A Computing Tier 2 Node Eric Fede – LAPP/IN2P3. 2 Eric Fede – 1st Chinese-French Workshop Plan What is a Tier 2 –Context and definition To be a Tier 2.

Slides:



Advertisements
Similar presentations
Alastair Dewhurst, Dimitrios Zilaskos RAL Tier1 Acknowledgements: RAL Tier1 team, especially John Kelly and James Adams Maximising job throughput using.
Advertisements

T1 at LBL/NERSC/OAK RIDGE General principles. RAW data flow T0 disk buffer DAQ & HLT CERN Tape AliEn FC Raw data Condition & Calibration & data DB disk.
S. Gadomski, "ATLAS computing in Geneva", journee de reflexion, 14 Sept ATLAS computing in Geneva Szymon Gadomski description of the hardware the.
23/04/2008VLVnT08, Toulon, FR, April 2008, M. Stavrianakou, NESTOR-NOA 1 First thoughts for KM3Net on-shore data storage and distribution Facilities VLV.
1 Data Storage MICE DAQ Workshop 10 th February 2006 Malcolm Ellis & Paul Kyberd.
Large scale data flow in local and GRID environment V.Kolosov, I.Korolko, S.Makarychev ITEP Moscow.
1 Deployment of an LCG Infrastructure in Australia How-To Setup the LCG Grid Middleware – A beginner's perspective Marco La Rosa
Status of WLCG Tier-0 Maite Barroso, CERN-IT With input from T0 service managers Grid Deployment Board 9 April Apr-2014 Maite Barroso Lopez (at)
E-Infrastructure hierarchy Networking and Computational facilities in Armenia ASNET AM Network Armenian National Grid Initiative Armenian ATLAS site (AM-04-YERPHI)
Computing for ILC experiment Computing Research Center, KEK Hiroyuki Matsunaga.
Status of the DESY Grid Centre Volker Guelzow for the Grid Team DESY IT Hamburg, October 25th, 2011.
Computing Infrastructure Status. LHCb Computing Status LHCb LHCC mini-review, February The LHCb Computing Model: a reminder m Simulation is using.
INTRODUCTION The GRID Data Center at INFN Pisa hosts a big Tier2 for the CMS experiment, together with local usage from other HEP related/not related activities.
A short introduction to the Worldwide LHC Computing Grid Maarten Litmaath (CERN)
Quarterly report ScotGrid Quarter Fraser Speirs.
Overview of day-to-day operations Suzanne Poulat.
LCG Service Challenge Phase 4: Piano di attività e impatto sulla infrastruttura di rete 1 Service Challenge Phase 4: Piano di attività e impatto sulla.
11/30/2007 Overview of operations at CC-IN2P3 Exploitation team Reported by Philippe Olivero.
Your university or experiment logo here Caitriana Nicholson University of Glasgow Dynamic Data Replication in LCG 2008.
Organisation Management and Policy Group (MPG): Responsible for setting and policy decisions and resolving any issues concerning fractional usage, acceptable.
The LHC Computing Grid – February 2008 The Worldwide LHC Computing Grid Dr Ian Bird LCG Project Leader 25 th April 2012.
Main title ERANET - HEP Group info (if required) Your name ….
Main title HEP in Greece Group info (if required) Your name ….
WLCG Service Requirements WLCG Workshop Mumbai Tim Bell CERN/IT/FIO.
GridPP Deployment & Operations GridPP has built a Computing Grid of more than 5,000 CPUs, with equipment based at many of the particle physics centres.
And Tier 3 monitoring Tier 3 Ivan Kadochnikov LIT JINR
WLCG GridKa+T2s Workshop Site Report --- Presenter, Site, Country.
Ian Bird LHC Computing Grid Project Leader LHC Grid Fest 3 rd October 2008 A worldwide collaboration.
Tim 18/09/2015 2Tim Bell - Australian Bureau of Meteorology Visit.
BNL Tier 1 Service Planning & Monitoring Bruce G. Gibbard GDB 5-6 August 2006.
Caitriana Nicholson, CHEP 2006, Mumbai Caitriana Nicholson University of Glasgow Grid Data Management: Simulations of LCG 2008.
…building the next IT revolution From Web to Grid…
Grid User Interface for ATLAS & LHCb A more recent UK mini production used input data stored on RAL’s tape server, the requirements in JDL and the IC Resource.
BalticGrid-II Project BalticGrid-II Kick-off meeting, May 2008, Vilnius SA2 - Network Provisioning Katrina Sataki, IMCS UL.
1 User Analysis Workgroup Discussion  Understand and document analysis models  Best in a way that allows to compare them easily.
CERN IT Department CH-1211 Genève 23 Switzerland t Frédéric Hemmer IT Department Head - CERN 23 rd August 2010 Status of LHC Computing from.
CERN-IT Oracle Database Physics Services Maria Girone, IT-DB 13 December 2004.
Computing Coordination Aspects for HEP in Germany International ICFA Workshop on HEP Networking, Grid and Digital Divide Issues for Global e-Science nLCG.
1 Andrea Sciabà CERN Critical Services and Monitoring - CMS Andrea Sciabà WLCG Service Reliability Workshop 26 – 30 November, 2007.
Slide David Britton, University of Glasgow IET, Oct 09 1 Prof. David Britton GridPP Project leader University of Glasgow UK-T0 Meeting 21 st Oct 2015 GridPP.
Status Organization Overview of Program of Work Education, Training It’s the People who make it happen & make it Work.
Testing and integrating the WLCG/EGEE middleware in the LHC computing Simone Campana, Alessandro Di Girolamo, Elisa Lanciotti, Nicolò Magini, Patricia.
U.S. ATLAS Computing Facilities Overview Bruce G. Gibbard Brookhaven National Laboratory U.S. LHC Software and Computing Review Brookhaven National Laboratory.
Tier 1 at Brookhaven (US / ATLAS) Bruce G. Gibbard LCG Workshop CERN March 2004.
Participation of JINR in CERN- INTAS project ( ) Korenkov V., Mitcin V., Nikonov E., Oleynik D., Pose V., Tikhonenko E. 19 march 2004.
Ian Bird WLCG Networking workshop CERN, 10 th February February 2014
Data transfers and storage Kilian Schwarz GSI. GSI – current storage capacities vobox LCG RB/CE GSI batchfarm: ALICE cluster (67 nodes/480 cores for batch.
CMS: T1 Disk/Tape separation Nicolò Magini, CERN IT/SDC Oliver Gutsche, FNAL November 11 th 2013.
ATLAS Distributed Computing ATLAS session WLCG pre-CHEP Workshop New York May 19-20, 2012 Alexei Klimentov Stephane Jezequel Ikuo Ueda For ATLAS Distributed.
WLCG Status Report Ian Bird Austrian Tier 2 Workshop 22 nd June, 2010.
Meeting with University of Malta| CERN, May 18, 2015 | Predrag Buncic ALICE Computing in Run 2+ P. Buncic 1.
Dominique Boutigny December 12, 2006 CC-IN2P3 a Tier-1 for W-LCG 1 st Chinese – French Workshop on LHC Physics and associated Grid Computing IHEP - Beijing.
SAM architecture EGEE 07 Service Availability Monitor for the LHC experiments Simone Campana, Alessandro Di Girolamo, Nicolò Magini, Patricia Mendez Lorenzo,
November 28, 2007 Dominique Boutigny – CC-IN2P3 CC-IN2P3 Update Status.
The status of IHEP Beijing Site WLCG Asia-Pacific Workshop Yaodong CHENG IHEP, China 01 December 2006.
ATLAS Computing: Experience from first data processing and analysis Workshop TYL’10.
1-2 March 2006 P. Capiluppi INFN Tier1 for the LHC Experiments: ALICE, ATLAS, CMS, LHCb.
ScotGRID is the Scottish prototype Tier 2 Centre for LHCb and ATLAS computing resources. It uses a novel distributed architecture and cutting-edge technology,
Bob Jones EGEE Technical Director
A Dutch LHC Tier-1 Facility
Report from WLCG Workshop 2017: WLCG Network Requirements GDB - CERN 12th of July 2017
Grid site as a tool for data processing and data analysis
Belle II Physics Analysis Center at TIFR
Jan 12, 2005 Improving CMS data transfers among its distributed Computing Facilities N. Magini CERN IT-ES-VOS, Geneva, Switzerland J. Flix Port d'Informació.
Data Challenge with the Grid in ATLAS
EGEE VO Management.
Grid Computing.
Simulation use cases for T2 in ALICE
LHC Data Analysis using a worldwide computing grid
The LHCb Computing Data Challenge DC06
Presentation transcript:

A Computing Tier 2 Node Eric Fede – LAPP/IN2P3

2 Eric Fede – 1st Chinese-French Workshop Plan What is a Tier 2 –Context and definition To be a Tier 2 –Objectives and commitments Building a Tier 2 –Tier 2 specifications –Infrastructures –Team Daily works on Tier 2 Conclusion

3 Eric Fede – 1st Chinese-French Workshop What is a Tier 2 Context –To provide computing and storage needed to process and analyse the data gathered by LHC, the Worldwide LHC Grid Computing project was defined. –WLCG define 4 type of computing resource : Tier 0, Tier 1, Tier 2 and Tier 3 Tier 2 definition (usual version) –Centres consisting of one or several collaborating computing facilities, which can store sufficient data and provide adequate computing power for end-user analysis tasks and Monte Carlo simulation. –Share between analysis and simulation depend of the application.  Only simulation for LHCb  Both for ATLAS and CMS

4 Eric Fede – 1st Chinese-French Workshop What is a Tier 2 Tier-1 Data acquisition Tier-2

5 Eric Fede – 1st Chinese-French Workshop How to be a Tier2 ?

6 Eric Fede – 1st Chinese-French Workshop To be a Tier 2 Tiers 2 have to declare to be agree with a Memorandum Of Understanding who define roles, objectives and commitment of Tiers 2. – Provisioning –Provisioning a managed disk storage  Providing permanent data storage for files and database.  Providing temporary data storage for files and database.  This storage should be “grid enable” according to standards –Provisioning a access to the stored data by other centre of the grid  Any Tier 2 may have access data to any Tier 1  In particular, the network bandwidth and services for data exchange with associate Tier 1 should be ensure.

7 Eric Fede – 1st Chinese-French Workshop To be a Tier 2 –Provisioning a processing capacity  Should be access by the grid standards –Provisioning some services according to the experiment requirements  Sharing grid services (resource broker,…..)  Dedicated to experiences ( files catalogue, VOBox,…)  … Level of availability –Average availability measured on annual basis need to be upper at 95 % ( 18 days per year of unavailability) –Responding delay to operational problems : from 2 hours to 72 depending of the services affected and the periods where the problem occur. Tier’s 2 have a real long term role of services provider

8 Eric Fede – 1st Chinese-French Workshop Building a Tier 2

9 Eric Fede – 1st Chinese-French Workshop Building a Tier 2 Tier 2 workspace The site has to ask himself some questions about her Tier 2 to define which Tier 2 it want being. –Which applications will be accepted on the T2 ?  Could influenced the Tier capacity and organisation For the LCG experiences : from 1 to 4. Majority of Tier 2 supported one or more none LHC VO. –To be or not be a 100% grid site ?  Share resources between grid users and non grid users Some sites need to provide some computing resources out of grid aspects. oA part of the resources (storage, computing) can be acceded by a none grid process

10 Eric Fede – 1st Chinese-French Workshop Building a Tier 2 –What are the constraints of my site ?  In terms of infrastructure.  In terms of manpower.  Both aspects are for a part linked to the next point. –The budget aspect  The financing of the Tier is perennial or not ?  Useful to anticipate the infrastructure and resources evolution.

11 Eric Fede – 1st Chinese-French Workshop Building a Tier 2 –What type of tier we want ?  A site = a Tier 2 This is the first view that we have about a tier2.  A set of sites = one Tier 2 On this view a Tier 2 is a federation of site. Must establish unified technical and operational management.  Today 50 % of the Tiers 2 are a federation of sites ( from 2 to 8 geographical sites ) Italy Tiers 2 are dedicated site. UK Tiers 2 are all a federation of site. France, both approach are done. –… On black, information extracted from a survey of T2s done this summer (25 answers)

12 Eric Fede – 1st Chinese-French Workshop Building a Tier 2 Infrastructure requested (hardware) –Networking  A fast network connection to national research an education networks ( in order to be connected to the associate Tier 1).  Today, all the Tier 2 ( which have participates to the survey) have at less 1Gb/s.  A fast network core to interconnect the Tier 2 elements/services. This task can spend many time because the number of partner involved on network infrastructure can be large. On black, information extracted from a survey of T2s done this summer (25 answers)

13 Eric Fede – 1st Chinese-French Workshop Building a Tier 2 Infrastructure requested (hardware) –Computing  A significant computing power.  No rules exist about the lower limit to be a Tier 2, but : ∑ Tier 2 = ∑ Tier 1  Today the Tier2 which support only one VO provide (or will provide) more than 400 cpu and more than 800 cpu for the Tier 2 which support 4 VO. Wait the last period for buying the computing can give the feeling that we have more performances for less cpu, but growing resources too fast is not easy. On black, information extracted from a survey of T2s done this summer (25 answers)

14 Eric Fede – 1st Chinese-French Workshop Building a Tier 2 Infrastructure requested (hardware) –Storage  Tier 2 will provide only disk storage, no MSS.  The size of the storage is from 50 TB to 800 TB.  Not really proportional to the number of VO. –Computing room infrastructure  Probably the most important infrastructure element, because other ( computing, storage,…) are dependent on her.  Space on computer rooms.  Cooling.  Electric power. The principal reason of downtime during the last year on LCG sites is related to a cooling or power problem.

15 Eric Fede – 1st Chinese-French Workshop Building a Tier 2 Infrastructure requested (hardware) Concerning the hardware infrastructure, the word to keep on mind is ANTICIPATE  Power, cooling requirements  Computing, storage provisioning  Time to obtain an adequate network link

16 Eric Fede – 1st Chinese-French Workshop Building a Tier 2 Infrastructure requested (services) –Deployment and configuration of  Operating system.  Grid middleware.  Maybe application software.  Not trivial because size of resources ( x hundred of cpu and disk) grid services complexity grid services dependency. a coherent system is essentiel  Needs expertises.  The usage of some tools can be useful. Yaim, Quattor,…..

17 Eric Fede – 1st Chinese-French Workshop Building a Tier 2 –Storage services  Management of the local storage solution (SAN, NAS, Castor,…).  Management of the sharing file system. Use by workers to “mount” some storage area (application software, working directory,…) Many candidate (GPFS, Luste, AFS,…)  Management of the grid storage service (DPM, dcache,…). Keeping the namespace coherent with the data  Management of the channels for access to storage. Finding the good parameter to optimize the transfer ( # stream) –Computing services  The job manager choice (PBS, LSF,Torque, BQS,…).  Definition of the scheduler rules. The efficiency of the site (computing point of view) depend of this.  Job parallel compatibility (MPI deployment)

18 Eric Fede – 1st Chinese-French Workshop Building a Tier 2 –Monitoring services  The control and the management of the Tier 2 require having a global view of the resource. Services overloaded (RB), scheduling efficiency, storage availability,..  Required to define what are the interesting parameters to supervise  Important in order to “solve” the problems before they comes and consequently increase the Tier 2 availability. –Control services  Mainly the control of the hardware resources (IPMI controller) Shutting down properly resources in case of power failure. Starting properly resources when the power come back. ….

19 Eric Fede – 1st Chinese-French Workshop Building a Tier 2 Team –The main activities of the people involved on the Tier 2 are operational activities.  Involved on the deployment (hardware and software).  Services maintenance and migration.  Support for local user (not really requested by the Tier 2 activity, but in practice you have to do a part of the users support). –A part of the teams are also involved on the relations and coordination with other LCG partner and sites. –Majority of the Tier 2 have between 4 and 6 FTE (Full Time Equivalent). But change strongly from site to site. Expertise of people involved on the teams is crucial to the good Tier 2 life.

20 Eric Fede – 1st Chinese-French Workshop Daily works This is the work of every computing centre Management of resources –Optimize the resources usage.  By a tuning of the services and resources. –Evolution and migration of the resources.  New hardware, new services. Security actions – Follow the WLCG security recommendations and actions. Operation actions –Coordination with the associate Tier 1 and with others WLCG Tiers. –Follow the grid software news, because WLCG services are alive. Some of them can die, some new will have to be deployed. –…..

21 Eric Fede – 1st Chinese-French Workshop Conclusion A Tier 2 is : –A commitment on WLCG project. –An hardware infrastructure. –Providing and managing services. –A daily work to keep the system up and running. –Teams and technical competencies. Keep on mind that a Tier 2 on the next years will be what was a Tier 1 today.

22 Eric Fede – 1st Chinese-French Workshop THANK YOU QUESTIONS ?