Upload
jada-bird
View
214
Download
1
Tags:
Embed Size (px)
Citation preview
Tony Doyle GridPP Oversight Committee15 May 2002
Tony Doyle - University of Glasgow
Document MappingDocument Mapping
Exec SummaryExec Summary GoalsGoals Metrics for successMetrics for success Project ElementsProject Elements
Risks/DependenciesRisks/Dependencies(and (and
mechanisms)mechanisms) SummarySummary
PMB-02-EXECPMB-02-EXEC PMB-01-VISIONPMB-01-VISION PMB-02-EXECPMB-02-EXEC Gantt Charts,Gantt Charts,
PMB-05-LCG,PMB-05-LCG, TB-01-Q5-Report, TB-01-Q5-Report, TB- TB-02-UKRollout, PMB-06-02-UKRollout, PMB-06-TierAstatus, PMB-04-TierAstatus, PMB-04-ResourcesResources
PMB-03-STATUS, PMB-03-STATUS, PMB-07-INSTRUMENTSPMB-07-INSTRUMENTS
PMB-02-EXECPMB-02-EXEC
Tony Doyle - University of Glasgow
OutlineOutline
The Vision The Vision Thing…Thing…
GridGrid1.1. ScaleScale2.2. IntegrationIntegration3.3. DisseminationDissemination4.4. LHC AnalysesLHC Analyses5.5. Other AnalysesOther Analyses
6.6. DataGridDataGrid
7.7. LCGLCG
8.8. InteroperabilityInteroperability
9.9. InfrastructureInfrastructure
10.10.FinanacesFinanaces SummarySummary
Tony Doyle - University of Glasgow
GridPP DocumentsGridPP Documents
GridPP Project Management Board
Vision Statement
From Web to Grid -Building the next IT Revolution
DOCUMENT IDENTIFIER: GridPP-PMB-01-Vision
Date: 07/05/2002
Version: 1.0
Document status: FINAL
Author Tony Doyle
Tony Doyle - University of Glasgow
GridPP VisionGridPP Vision
From Web to Grid - From Web to Grid - Building the next IT Building the next IT RevolutionRevolution
PremisePremise
The next IT revolution will The next IT revolution will be the Grid. The Grid is a be the Grid. The Grid is a practical solution to the practical solution to the data-intensive problems that data-intensive problems that must be overcome if the must be overcome if the computing needs of many computing needs of many scientific communities and scientific communities and industry are to be fulfilled industry are to be fulfilled over the next decade.over the next decade.
AimAim
The GridPP Collaboration The GridPP Collaboration aims to develop and deploy aims to develop and deploy the largest-scale science the largest-scale science Grid in the UK for use by the Grid in the UK for use by the worldwide particle physics worldwide particle physics community.community.
Many Challenges..Shared distributed
infrastructure For all experiments
Tony Doyle - University of Glasgow
GridPP ObjectivesGridPP Objectives
1. SCALE: GridPP will deliver the Grid software 1. SCALE: GridPP will deliver the Grid software (middleware) and hardware infrastructure to (middleware) and hardware infrastructure to enable the testing of a prototype of the Grid for enable the testing of a prototype of the Grid for the LHC of significant scale. the LHC of significant scale.
2. INTEGRATION: The GridPP project is designed to 2. INTEGRATION: The GridPP project is designed to integrate with the existing Particle Physics integrate with the existing Particle Physics programme within the UK, thus enabling early programme within the UK, thus enabling early deployment and full testing of Grid technology deployment and full testing of Grid technology and efficient use of limited resources. and efficient use of limited resources.
3. DISSEMINATION: The project will disseminate the 3. DISSEMINATION: The project will disseminate the GridPP deliverables in the multi-disciplinary e-GridPP deliverables in the multi-disciplinary e-science environment and will seek to build science environment and will seek to build collaborations with emerging non-PPARC Grid collaborations with emerging non-PPARC Grid activities both nationally and internationally.activities both nationally and internationally.
4. UK PHYSICS ANALYSES (LHC): The main aim is to 4. UK PHYSICS ANALYSES (LHC): The main aim is to provide a computing environment for the UK provide a computing environment for the UK Particle Physics Community capable of meeting Particle Physics Community capable of meeting the challenges posed by the unprecedented data the challenges posed by the unprecedented data requirements of the LHC experiments.requirements of the LHC experiments.
5. UK PHYSICS ANALYSES (OTHER): The process of 5. UK PHYSICS ANALYSES (OTHER): The process of creating and testing the computing environment creating and testing the computing environment for the LHC will naturally provide for the needs for the LHC will naturally provide for the needs of the current generation of highly data of the current generation of highly data intensive Particle Physics experiments: these intensive Particle Physics experiments: these will provide a live test environment for GridPP will provide a live test environment for GridPP research and development.research and development.
6. DATAGRID: Grid technology is the framework used to 6. DATAGRID: Grid technology is the framework used to develop this capability: key components will be develop this capability: key components will be developed as part of the EU DataGrid project and developed as part of the EU DataGrid project and elsewhere.elsewhere.
7. LCG: The collaboration builds on the strong 7. LCG: The collaboration builds on the strong computing traditions of the UK at CERN. The computing traditions of the UK at CERN. The CERN working groups will make a major CERN working groups will make a major contribution to the LCG research and contribution to the LCG research and development programme.development programme.
8. INTEROPERABILITY: The proposal is also integrated 8. INTEROPERABILITY: The proposal is also integrated with developments from elsewhere in order to with developments from elsewhere in order to ensure the development of a common set of ensure the development of a common set of principles, protocols and standards that can principles, protocols and standards that can support a wide range of applications. support a wide range of applications.
9. INFRASTRUCTURE: Provision is made for facilities at 9. INFRASTRUCTURE: Provision is made for facilities at CERN (Tier-0), RAL (Tier-1) and use of up to four CERN (Tier-0), RAL (Tier-1) and use of up to four Regional Centres (Tier-2).Regional Centres (Tier-2).
10. OTHER FUNDING: These centres will provide a focus 10. OTHER FUNDING: These centres will provide a focus for dissemination to the academic and for dissemination to the academic and commercial sector and are expected to attract commercial sector and are expected to attract funds from elsewhere such that the full funds from elsewhere such that the full programme can be realised.programme can be realised.
(…. WHAT WE SAID WE COULD DO (…. WHAT WE SAID WE COULD DO IN THE PROPOSAL)IN THE PROPOSAL)
Tony Doyle - University of Glasgow
Grid – A Single ResourceGrid – A Single Resource
Peta Bytes of data storage
Many millions
of events
Many samples
Distributed resources
Many 1000s of computers required
GRIDA unified approach
Worldwide collaboration
Various conditions
Heterogeneous operating systems
GRIDA unified approach
Tony Doyle - University of Glasgow
Grid - What’s been happening?Grid - What’s been happening?
A lot…
GGF4, OGSA and support of IBM (and others) GGF4, OGSA and support of IBM (and others) [as opposed to .NET development framework and passports
to access services] Timescale? September 2002
W3C architecture for web servicesW3C architecture for web services Chose (gzipped) XML as opposed to other solutions for
metadata descriptions… and web-based interfaces
linux linux [as opposed to other platforms… lindows??]
C++ (experiments) and C, Java (middleware) APIsC++ (experiments) and C, Java (middleware) APIs [mono - Open Source implementation of the .NET
Development Framework??]
OGSA
GRIDA unified approach
Tony Doyle - University of Glasgow
GridPP ContextGridPP Context
Provide architecture and middleware
Use the Grid with simulated data
Use the Grid with real data
Future LHC Experiments
Running US Experiments
Build Tier-A/prototype Tier-1 and Tier-2 centres in the UK and join worldwide
effort to develop middleware for the
experiments
Tony Doyle - University of Glasgow
EDG TestBed 1 StatusEDG TestBed 1 Status
Web interface Web interface showing status of showing status of (~400) servers at (~400) servers at testbed 1 sitestestbed 1 sites
GRIDA unified approach
GRIDextend to all expts
Tony Doyle - University of Glasgow
LHC computing at a glanceLHC computing at a glance
The investment in LHC computing will be massiveThe investment in LHC computing will be massive LHC Review estimated 240MCHF (before LHC delay) 80MCHF/y afterwards
These facilities will be distributedThese facilities will be distributed Political as well as sociological and practical reasons
Europe:267 institutes, 4603 users
Elsewhere: 208 institutes, 1632 users
Eng. & accel. services
Infrastructure (non-physics)
non-LHC share of base physics svcs & infrastr.
LHC share of base physics services &
infrastructure
Physics WANComputer centre
refurbishment
PrototypeOutsourced
administration & operation
Tier 0 investment
Tier 1 investment
0
10
20
30
40
50
60
2001 2002 2003 2004 2005 2006 2007 2008
year
MC
HF
Funding available (MTP)
1. s1. sccaallee
Tony Doyle - University of Glasgow
RTAG StatusRTAG Status
6 RTAGs created to date: 6 RTAGs created to date: RTAG1 (Persistency Framework; status: completed) RTAG2 (Managing LCG Software; status: running) RTAG3 (Math Library Review; status: running) RTAG4 (GRID Use Cases; status: starting) RTAG5 (Mass Storage; status: running) RTAG6 (Regional Centres; status: starting)
Two more in advanced state of preparation:Two more in advanced state of preparation: Simulation components Data Definition Tools
7. LCG7. LCG
Tony Doyle - University of Glasgow
Fabrics & Grid DeploymentFabrics & Grid Deployment
LCG Level 1 Milestone: deploy a LCG Level 1 Milestone: deploy a Global Grid Service Global Grid Service within within 1 year1 year sustained 24 X 7 service including sites from three continents
identical or compatible Grid middleware and infrastructure
several times the capacity of the CERN facility and as easy to use
Ongoing work at CERN to increase automation and Ongoing work at CERN to increase automation and streamline configuration, especially for migration to streamline configuration, especially for migration to RedHat 7.2.RedHat 7.2.
Aim to phase out old CERN solutions by mid-2003.
7. LCG7. LCG
Tony Doyle - University of Glasgow
LCG TimelineLCG Timeline
2002 200520042003
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4Q1 Q2 Q3 Q4Q1 Q2 Q3 Q4
Prototype of Hybrid Event Store (Persistency Framework)
Hybrid Event Store available for general users
Distributed production using grid services
First Global Grid Service (LCG-1) available
Distributed end-user interactive analysis
Full Persistency Framework
LCG-1 reliability and performance targets
“50% prototype” (LCG-3) available
LHC Global Grid TDR
applicationsapplications
gridgrid
1. 1. titimmeessccaallee
Tony Doyle - University of Glasgow
Be a part of this?Be a part of this?
LCG DevelopmentLCG Development– – Long Term Attachment at CERNLong Term Attachment at CERN This will enable Grid developments This will enable Grid developments
in the UK to be (more) fully in the UK to be (more) fully integrated with long-term Grid integrated with long-term Grid development plans at CERN.development plans at CERN.
The proposed mechanism is:The proposed mechanism is: 1. submit a short one-page outline 1. submit a short one-page outline
of current and proposed work, of current and proposed work, noting how this work can best be noting how this work can best be developed within a named team at developed within a named team at CERN, by e-mail to the GridPP CERN, by e-mail to the GridPP Project Leader (Tony Doyle) and Project Leader (Tony Doyle) and GridPP CERN Liaison (Tony Cass).GridPP CERN Liaison (Tony Cass).
2. This case will be discussed at 2. This case will be discussed at the following weekly GridPP PMB the following weekly GridPP PMB meeting and outcomes will be meeting and outcomes will be communicated as soon as communicated as soon as possible by e-mail following that possible by e-mail following that meeting. meeting.
NotesNotes1. The minimum period for LTA is 3 1. The minimum period for LTA is 3
months. It is expected that a work months. It is expected that a work programme will be typically for 6 programme will be typically for 6 months (or more).months (or more).
2. Prior DataGrid and LHC (or other) 2. Prior DataGrid and LHC (or other) experiments' Grid work are experiments' Grid work are normally expected.normally expected.
3. It is worthwhile reading3. It is worthwhile readinghttp://http://cerncern..chch//lcglcg//pebpeb/applications/applications in order to get an idea of the areas in order to get an idea of the areas
covered, and the emphasis placed, covered, and the emphasis placed, by the LCG project on specific by the LCG project on specific areas (building upon DataGrid and areas (building upon DataGrid and LHC experiments' developments).LHC experiments' developments).
4. Please send all enquiries and 4. Please send all enquiries and proposals to:proposals to:
Tony Doyle <[email protected]> Tony Doyle <[email protected]> andand
Tony CASS <[email protected]>Tony CASS <[email protected]>
Tony Doyle - University of Glasgow
Summary of LCGSummary of LCG
Project got under way early this yearProject got under way early this year Launch workshop and early RTAGs give good input for high-level Launch workshop and early RTAGs give good input for high-level
planning …planning … … … to be presented to LHCC in Julyto be presented to LHCC in July New plan takes account of first beam in 2007New plan takes account of first beam in 2007 No serious problems foreseen in synchronising LCG plans with No serious problems foreseen in synchronising LCG plans with
those of the experimentsthose of the experiments Collaboration with the many Grid projects needs more workCollaboration with the many Grid projects needs more work Technical collaboration with the Regional Centres has to be Technical collaboration with the Regional Centres has to be
establishedestablished Recruitment of special staff going well (but need to keep the Recruitment of special staff going well (but need to keep the
recruitment momentum going)recruitment momentum going) Serious problem with materials fundingSerious problem with materials funding
7. LCG7. LCG
Tony Doyle - University of Glasgow
Building upon SuccessBuilding upon Success
The most important criterion for establishing the status of The most important criterion for establishing the status of this project was the European Commission review on this project was the European Commission review on March 1st 2002. March 1st 2002.
The review report of project IST-2000-25182 DATAGRID is The review report of project IST-2000-25182 DATAGRID is available from PPARC. available from PPARC.
The covering letter states “As a general conclusion, the The covering letter states “As a general conclusion, the reviewers found that the overall performance of the project reviewers found that the overall performance of the project is good and in some areas beyond expectations.” is good and in some areas beyond expectations.”
The reviewers state “The deliverables due for the first The reviewers state “The deliverables due for the first review were in general of excellent quality, and all of them review were in general of excellent quality, and all of them were available on time… All deliverables are approved. The were available on time… All deliverables are approved. The project is doing well, exceeding expectations in some project is doing well, exceeding expectations in some areas, and coping successfully with the challenges due to areas, and coping successfully with the challenges due to its size.” its size.”
6. DataGrid6. DataGrid
Tony Doyle - University of Glasgow
6. DataGrid6. DataGrid
Tony Doyle - University of Glasgow
WP1 – Workload Management WP1 – Workload Management (Job Submission)(Job Submission)
1. Authenticationgrid-proxy-init
2. Job submission to DataGriddg-job-submit
3. Monitoring and controldg-job-statusdg-job-canceldg-job-get-output
4. Data publication and replication (WP2)globus-url-copy, GDMP
5. Resource scheduling – use of CERN MSS
JDL, sandboxes, storage elements
Important to implement this
for all experiments…
6. DataGrid6. DataGrid
Tony Doyle - University of Glasgow
WP2 - SpitfireWP2 - Spitfire 6. DataGrid6. DataGrid
Tony Doyle - University of Glasgow
WP3 - R-GMA WP3 - R-GMA
Consumer Servlet
RegistryAPI
Consumer Servlet
RegistryAPI
Consumer Servlet
RegistryAPI
Consumer Servlet
RegistryAPI
Sensor Code
ProducerAPI
Application Code
ConsumerAPI
ProducerServlet
RegistryAPI
Registry Servlet
SchemaAPI
Schema Servlet
“Event Dictionary”
Application Code
ArchiverAPI
DBProducer
DBProducerServlet
Archiver Servlet
ConsumerAPIConsumer
APIConsumerAPIConsumer
API
User code here.Builds on R-GMA
Database Structures.
User code monitors output here.
6. DataGrid6. DataGrid
Tony Doyle - University of Glasgow
The LCFG Architecture
LCFG Source Files
mkxprof
Web Server
XML Profi le
ldxprofldxprof
GenericComponent
GenericComponent
rdxprofrdxprof
LCFG Components
DBM File
Server
Client
WP4 - LCFG WP4 - LCFG
6. DataGrid6. DataGrid
Tony Doyle - University of Glasgow
Interface
Queue M
anager
Request Manager
Pipe Manager
Tape
Disk
Named Pipe
1
2
3
4
5
6
78
Interface Layer The Core and the Bottom Layer
MS
MHandler
Named Pipe
Named Pipe
Named Pipe
Pipe Store
Network
Data Flow Diagram for SEData Flow Diagram for SEWP5 – Storage ElementWP5 – Storage Element
A consistent interface to MSS.A consistent interface to MSS. MSSMSS
CastorCastorHPSSHPSSRAID arraysRAID arraysSRMSRMDMFDMFEnstoreEnstore
Interfaces Interfaces GridFTPGridFTPGridRFIOGridRFIO/grid/gridOGSAOGSA
6. DataGrid6. DataGrid
Tony Doyle - University of Glasgow
WP6 - TestBed 1 StatusWP6 - TestBed 1 Status
Web interface Web interface showing status of showing status of (~400) servers at (~400) servers at testbed 1 sitestestbed 1 sites
GRIDextend to all expts
6. DataGrid6. DataGrid
Tony Doyle - University of Glasgow
WP7 – Network MonitoringWP7 – Network Monitoring 6. DataGrid6. DataGrid
Tony Doyle - University of Glasgow
WP7 - EDG AuthorisationWP7 - EDG Authorisationgrid-mapfilegrid-mapfile generation generation
o=testbed,dc=eu-datagrid, dc=org
CN=Franz Elmer
ou=People
CN=John Smith
mkgridmap grid-mapfile
VOVODirectoryDirectory
““AuthorizationAuthorizationDirectory”Directory”
CN=Mario Rossi
o=xyz,dc=eu-datagrid, dc=org
CN=Franz ElmerCN=John Smith
Authentication Certificate
Authentication Certificate
Authentication Certificate
ou=People ou=Testbed1 ou=???
local users ban list
6. DataGrid6. DataGrid
Tony Doyle - University of Glasgow
1. Realistic Large-Scale Tests1. Realistic Large-Scale Tests Reliability! Need reliable dg-job-*
command suite 2. Data management2. Data management
Reliability! Need reliable gdmp-* command suite, file-transfer commands
3. Mass Storage Support3. Mass Storage Support Working access to MSS (CASTOR
and HPSS at CERN, Lyon) 4. Lightweight User Interface4. Lightweight User Interface
Put on a laptop or std. Desktop machine
5. Portability Demonstrable portability of
middleware: a) use other resources, b) debugging
6. Scratch Space Job requests X amount of
scratch space to be available during execution, system tells job where it is
7. Output File Support JDL support for output files:
specify where output should go in JDL, not in job script
WP8 - ApplicationsWP8 - Applications 6. DataGrid6. DataGrid
Tony Doyle - University of Glasgow
Expt. FeedbackExpt. Feedback 4. and 5. Expts4. and 5. Expts
Tony Doyle - University of Glasgow
= Minimal = Minimal e-Bureaucracye-Bureaucracy
8. Interoperability8. Interoperability
5. Other Expts5. Other Expts
Tony Doyle - University of Glasgow
GRID JOB SUBMISSION GRID JOB SUBMISSION – External User Experience– External User Experience 5. Other Expts5. Other Expts
Tony Doyle - University of Glasgow
Things Missing, apparently…Things Missing, apparently… 5. Other Expts5. Other Expts
Tony Doyle - University of Glasgow
Expt. FeedbackExpt. Feedback 4. and 5. Expts4. and 5. Expts
Tony Doyle - University of Glasgow
GridPP PosterGridPP Poster 3. Dissemination3. Dissemination
Tony Doyle - University of Glasgow
Tier 1/A EDG PosterTier 1/A EDG Poster 3. Dissemination3. Dissemination
Tony Doyle - University of Glasgow
BaBar PosterBaBar Poster 3. Dissemination3. Dissemination
Tony Doyle - University of Glasgow
LHCb PosterLHCb Poster 3. Dissemination3. Dissemination
Tony Doyle - University of Glasgow
ScotGRID PosterScotGRID Poster 3. Dissemination3. Dissemination
Tony Doyle - University of Glasgow
Identifiable Progress...Identifiable Progress...
t0
t1
3. Dissemination3. Dissemination
Tony Doyle - University of Glasgow
WebLogWebLog
Allows every area/sub group to have its own 'news' pagesAllows every area/sub group to have its own 'news' pages
Tony Doyle - University of Glasgow
GridPP & Core e-Science CentresGridPP & Core e-Science Centres
NeSCNeSC Close ties, hosted 2nd GridPP Collaboration Meeting,
Collaboration on EDIKT Project? Training...
BelfastBelfast Replied but not yet up and running.
CambridgeCambridge Close ties, hosted 3rd GridPP Collaboration Meeting. Share one
post with GridPP. Will collaborate on ATLAS Data Challenges.
CardiffCardiff Replied - contacts through QM (Vista) and Brunel GridPP Group.
Written formally to all e-Science centres inviting contact and Written formally to all e-Science centres inviting contact and collaboration with GridPP.collaboration with GridPP.
3. Dissemination3. Dissemination
Tony Doyle - University of Glasgow
GridPP & Core e-Science CentresGridPP & Core e-Science Centres
LondonLondon No formal reply but close contacts through IC HEP Group.
IC will host 5th GridPP Collaboration Meeting.
ManchesterManchester No collab. projects so far. Manchester HEP Group will host
4th GridPP Collaboration Meeting.
NewcastleNewcastle In contact - Database projects?
OxfordOxford Close ties, collaboration between Oxford HEP Group and
GridPP on establishment of central Tier-2 centre? CS/Core-GridPP-EDG links? Probably host 6th GridPP Collaboration Meeting.
SouthamptonSouthampton Replied but no collaboration as yet.
3. Dissemination3. Dissemination
Tony Doyle - University of Glasgow
GLUEGLUE
How do we integrate with developments from elsewhere in How do we integrate with developments from elsewhere in order to ensure the development of a common set of order to ensure the development of a common set of principles, protocols and standards that can support a principles, protocols and standards that can support a wide range of applications?wide range of applications?
GGF… GGF… Within the Particle Physics community, these ideas are Within the Particle Physics community, these ideas are
currently encapsulated in the Grid Laboratory Uniform currently encapsulated in the Grid Laboratory Uniform Environment (Environment (GLUEGLUE). ).
Recommend this as a starting point for the wider Recommend this as a starting point for the wider deployment of Grids across the Atlantic. See deployment of Grids across the Atlantic. See http://www.http://www.hicbhicb.org/glue/GLUE-v0.1.doc.org/glue/GLUE-v0.1.doc (Ruth Pordes (Ruth Pordes et al.)et al.)
8. Interoperability8. Interoperability
Tony Doyle - University of Glasgow
8. Interoperability8. Interoperability
Tony Doyle - University of Glasgow
UK Tier-A/prototype Tier-1 CentreUK Tier-A/prototype Tier-1 Centre
RolesRolesTier-A Centre for BaBar Tier-A Centre for BaBar EDG testbed(s)EDG testbed(s)LCG prototype Tier-1 Centre LCG prototype Tier-1 Centre prototype Tier-1 for LHC experiments (Data Challenges prototype Tier-1 for LHC experiments (Data Challenges independent independent of LCG development…)of LCG development…)Interworking with other UK resources (JIF, JREI, eSC) Interworking with other UK resources (JIF, JREI, eSC) = = UK portalUK portalexisting LEP, DESY and non-accelerator experimentsexisting LEP, DESY and non-accelerator experiments
PurchasesPurchases First year = Hardware Advisory Group (HAG1)First year = Hardware Advisory Group (HAG1) Determine balance between cpu, disk, and tape Determine balance between cpu, disk, and tape Experts on specific technologiesExperts on specific technologies Propose more HAGs (2 and 3).. Propose more HAGs (2 and 3).. Needs to be successful in all roles...Needs to be successful in all roles...
9. Infrastructure9. Infrastructure
Tony Doyle - University of Glasgow
Rollout of the UK Grid for PPRollout of the UK Grid for PP
Operational stability of GridPP middleware = Testbed teamOperational stability of GridPP middleware = Testbed team The “gang of four” … Andrew McNab, Steve Traylen, Dave The “gang of four” … Andrew McNab, Steve Traylen, Dave
Colling (other half) and Owen MoroneyColling (other half) and Owen Moroney Ensures the release of “Testbed” quality EDG softwareEnsures the release of “Testbed” quality EDG software
documentation lead for other system managers in terms of implementation pre-defined software cycle releases (2 months..)
Subject of the Rollout Plan… “Planning for EDG Testbed Subject of the Rollout Plan… “Planning for EDG Testbed software deployment and support at participating UK software deployment and support at participating UK sites” (Pete Clarke, John Gordon)sites” (Pete Clarke, John Gordon)
LCG is the proposed mechanism by which the EDG LCG is the proposed mechanism by which the EDG testbed at CERN becomes an LCG Grid Service. The testbed at CERN becomes an LCG Grid Service. The evolution of the EDG testbed to the LCG Grid Service will evolution of the EDG testbed to the LCG Grid Service will take account of both EDG and US grid technology. Need to take account of both EDG and US grid technology. Need to take account of this..take account of this..
9. Infrastructure9. Infrastructure
Tony Doyle - University of Glasgow
Longer Term..Longer Term..
LCG Grid ServiceLCG Grid Service Takes account of EDG and US grid technology Takes account of EDG and US grid technology A large-scale Grid resource, consistent with the LCG A large-scale Grid resource, consistent with the LCG
timeline, within the UK.timeline, within the UK. Scale in UK? 0.5 Pbytes and 2,000 distrib. CPUsScale in UK? 0.5 Pbytes and 2,000 distrib. CPUs
= GridPP in Sept 2004 = GridPP in Sept 2004 ““50% prototype”50% prototype”
2002 200520042003
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4Q1 Q2 Q3 Q4Q1 Q2 Q3 Q4
Prototype of Hybrid Event Store (Persistency Framework)
Hybrid Event Store available for general users
Distributed production using grid services
First Global Grid Service (LCG-1) available
Distributed end-user interactive analysis
Full Persistency Framework
LCG-1 reliability and performance targets
“50% prototype” (LCG-3) available
LHC Global Grid TDR
applicationsapplications
gridgrid
2002 200520042003
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4Q1 Q2 Q3 Q4Q1 Q2 Q3 Q4
2002 200520042003
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4Q1 Q2 Q3 Q4Q1 Q2 Q3 Q4
Prototype of Hybrid Event Store (Persistency Framework)
Hybrid Event Store available for general users
Distributed production using grid services
First Global Grid Service (LCG-1) available
Distributed end-user interactive analysis
Full Persistency Framework
LCG-1 reliability and performance targets
“50% prototype” (LCG-3) available
LHC Global Grid TDR
applicationsapplications
gridgrid
9. Infrastructure9. Infrastructure
Tony Doyle - University of Glasgow
£17m 3-Year Project£17m 3-Year Project
£3.78m
£5.67m
£3.66m
£1.99m
£1.88m
CERN
DataGrid
Tier - 1/A
Applications
Operations
Five componentsFive components Tier-1/A = Hardware + ITD Support Staff DataGrid = DataGrid Posts + PPD Staff Applications = Experiments Posts Operations = Travel + Management + Early Investment CERN = LCG posts + Tier-0 + LTA
10. Finances10. Finances
Dave Dave BrittonBritton
Tony Doyle - University of Glasgow
1. Recruitment1. Recruitment
EDG Funded PostsEDG Funded Posts (Middleware/Testbed) (Middleware/Testbed) All 5 in post + 1 additional
EDG Unfunded Posts EDG Unfunded Posts (Middleware/Testbed)(Middleware/Testbed) 15 out of 15 in post
GridPP PostsGridPP Posts (Applications + Tier1/A) (Applications + Tier1/A) Allocated Dec 2001 13 out of 15 in post
CERN PostsCERN Posts First Round = 105 Applicants, 12 Offers, 9 Accepted 4 in Applications, 2 Data Management, 3 Systems Second Round = 140 applicants, 9 Offers Third Round ~ 70 Applicants Aim ~ 28 posts
Status Feb'02
Owen Moroney starts Jan 2002Frederic Brochu (starts 1st Apr 02, part funded by CeSC)Regina Tam started 7th Jan 02Steve Traylen started 14th Jan 02Alexander Holt from 1st Aug 01Gavin McCance started 1st Sep 01 (1FTE for 3 yrs)Will Bell started 1st Nov 01 (1FTE for 2 yrs)Dave Colling from 1st Oct 01 (WP1&6)Phillip Lewis from 1st Oct 01 (WP8)Michael George started 1st Oct 01Andrew McNab started 1st Nov 01Started ~15th Apr 02Started 1st Jan 02Arijeet Datta started 21st Jan 02Mike Gardner started 28th Jan 02Paul Mealor started 1st Jul 01
WP3 : Antony Wilson started 4th Jun 01 Laurence Field started 28th Aug 01 Xiaomei Zhu started 28th Jan 02 Manish Soni started 2nd Apr 02WP5 : Owen Synge started 27th Nov 01 Timothy Eves started 18th Dec 01WP8 : Stephen Burke started 1st Nov 01
Tony Doyle - University of Glasgow
2. Monitoring Staff Effort [SM]2. Monitoring Staff Effort [SM] Robin Robin MiddletonMiddleton
Tony Doyle - University of Glasgow
3. Progress towards deliverables..3. Progress towards deliverables..
GridPP Experimental Support :Milestone/Deliverable specification and schedule
Milestone/Deliverable Specification
Experiment: LHCb Due: 2002-Q2
Title Technology review document
D/M [Type] [D] DocumentDescription ???? what are you actually reviewing and for what purpose ????Dependencies:(if any)
None
Milestone/Deliverable Specification
Experiment: LHCb Due: 2002-Q2
Title Detailed requirements, architecture and design document
D/M [Type] [D] DocumentDescription ???This document will establish the LHCb requirements for its Grid
based analysis and MC production, the architecture it wishes to use inassembling components, and the detailed design of the first prototype.???
Dependencies:(if any)
None
Milestone/Deliverable Specification
Experiment: LHCb Due: 2002-Q3
Title First prototype of Grid interface
D/M [Type] [M] Prototype & demonstrationDescription Interface originally based on command lines modelled on current Atlas
and LHCb production tools. It will allow simple Gaudi “analysis” jobs tobe submitted to EDG testbed sites in addition to MC production jobs. Ageneric script syntax will need to be developed and job submission scriptsshould be automatically generated. The scripts have to have an errorcatching mechanism. Functionality will include:
DataGrid job submission (WP1 tools - already exist)
GridPP Experimental Support :Milestone/Deliverable specification and schedule
Milestone/Deliverable Specification
Experiment: SAM Core Development Due: 2002-Q2
Title Technology Review
D/M [Type] [D] Technology Review reportDescriptionDependencies:(if any)
None
Milestone/Deliverable Specification
Experiment: SAM Core Development Due: 2002-Q2
Title Architecture Design
D/M [Type] [D] Document specifying architecture designDescription In parallel with the technology review, the architecture to be used for the
upgrade of SAM will be specified.Dependencies:(if any)
None
Milestone/Deliverable Specification
Experiment: SAM Core Development Due: 2002-Q3
Title Demonstration of phase-1 SAM development
D/M [Type] [M] Report on demonstrationDescription The first modifications to SAM will be
- the inclusion of GridFTP as a transport option.- the inclusion of Condor-G to allow submission of a job to a remote
site (site chosen by hand)- Decentralisation of existing SAM information services and development of a
prototype of information services that provides information on availableresources and tracks them.
This milestone will be a demonstration of these upgrades operatingsuccessfully by submitting jobs to key sites in the US and UK
GridPP Experimental Support:Milestone/Deliverable specification and schedule
QCDGrid
Milestone/Deliverable Specification
Experiment: QCDGrid Due: 2002-Q2
Title Develop an XML Schema for lattice QCD Calculations
D/M [Type] [M] Prototype XML Schema for meta-data catalogue (see nextdeliverable)
Description We aim to develop an XML schema, to define the format of the meta-datadocuments of the lattice QCD data files in an extensible and scientificallymeaningful manner.
Dependencies:(if any)
None
Milestone/Deliverable Specification
Experiment: QCDGrid Due: 2002-Q2
Title Develop a meta-data data catalogue for QCD simulation data
D/M [Type] [M] Meta-data catalogue and browser demonstratorDescription We aim to develop a XML Database Server (a meta-data catalogue) for
storing and querying the lattice QCD meta-data. Access to this meta-datawill be via command-line tools and through a Browser (see nextdeliverable).
Dependencies:(if any)
None
Milestone/Deliverable Specification
Experiment: QCDGrid Due: 2002-Q2
Title Develop a Browser for interrogating the meta-data data catalogue
D/M [Type] [M] Meta-data catalogue and browser demonstratorDescription We aim to develop a Browser to allow users to query the meta-data
catalogue. The browser will supply a single interface to the meta-datacatalogue, and ultimately to the data catalogue (see next deliverable).
Dependencies:(if any)
None
Pete Pete ClarkeClarke
Tony Doyle - University of Glasgow
-1. Next steps..-1. Next steps..
O(100k)O(100k) CLRC support through to Sept 04 Other experiments – unfunded in peer review process Tier-2 centres – unfunded initially
£2.3m £2.3m eDIKT (e-Data, Information and Knowledge Transformation) [SHEFC] Particle
Physics = application area - assignment of two (of twelve) FTEs in initial planning. Discussions ongoing with EPCC.
O(€O(€100m)100m) The first call for Framework VI will be early next year. Call out now for expressions of interest for new networks and integrated
projects. Draft document led by David Williams (CERN) “Enabling Grids and e-Science
in Europe” plans to extend the current paradigms with CERN at its focus as the European e-Science Centre.
We believe this is the right approach. Incorporates the UK’s e-Science agenda, adding a European dimension. It also
recognises the central role of CERN and builds upon the recent successes of EDG.
PPARC Contact: Neil Geddes
10. Finances10. Finances
Tony Doyle - University of Glasgow
Andrew McNab - Manchester HEP - 10 May 2002
Green Dot G1.1.3 G2.0(b) EDG-CEBabar-CEBirmingham y y yBristol y y y y yBrunel y yCambridge yEdinburgh y yGlasgow y yImperial y y yLancaster y yLiverpool y yManchester y y y y yOxford y yQMUL y y yRAL y y y yRHUL y yUCL y
Testbed Status OverviewTestbed Status Overview MetricsMetrics
Tony Doyle - University of Glasgow
What is in place in the UK testbed?(an RB centric view of the world)Only GridPP and Babar VOs
Imperial
R. B.
JSS
II
LB
Bristol
Replica Catalogue
Imperial
CE, SE, UIRAL
CE, SE, UI
Birmingham
CE, UI
Liverpool
CE, UI
Bristol
CE, UI
QMUL
CE, UI
RHUL
CE, UI
IN2P3-Babar
UI
MetricsMetrics
Tony Doyle - University of Glasgow
Grid Support CentreGrid Support Centre
UKHEP CA uses primitive technologyUKHEP CA uses primitive technology It works but takes effort 201 personal certs issued (98 still valid) 119 other certs issued (93 still valid)
GSC will run a CA for UK escience CAGSC will run a CA for UK escience CA Uses openCA; Registration Authority uses web
We plan to use itWe plan to use it Namespace identifies RA, not Project
Through GSC we have access to skills of Through GSC we have access to skills of CLRC eSCCLRC eSC
Use helpdesk to formalise support later Use helpdesk to formalise support later in the rolloutin the rollout
8. Interoperability8. Interoperability
UKUK
e-Sciencee-Science
CertificationCertification
AuthorityAuthority
MetricsMetrics
Tony Doyle - University of Glasgow
SummarySummary
A vision is only useful if its A vision is only useful if its sharedshared
Grid success is fundamental Grid success is fundamental for PPfor PP
1.1. Scale in UK? 0.5 Pbytes and Scale in UK? 0.5 Pbytes and 2,000 distrib. CPUs 2,000 distrib. CPUs
GridPP in Sept 2004 GridPP in Sept 2004
2.2. Integration – ongoing.. Integration – ongoing..
3.3. Dissemination – external Dissemination – external and internaland internal
4.4. LHC Analyses – ongoing LHC Analyses – ongoing feedback mechanism..feedback mechanism..
5.5. Other Analyses – closely Other Analyses – closely integrated using EDG toolsintegrated using EDG tools
6.6. DataGrid - major investment = DataGrid - major investment = must be (and is so far) successful must be (and is so far) successful
7.7. LCG – Grid as a Service LCG – Grid as a Service
8.8. Interoperability – sticky subjectInteroperability – sticky subject
9.9. Infrastructure – Tier-A/1 in place, Infrastructure – Tier-A/1 in place, Tier-2’s to follow… Tier-2’s to follow…
10.10. Finances – (very well) under control Finances – (very well) under control Next steps on framework VI..Next steps on framework VI.. CERN = EU’s e-science centre?CERN = EU’s e-science centre? Co-operation required with other Co-operation required with other
disciplines/industrydisciplines/industry
11.11. Monitoring mechanisms in placeMonitoring mechanisms in place
12.12. Emphasis on deliverablesEmphasis on deliverables
Tony Doyle - University of Glasgow
ExecutiveExecutive22 Summary Summary
Significant progress... Significant progress... Project is now well defined Project is now well defined
in a broad sense and is in a broad sense and is progressing on a series of progressing on a series of fronts.fronts.
We have responded and We have responded and outlined our plans to outlined our plans to address the concerns of the address the concerns of the last OC concerning:last OC concerning:
1. WP5;
2. Rollout plan;
3. Monitoring instruments;
4. Metrics for success.
The project has demonstrated The project has demonstrated progress in:progress in:
1. Widespread deployment of EDG testbeds in the UK;
2. Integration with specific experimental areas (BaBar, UKDMC and LISA); and
3. Demonstrating Grid deployment in the UK at the NeSC opening.
We see various challenges ahead:We see various challenges ahead:1. Development of more detailed
metrics and monitoring of outputs;2. Management of changes due to
external developments (e.g. OGSA);3. Development of Tier-2 deployment;4. Engagement of the UK HEP
community; and5. Future funding initiatives such as
Framework VI.