1 the national grid service: an overview stephen pickles university of manchester technical...

1

The National Grid Service:An Overview

Stephen Pickles <stephen.pickles@manchester.ac.uk>

University of Manchester

Technical Director, GOSC

Towards an NGS User Induction Course, NeSC, Edinburgh, 8 December 2004

2

Outline

• Context– UK e-Science Programme– NGS and GOSC– ETF, EGEE, OMII, TeraGrid...

• Core Services– Globus, SRB, OGSA-DAI,...

• Operations Infrastructure• Web sites• Current Status

3

Context

• UK e-Science Programme• Grid Operations Support Centre• National Grid Service• Relationships to

– EGEE– Engineering Task Force (ETF)– Open Middleware Infrastructure Institute

(OMII)

4

UK e-Science Programme

UK Grid OperationsSupport Centre

NGS

Key Components

5

GOSC

The Grid Operations Support Centre is a distributed “virtual centre” providing deployment and operations support for the UK e-Science programme.

6

GOSC Roles UK Grid Services

National Services Authentication, authorisation, certificate management, VO

management, security, network monitoring, help desk + support centre.

NGS Services Job submission, simple registry, data transfer, data access and

integration, resource brokering, monitoring and accounting, grid management services, workflow, notification, operations centre.

NGS core-node Services CPU, (meta-) data storage, key software

Services to be coordinated with others (eg OMII, NeSC, LCG):

Integration testing, compatibility & Validation Tests, User Management, training

Administration: Policies and acceptable use conditions SLA’s, SLD’s Coordinate deployment and Operations

7

One Stop Shop

Click for help

8

Authentication, authorisation, certificate management, VO management, security.

Helpdesk

FAQ

9

GOSC does not...

• Run a repository• Develop software (much)

– contribute to developments to influence/adapt cf • “I’ve got one of those you can have. You just need to …”

• (Training – Edinburgh/NeSC are part of GOSC)• Provide support for Access Grid

– use Access Grid Support Centre instead– http://www.agsc.ja.net/

• Do extensive user hand-holding and application support– Need the e-Science Centres– Priorities will be driven by users

10

National Grid Service

11

NGS - A production Grid

* LeedsManchester *

* OxfordRAL *

Level-2 Grid

National Grid Service

* DL

12

NGS “Today”

Projectse-Mineralse-MaterialsOrbital Dynamics of GalaxiesBioinformatics (using BLAST) GEODISE projectUKQCD Singlet meson projectCensus data analysis MIAKT projecte-HTPX project.RealityGridConvertGrid (ESRC)Integrative Biology

Users LeedsOxfordUCLCardiffSouthamptonImperialLiverpoolSheffieldCambridgeEdinburghQUBBBSRCCCLRCManchester

InterfacesOGSI::LiteOGSI::LiteWSRF::LiteWSRF::Lite

13

GOSC Timeline

Q2 Q4 Q2 Q3Q1Q4Q3Q2Q1Q3

2004 20062005

EGEE gLite alpha release

gLite release 1

OMII release

NGS Expansion(Bristol, Cardiff…)

OGSA-DAI

WS plan

NGS ProductionService

NGS WS Service

EGEE gLite releaseOMII Release

NGS Expansion

WS2 plan

NGS WS Service 2

“Tomorrow”

Web Services-based National Grid Infrastructure

14

Core nodes:

Need UK e-Science certificate (1-2 days)

Apply through NGS web site (1-2 weeks)

http://www.ngs.ac.uk

15

Gaining AccessNGS core nodes

• data nodes at RAL and Manchester

• compute nodes at Oxford and Leeds

• free at point of use• apply through NGS web site• light-weight peer review

– 1-2 weeks

• all access is through digital X.509 certificates– from UK e-Science CA– or recognized peer

National HPC services

• HPCx

• CSAR

Must apply separately to research councils

Digital certificate and Conventional (username/

password) access supported

16

UofA

HPCx

UofD

GOSC

NGS Core Nodes: Host core services, coordinate integration, deployment and support+free to access resources for all VOs. Monitored interfaces + services

NGS Partner Sites: Integrated with NGS, some services/resources available for all VOs Monitored interfaces + services

NGS Affiliated Sites: Integrated with NGS, support for some VO’sMonitored interfaces (+security etc.)

RAL Oxford

LeedsMan. CSAR

UofB

UofC

CommercialProvider

PSRE

17

Joining the NGS

Resource providers join the NGS by• Adopting NGS acceptable use and security policies• Run compatible middleware

– as defined by NGS Minimum Software Stack– and verified by compliance test suite

• Support monitoring and accountingTwo levels• NGS affiliates• NGS partners

– also provide significant resources or services to NGS users

More later

18

• Start from LCG2• “Harden” middleware

• Expand applications• €32 from EU

– The EU Grid Infrastructure– 50% deployment/operations– lots at CERN

• matched by PP

• UK + I– training– GOC dev + ops– regional deployment+support

• T2 coordinators

19

LCG

•Used for batch production Now.

• worldwide de facto standards

•Currently trying to interface analysis S/W

• on top of gLite from EGEE

•Need to move out of the physics dept’s

20

GRIDPP • UK Contribution to LCG

– Also supports current users• GridPP1 (2001-2004)

– 33% deployment/operations– 33% middleware dev.– 33% applications

• GridPP2 (2004-2007)– 60% deployment/operations

• LCG2 -> EGEE– 15% middleware dev.– 25% applications

• Not just PPARC funding– Universities

• support LHC• benefit from grid experience• You should know/meet these

people

21

Grid Operations CentreResponsibilities in EGEE

UK focused screen

UKERNA work to be integrated

22

OMII

23

UKCampus

andother Grids

Managing middleware evolution

• Core of GOSC built around experience in deploying and running National Grid Service (NGS) – Support service

• Important to coordinate and integrate this with deployment and operations work in EGEE, LCG and similar projects. – e.g. EGEE – low level services, CA, GOC, CERT...

• Focus on deployment and operations, NOT development.

OMII

ETF

NGSOther software sources Software with proven

capability & realistic deployment experience

‘Gold’ services

Prototypes &specifications

Feedback & future requirements

EGEE…

Deployment/testing/advice

Operations

24

(Grid) Engineering Task Force

• Originally built the UK Level-2 Grid using hardware resources volunteered by the UK e-Science Centres– very heterogeneous– exposed need for dedicated resources

• 0.5 FTE at each e-Science centre• Now primarily conducting pre-deployment evaluation and

testing for NGS• Currently evaluating:

– Condor (nearly complete)– Globus Toolkit version 4– Innergrid and Outergrid– OMII distribution– gLite from EGEE (awaiting release candidate January ’05)

• Also doing portal work for NGS linked to JISC VRE developments

• UDDI work complete

25

NGS Core Services - Globus

• Globus Toolkit version 2– GT 2.4.3 from VDT 1.2

• Job submission (GRAM)• File transfer (GridFTP)• Shell (GSI-SSH)• Information Services (MDS/GIIS/GRIS)

– Information providers from GLUE schema

• More from Steve Pickering later today

26

NGS Core Services - SRB

• Storage Resource Broker from SDSC

• Location transparent access to storage• Metadata catalog• Replica management

• Clients on compute nodes• Servers on data nodes

• More from Andy Richards later today

27

NGS Core Services – Oracle

• Oracle 9i database• Only on data nodes

• Populated by users/data providers

• Infrastructure maintained by NGS database administrators

• Used directly or via OGSA-DAI

• More from Matt Ford later today

28

NGS Core Services – OGSA-DAI

• Open Grid Services Architecture (OGSA)• Database Access and Integration (DAI)• Developed by UK e-Science projects OGSA-DAI

and DAIT• OGSA-DQP (Distributed Query Processor)• Experimental service based on OGSI/GT3 on

Manchester data node only– will consider WS-I and WSRF flavours when in final

release• Uses Oracle underneath• Early users from e-Social Science (ConvertGrid)

• More from Matt Ford later today

29

NGS Core Services - otherOperated by GOSC for NGS and UK e-Science programme

In production:• Certificate Authority• Information Services (MDS/GIIS)• MyProxy server• Integration tests and database• Cluster monitoring• LCG-VOIn testing:• VOMS• EDG Resource BrokerIn development• Accounting• Portal (see Dharmesh’s demonstration, Friday)

30

NGS Organisation• Operations Team

– led by Andy Richards– representatives from all NGS core nodes– meets weekly by Access Grid– day-to-day operational and deployment issues– reports to Technical Board

• Technical Board– led by Stephen Pickles– representatives from all sites and GOSC– meets bi-weekly by Access Grid– deals with policy issues and high-level technical strategy– sets medium term goals and priorities– reports to Management Board

• Management Board (a.k.a. Steering Committee)– meets quarterly– representatives from funding bodies, partner sites and major

stakeholders– sets long term priorities

31

Web Sites

• NGS– http://www.ngs.ac.uk

• GOSC– http://www.grid-support.ac.uk

• CSAR– http://www.csar.cfs.ac.uk

• HPCx– http://www.hpcx.ac.uk

32Google search for “Grid Support”

33

34

35

36

37

38

Production StatusTODAY• 4 JISC-funded core nodes operational• 136 Users registered (36 since 1 September ’04)• Grid enabled – Globus v2 (VDT distribution v1.2) at present• BDII information service (GLUE + MDS Schemas)• Data Services – Oracle, SRB and OGSA-DAI• Growing base of user applications• MyProxy and CA services provided by GOSC• VO Management Software – LCG-VO • User support: Helpdesk

Next…• NGS Portal• Resource Broker• SRB production service• Accounting• EGEE VOMS• Move from User- to Project/VO-based registration• Other Middleware [gLite/OMII etc…]

39

UKGSC: Queue Totals For Month Prior To (10:11 AM, 26/11/2004)

• Certification still dominant query type• General contains GGUS related queries from EGEE/GGUS helpdesk

• (A separate queue for this new traffic is to be created)

Helpdeskhttp://www.grid-support.ac.uk

support@grid-support.ac.uk

40

User registrations so far…

Number of Registered NGS Users

0

20

40

60

80

100

120

140

160

04 April 2004 24 May 2004 13 July 2004 01 September2004

21 October2004

10 December2004

Date

Nu

mb

er o

f U

sers

NGS UserRegistrations

Linear (NGS UserRegistrations)

41

Recent Developments

• NGS newsletter• OMII helpdesk• Resource Broker (LCG) available for early

adopters• TeraGrid

– UK certificates accepted on TeraGrid– Looking at INCA monitoring with Jenny Schopf

• Trial of accounting software from MCS project– using GGF Usage Record draft standard (as EGEE)

• Meeting with Open Science Grid’s iGOC• Collect user exemplars

– Agreed form of acknowledgement

42

Other Developments

• VO server (LCG/EGEE) being trialled by RealityGrid– also CCLRC e-Science

• EGEE VOMS now available• EGEE meeting in den Haag• Security = NGS + GridPP + EGEE• UKERNA developments on Security and Network monitoring

• Negotiations with vendors on software licenses– NAG – positive discussions

• If user A has a license X, they can use any other site with a license X • To be formalised

– Gaussian• Need a Commercial license ($20k-30k)

– Matlab• Currently have a workable solution (binaries) but next release will

break• Mathworks aware but no solution yet

43

The Last Slide• We are in the game of providing a service

– built on leading (sometimes bleeding) edge academic stuff !

• The challenge– not the latest and greatest grid– not what any given user wants

• The solution– want to make it work

• for our researchers• for our institutions

– and maintain compatibility with EGEE, TeraGrid– and accommodate OMII– and expand, bringing in more partners

• Sign people/users up (to the vision)– “get out more”