geneva) and StarLight (Chicago) around July 2002.around July 2002.

Dedicated to research (no production traffic) In principle open to other EU Grid projects as

well as ESA for demonstrations


Goals & outcomesGoals & outcomes• End to end Gigabit Ethernet End to end Gigabit Ethernet

performance using innovative high performance using innovative high performance transport protocols.performance transport protocols.

• Assess inter-domain QoS and Assess inter-domain QoS and bandwidth reservation techniques.bandwidth reservation techniques.

• Interoperability between some major Interoperability between some major GRID projects in Europe and North GRID projects in Europe and North America America

DataGrid, possibly other EU funded Grid projects PPDG, GriPhyN, Teragrid, iVDGL (USA)

NLSURFnet

CERN

UKSuperJANET4

Abilene

ESNET

MREN

ITGARR-B

GEANT

NewYork

STAR-TAP

STAR-LIGHT

DataTAG project

Major 2.5 Gbps circuits between Europe & USA


Project Project positioningpositioning

Why yet another 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, which makes them basically not suitable for advanced networking experiments that require a great deal of operational flexibility in order to investigate new application driven network services, 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) Concerns:Concerns:

How far beyond Starlight can DataTAG extend? How fast will US research network infrastructure match that

of Europe!


The STAR LIGHTThe STAR LIGHT Next generation STAR TAP with the

following main distinguishing features: Neutral location (Northwestern University) 1/10 Gigabit Ethernet based Multiple local loop providers Optical switches for advanced experiments

The STAR LIGHT will provide 2*622 Mbps ATM connection to the STAR TAP

Started in July 2001 Also hosting other advanced networking

projects in Chicago & State of Illinois N.B. Many European Internet Exchanges

Points have already been deployed along the same lines.


Major Grid networking Major Grid networking issuesissues

QoS (Quality of Service) QoS (Quality of Service) still largely unresolved on a wide scale because of complexity of deployment

TCP/IP performance over high bandwidth, long distance networksTCP/IP performance over high bandwidth, long distance networks The loss of a single packet will affect a 10Gbps stream with 200ms RTT (round

trip time) for 5 hours. During that time the average throughput will be 7.5 Gbps.

On the 2.5Gbps DataTAG circuit with 100ms RTT, this translates to 38 minutes recovery time, during that time the average throughput will be 1.875Gbps.

Line Error ratesLine Error rates A 2.5 Gbps circuit can absorb 0.2 Million packets/second Bit error rates of 10E-9 means one packet loss every 250 mseconds Bit error rates of 10E-11 means one packet loss every 25 seconds

End to end performance in the presence of firewallsEnd to end performance in the presence of firewalls There is a lack of products, can we rely on products becoming available or

should a new architecture be evolved? Evolution of LAN infrastructure to 1Gbps/10Gbps Evolution of LAN infrastructure to 1Gbps/10Gbps

Uniform end to end bandwidth


Single stream vs Multiple streams Single stream vs Multiple streams effect of a single packet loss (e.g. link error, buffer overflow)effect of a single packet loss (e.g. link error, buffer overflow)

Time

Thro

ughp

ut G

bps

10

5

7 5

2.5

Avg. 4.375 Gbps

Avg. 3.75 Gbps

Avg. 7.5 Gbps

Avg. 6.25 Gbps

Streams/Throughput 10 5

1 7.5 4.375

2 9.375

T = 2.37 hours! (RTT=200msec, MSS=1500B)

T T T T


Workplan (1)Workplan (1) WP1: Provisioning & Operations (P. Moroni/CERN)WP1: Provisioning & Operations (P. Moroni/CERN)

Will be done in cooperation with DANTE & National Research & Education Networks (NREN)

Two main issues: Procurement (largely done already for what concerns the circuit,

equipment still to be decided). Routing, how can the DataTAG partners access the DataTAG

circuit across GEANT and their national network? Funded participants: CERN(1FTE), INFN (0.5FTE)

WP5: Information dissemination and exploitation (CERN)WP5: Information dissemination and exploitation (CERN) Funded participants: CERN(0.5FTE)

WP6: Project management (CERN)WP6: Project management (CERN) Funded participants: CERN(2FTE)


Workplan (2)Workplan (2) WP2: High Performance Networking (Robin WP2: High Performance Networking (Robin

Tasker/PPARC)Tasker/PPARC) High performance Transport

tcp/ip performance over large bandwidth*delay networks

Alternative transport solutions using:Modified TCP/IP stackUDP based transport conceptually similar to rate based

TCP End to end inter-domain QoS Advance network resource reservation

Funded participantsFunded participants: PPARC (2FTE), INFN (2FTE), UvA : PPARC (2FTE), INFN (2FTE), UvA (1FTE), CERN(1FTE)(1FTE), CERN(1FTE)


Workplan (3)Workplan (3) WP3: Bulk Data Transfer & Application WP3: Bulk Data Transfer & Application

performance monitoring (Cees deLaat/UvA)performance monitoring (Cees deLaat/UvA) Performance validation End to end user performance

Validation Monitoring Optimization

Application performance Netlogger

Funded participantsFunded participants: UvA (2FTE), CERN(0.6FTE): UvA (2FTE), CERN(0.6FTE)


WP4 WorkplanWP4 Workplan (Antonia Ghiselli & Cristina Vistoli / INFN)(Antonia Ghiselli & Cristina Vistoli / INFN)

Main Subject: Main Subject: Interoperability between EU and US Grids services Interoperability between EU and US Grids services

from DataGrid, GriPhyN, PPDG and in from DataGrid, GriPhyN, PPDG and in collaboration with iVDGL, for the HEP applications.collaboration with iVDGL, for the HEP applications.

Objectives:Objectives: Produce an assessment of interoperability solutions Provide test environment to LHC Applications to

extend existing use-cases to test interoperability of the grid components

Provide input to a common Grid LHC architecture Plan EU-US Integrated grid deployment

Funded participantsFunded participants: INFN (6FTE), PPARC (1FTE), UvA : INFN (6FTE), PPARC (1FTE), UvA (1FTE)(1FTE)


WP4 TasksWP4 TasksAssuming the same grid basic services (gram,gsi,gris)Assuming the same grid basic services (gram,gsi,gris)between the differen grid projects, the main issues are:between the differen grid projects, the main issues are: 4.1 resource discovery, coord. C.Vistoli4.1 resource discovery, coord. C.Vistoli 4.2 security/authorization, coord. R.Cecchini 4.2 security/authorization, coord. R.Cecchini 4.3 interoperability of collective services 4.3 interoperability of collective services

between EU-US grid domains, coord. F.Donno between EU-US grid domains, coord. F.Donno 4.4 test applications, contact people from 4.4 test applications, contact people from

each application :each application : Atlas / L.Perini, CMS / C.Grandi, Atlas / L.Perini, CMS / C.Grandi, Alice / P.CerelloAlice / P.Cerello


DataTAG/WP4 frameworkDataTAG/WP4 framework and relationships and relationships

Grid projects:

GridInteroperabilityActivities:

Applications:

DataGrid PPDG Griphyn Globus Condor

DataTAG/WP4 iVDGLiVDGL

LCG

HICB/HIJTB GGF

Integration

LHC experiments CDF Babar ESA …..

stardardization

…..

…..

input feedback

Proposals


WP4.1 - Resource DiscoveryWP4.1 - Resource DiscoveryObjectivesObjectives

Enabling an interoperable system that allows for Enabling an interoperable system that allows for the discovery and access of the Grid services the discovery and access of the Grid services available at participant sites of all Grid domains, available at participant sites of all Grid domains, in particular between EU and US Grids.in particular between EU and US Grids.

Compatibility of the Resource Discovery System Compatibility of the Resource Discovery System with the existent components/services of the with the existent components/services of the available GRID systems.available GRID systems.


Task 4.1 Time PlanTask 4.1 Time Plan Reference agreement document on resource Reference agreement document on resource

discovery schemadiscovery schemaby 31st of May 2002by 31st of May 2002

““INTERGRID VO” MDS testINTERGRID VO” MDS testby 31st of July 2002by 31st of July 2002

Evaluation of the interoperability of multiple Evaluation of the interoperability of multiple Resource Discovery Systems (FTree, MDS, Resource Discovery Systems (FTree, MDS, etc…)etc…) by 30th of September 2002by 30th of September 2002

Network ElementNetwork Elementby 31st of December 2002by 31st of December 2002

Impact of the new Web Services TechnologyImpact of the new Web Services Technologyby 31st of June 2003by 31st of June 2003

Identify missing components.Identify missing components.by 31st of June 2003by 31st of June 2003

Final deployment.Final deployment.by 31st of December 2003by 31st of December 2003


Identify Authentication, Identify Authentication, Authorization and Accounting (AAA) Authorization and Accounting (AAA) mechanisms allowing mechanisms allowing interoperability between gridsinteroperability between grids

Compatibility of the AAA Compatibility of the AAA mechanisms mechanisms with the existing with the existing components/services of the components/services of the available GRID systems.available GRID systems.

WP4.2 - ObjectivesWP4.2 - Objectives


Reference documentReference document Issues

Minimum requirements for DataTAG CA’s; Analysis of available authorization tools

and policy languages and their suitability (in cooperation with the DataGrid Authorization WG);

Mapping of the policies of the VO domains;

Information exchange protocol between the authorization systems;

Feasibility study of an accounting system (in cooperation with the DataGrid WP1);

First draft: 31 July 2002Final version: 30 April 2003

DeploymentDeploymentFirst: 30 September 2002Final: 31 December 2003

Task 4.2 Time PlanTask 4.2 Time Plan


Identify grid elements in EU and US grid Identify grid elements in EU and US grid projects, Identify common components projects, Identify common components in the testbeds used by the HEP in the testbeds used by the HEP experiments for semi-production experiments for semi-production activities in EU and US activities in EU and US and classify and classify them in an architectural framework.them in an architectural framework.

Plan and Setup “InterGrid VO” Plan and Setup “InterGrid VO” environment with common EU-US environment with common EU-US services.services.

Deployment of an EU-US VO domain in Deployment of an EU-US VO domain in collaboration with iVDGL.collaboration with iVDGL.

WP4.3 / WP4.4 - WP4.3 / WP4.4 - ObjectivesObjectives


Task 4.3/4.4 Time Task 4.3/4.4 Time PlanPlan

Study of exp. Layout and Classification – first Study of exp. Layout and Classification – first resultresultby 31st of June 2002by 31st of June 2002

First deployment (already started)First deployment (already started)by 31st of September 2002by 31st of September 2002

First report of integration and First report of integration and interoperability issuesinteroperability issuesby 30th of December 2002by 30th of December 2002

First working version of a VO EU-US domainFirst working version of a VO EU-US domainby 31st of June 2003by 31st of June 2003

Complete deployment.Complete deployment.by 31st of December 2003by 31st of December 2003

The time plan will follow the schedule The time plan will follow the schedule of each experimentof each experiment

Documents

The EU DataTAG Project