1 Soonwook Hwang, Hangi Kim KISTI Seyong Kim Sejong University XQCD Workshop 2009 August 3-5, 2009 Large-scale Deployment of two-color QCD on the FKPPL.

Slides:



Advertisements
Similar presentations
ATLAS/LHCb GANGA DEVELOPMENT Introduction Requirements Architecture and design Interfacing to the Grid Ganga prototyping A. Soroko (Oxford), K. Harrison.
Advertisements

FP7-INFRA Enabling Grids for E-sciencE EGEE Induction Grid training for users, Institute of Physics Belgrade, Serbia Sep. 19, 2008.
GRID INTEROPERABILITY USING GANGA Soonwook Hwang (KISTI) YoonKee Lee and EunSung Kim (Seoul National Uniersity) KISTI-CCIN2P3 FKPPL Workshop December 1,
INFSO-RI Enabling Grids for E-sciencE EGEE Middleware The Resource Broker EGEE project members.
The EPIKH Project (Exchange Programme to advance e-Infrastructure Know-How) gLite Grid Services Abderrahman El Kharrim
1 Kihyeon Cho & Soonwook Hwang (KISTI) Super Belle with FKPPL VO & AMGA Data Handling.
1 Bridging Clouds with CernVM: ATLAS/PanDA example Wenjing Wu
Makrand Siddhabhatti Tata Institute of Fundamental Research Mumbai 17 Aug
LHC Experiment Dashboard Main areas covered by the Experiment Dashboard: Data processing monitoring (job monitoring) Data transfer monitoring Site/service.
KISTI’s Activities on the NA4 Biomed Cluster Soonwook Hwang, Sunil Ahn, Jincheol Kim, Namgyu Kim and Sehoon Lee KISTI e-Science Division.
5 November 2001F Harris GridPP Edinburgh 1 WP8 status for validating Testbed1 and middleware F Harris(LHCb/Oxford)
FESR Consorzio COMETA Grid Introduction and gLite Overview Corso di formazione sul Calcolo Parallelo ad Alte Prestazioni (edizione.
11 ALICE Computing Activities in Korea Beob Kyun Kim e-Science Division, KISTI
F.Fanzago – INFN Padova ; S.Lacaprara – LNL; D.Spiga – Universita’ Perugia M.Corvo - CERN; N.DeFilippis - Universita' Bari; A.Fanfani – Universita’ Bologna;
INFSO-RI Enabling Grids for E-sciencE Logging and Bookkeeping and Job Provenance Services Ludek Matyska (CESNET) on behalf of the.
EGEE-II INFSO-RI Enabling Grids for E-sciencE Grid application development with gLite and P-GRADE Portal Miklos Kozlovszky MTA SZTAKI.
The EDGI project receives Community research funding 1 EDGI Brings Desktop Grids To Distributed Computing Interoperability Etienne URBAH
:: ::::: ::::: ::::: ::::: ::::: ::::: ::::: ::::: ::::: ::::: ::::: :: GridKA School 2009 MPI on Grids 1 MPI On Grids September 3 rd, GridKA School 2009.
Enabling Grids for E-sciencE ENEA and the EGEE project gLite and interoperability Andrea Santoro, Carlo Sciò Enea Frascati, 22 November.
1 Overview of the Application Hosting Environment Stefan Zasada University College London.
Grid Technologies  Slide text. What is Grid?  The World Wide Web provides seamless access to information that is stored in many millions of different.
DataGrid WP1 Massimo Sgaravatto INFN Padova. WP1 (Grid Workload Management) Objective of the first DataGrid workpackage is (according to the project "Technical.
INFSO-RI Enabling Grids for E-sciencE Workload Management System Mike Mineter
Belle MC Production on Grid 2 nd Open Meeting of the SuperKEKB Collaboration Soft/Comp session 17 March, 2009 Hideyuki Nakazawa National Central University.
Group 1 : Grid Computing Laboratory of Information Technology Supervisors: Alexander Ujhinsky Nikolay Kutovskiy.
The EDGeS project receives Community research funding 1 SG-DG Bridges Zoltán Farkas, MTA SZTAKI.
Grid infrastructure analysis with a simple flow model Andrey Demichev, Alexander Kryukov, Lev Shamardin, Grigory Shpiz Scobeltsyn Institute of Nuclear.
F. Fassi, S. Cabrera, R. Vives, S. González de la Hoz, Á. Fernández, J. Sánchez, L. March, J. Salt, A. Lamas IFIC-CSIC-UV, Valencia, Spain Third EELA conference,
Stuart Wakefield Imperial College London Evolution of BOSS, a tool for job submission and tracking W. Bacchi, G. Codispoti, C. Grandi, INFN Bologna D.
Status of the LHCb MC production system Andrei Tsaregorodtsev, CPPM, Marseille DataGRID France workshop, Marseille, 24 September 2002.
November SC06 Tampa F.Fanzago CRAB a user-friendly tool for CMS distributed analysis Federica Fanzago INFN-PADOVA for CRAB team.
David Adams ATLAS ADA, ARDA and PPDG David Adams BNL June 28, 2004 PPDG Collaboration Meeting Williams Bay, Wisconsin.
EGEE-III INFSO-RI Enabling Grids for E-sciencE EGEE and gLite are registered trademarks WMSMonitor: a tool to monitor gLite WMS/LB.
June 24-25, 2008 Regional Grid Training, University of Belgrade, Serbia Introduction to gLite gLite Basic Services Antun Balaž SCL, Institute of Physics.
EGEE-III INFSO-RI Enabling Grids for E-sciencE Feb. 06, Introduction to High Performance and Grid Computing Faculty of Sciences,
T3 analysis Facility V. Bucard, F.Furano, A.Maier, R.Santana, R. Santinelli T3 Analysis Facility The LHCb Computing Model divides collaboration affiliated.
Overview of grid activities in France in relation to FKPPL FKPPL Workshop Thursday February 26th, 2009 Dominique Boutigny.
US LHC OSG Technology Roadmap May 4-5th, 2005 Welcome. Thank you to Deirdre for the arrangements.
1 Grid2Win: porting of gLite middleware to Windows Dario Russo INFN Catania
Glite. Architecture Applications have access both to Higher-level Grid Services and to Foundation Grid Middleware Higher-Level Grid Services are supposed.
Development of e-Science Application Portal on GAP WeiLong Ueng Academia Sinica Grid Computing
FP6−2004−Infrastructures−6-SSA E-infrastructure shared between Europe and Latin America Grid2Win: Porting of gLite middleware to.
EGEE-III INFSO-RI Enabling Grids for E-sciencE EGEE and gLite are registered trademarks Grid2Win : gLite for Microsoft Windows Roberto.
The GridPP DIRAC project DIRAC for non-LHC communities.
EGEE-II INFSO-RI Enabling Grids for E-sciencE Command Line Grid Programming Spiros Spirou Greek Application Support Team NCSR “Demokritos”
Global ADC Job Monitoring Laura Sargsyan (YerPhI).
ALICE-Korea Activities and Plans Yong-Wook Baek Kangnung National University* and L.P.C. Clermont-Ferrand* *France-Korea Particle Physics Laboratory (International.
INFSO-RI Enabling Grids for E-sciencE Using of GANGA interface for Athena applications A. Zalite / PNPI.
D.Spiga, L.Servoli, L.Faina INFN & University of Perugia CRAB WorkFlow : CRAB: CMS Remote Analysis Builder A CMS specific tool written in python and developed.
The GridPP DIRAC project DIRAC for non-LHC communities.
1 Tutorial:Initiation a l’Utilisation de la Grille EGEE/LCG, June 5-6 N. De Filippis CMS tools for distributed analysis N. De Filippis - LLR-Ecole Polytechnique.
Breaking the frontiers of the Grid R. Graciani EGI TF 2012.
EGEE-II INFSO-RI Enabling Grids for E-sciencE Overview of gLite, the EGEE middleware Mike Mineter Training Outreach Education National.
Consorzio COMETA - Progetto PI2S2 UNIONE EUROPEA Grid2Win : gLite for Microsoft Windows Elisa Ingrà - INFN.
Geant4 Simulation Study for Proton Therapy Proton Therapy Center Se Byeong Lee Medical Physicist, Ph.D. The 3 rd FKPPL workshop, March 8-9, 2011, Seoul,
Geant4 GRID production Sangwan Kim, Vu Trong Hieu, AD At KISTI.
INFN/IGI contributions Federated Clouds Task Force F2F meeting November 24, 2011, Amsterdam.
Antonio Fuentes RedIRIS Barcelona, 15 Abril 2008 The GENIUS Grid portal.
The EPIKH Project (Exchange Programme to advance e-Infrastructure Know-How) gLite Grid Introduction Salma Saber Electronic.
Enabling Grids for E-sciencE Claudio Cherubino INFN DGAS (Distributed Grid Accounting System)
GRID & Parallel Processing Koichi Murakami11 th Geant4 Collaboration Workshop / LIP - Lisboa (10-14/Oct./2006) 1 GRID-related activity in Japan Go Iwai,
1 March 8-9, 2011 Soonwook Hwang KISTI FKPPL VO: Status and Perspectives of Grid Infrastructure The 3 rd FKPPL KIAS.
EGI-InSPIRE RI EGI-InSPIRE EGI-InSPIRE RI EGI solution for high throughput data analysis Peter Solagna EGI.eu Operations.
U.S. ATLAS Grid Production Experience
INFN-GRID Workshop Bari, October, 26, 2004
Short update on the latest gLite status
LCG middleware and LHC experiments ARDA project
Particle Physics at KISTI
e-Science paradigm for CDF experiment
e-Science for High Energy Physics
Presentation transcript:

1 Soonwook Hwang, Hangi Kim KISTI Seyong Kim Sejong University XQCD Workshop 2009 August 3-5, 2009 Large-scale Deployment of two-color QCD on the FKPPL VO using Ganga

Outline  FKPPL VO (Virtual Organization) Grid  Ganga  High-level Grid job submission and management tool  Deployment of QCD Simulations on the Grid

FKPPL (France Korea Particle Physics Laboratory)  International Associated Laboratory between France and Korea  Promote joint cooperative activities (research projects) under a scientific research program in the area of  Particle Physics  LHC  ILC  E-Science  Bioinformatics  Grid Computing

FKPPL Scientific Projects Project nameCoordinatorsPartners ILC calorimeter (particle physics) Yongmann Yang, Ewha Jean-Claude Brient, LLR Ewha Womans Univ., Kangnung Nat. U niv.,LPC, LLR ILC electronics (particle physics) Jongseo Chai, SKKU Christoph de la Taille, LAL Sung Kyun Kwan Univ., Korea Institute of Radiological and medical Sciencces, Pohang Accel. Lab. LAL, LLR Grid computingS. Hwang, KISTI D. Boutigny, CC-IN2P3 KISTI, CC-IN2P3 WISDOM (in silico drug discovery) Doman Kim, CNU V. Breton, LPC Chonnam Nat. Univ., KISTI, Kangnung Nat. Univ., LPC ALICE ( heavy ions physics) Yongwook Baek, KNU Pascal Dupieux, LPC Kangnung Nat. Univ. LPC CDF (particle physics) Kihyeon Cho, KISTI Aurore Savoy-Navarro, LPNH E KISTI, LPNHE

FKPPL VO Grid

Objectives  Provide computing facilities needed to foster FKPPL scientific applications and experiments  Provide researchers and scientists in Korea and France with a production grid environment

FKPPL VO Grid Testbed ServiceHostSite UIkenobi.kisti.re.krKISTI VOMSpalpatine.kisti.re.krKISTI WMS/LBsnow.kisti.re.krKISTI SEccsrm02.in2p3.fr (0.5TB)CC-IN2P3 hansolo.kisti.re.kr (1.5TB)KISTI CEcclcgceli03.in2p3.fr (8000 CPU cores)CC-IN2P3 darthvader.kisti.re.kr (64 CPU cores)KISTI VOMS WMS CESEUICESE FKPPL VO KISTI IN2P3 LFC WIKI

gLite Grid Services on FKPPL VO User Interface (UI) User Interface (UI): The place where users logon to access the Grid Workload Management System (WMS) Workload Management System (WMS): Matches the user requirements with the available resources on the Grid File and replica catalog File and replica catalog: Location of grid files and grid file replicas Computing Element (CE) Computing Element (CE): A batch queue on a site’s computers where the user’s job is executed Storage Element (SE) Storage Element (SE): provides (large-scale) storage for files

9 Computing Element Storage Element IN2P3 Information System Submit job (executable + small inputs) Submit job query Retrieve output WMS User Interface publish state File and Replica Catalog VO Management Service (DB of VO users) query create proxy process Retrieve status & (small) output files Logging and bookkeeping Job status Logging Input file(s) Output file(s) Register file Job Submission Example

Utilization – CE Services # of jobsCPU used (SI2K Hour) Elapsed time (SI2K Hour) Oct Nov ,250- Dec ,410- Jan ,859,6402,092,550 Feb ,167,3507,732,280 Mar ,937,9205,068,990 Apr ,130,2405,666,100 May ,21050,560 Jun ,269,4403,593,100

User Support  FKPPL VO Wiki site   User Accounts on UI  20 User accounts has been created  FKPPL VO Membership Registration  7 Users have been registered at FKPPL VO membership

Application Support on FKPPL VO  Deployment of Geant4 applications  Cancel Treatment Plan  In collaboration with Dr. Jungwook Shin of National Cancer Center in Korea  Deployment of two-color QCD simulations  In collaboration with Prof. Seyong Kim of Sejong University

Introduction to Ganga

Ganga  easy-to-use user interface for job submission and management to  Specification, submission, bookkeeping, and post processing of computational tasks on a wide set of distributed resources  Provides a homogeneous environment for processing data on heterogeneous resources Grid Localho st Batch GangaGanga LCG/gLite Backend PBS or SGE Backend Local Backend LCG/gLite Cmd or Lib PBS or SGE Cmd or Lib Local Cmd or Lib Athena GAUDI ROOT Executabl e

Overall Architecture  The user interact with the Ganga Public Interface via  GUI  CLIP  Script  Plugins provided for different execution backends (e.g., Grid, Batch and Local) and applications (e.g., ROOT, GAUDI and Athena)  easily extended and customized to meet the needs of different user communities

Uniform Interface to Heterogeneous Resources  For the user, running a job interactively is no different than running on the Grid

Deployment of QCD Simulations on the FKPPL VO

Our Two-color QCD Applications (1/2)  Large-scale  A large number of simulation jobs to be run with a wide range of different parameters  In our case, we have planned to run a total of 360 different QCD jobs with a different parameter set  beta = [0.9, 1.0, 1.1, 1.2, 1.3, 1.4, 1.5, , 1.75, 1.8, 1.85, 1.9, 1.93, 1.95, 1.98, 2.0, 2.1, 2.2, 2.3] (20)  J = [0.04, 0.05, 0.06] (3)  mu = [0.0, 0.575, 0.65, 0.85, 0.9, 1] (6)  Independent  Each job runs independently  Long-duration  Each QCD job goes through 200 steps to complete, each step taking an average of 1 Hour, so each QCD job takes an average of 200 hours

 Need a computing facility to run a large number of jobs  FKPPL VO provides computing resources sufficient to run the 360 QCD jobs all together concurrently  Need some grid tool to effectively maintain such a large-scale jobs running on the Grid without having to know the details of the underlying Grid  Ganga seems to be appropriate as a tool for managing such large number of jobs on the Grid Our Two-color QCD Applications (2/2)

Issues relating to long-running jobs on the Grid  Long-running jobs often fail to complete on the Grid  It is not straightforward to successfully get done a long-duration job like our two-color QCD simulation on the Grid  A Grid proxy certificate expires before the job’s completion  By default, the proxy has a lifetime of 12 hours  Each Grid site has its own site operational policy such as the maximum CPU time for a job to be allowed to run at a time SitesCE NodesMaxCPUTime (Min) CC-IN2P3 Jobmanager-bqs-short16 Jobmanager-bqs-medium751 Jobmanager-bqs-long4753 KISTIJobmanager-lcgpbs-fkppl4880

Application-level Checkpointing/Restarting  We have modified the original two-color QCD simulation code to support an application- level checkpointing scheme  The two color QCD code takes 200 steps to complete  Once a QCD job is launched successfully on the Grid, a intermediate result is generated at each step and saved into the checkpoint server  When a QCD job is detected to be stopped for some reason,  Ganga restarts it from where it has left by resubmitting it along with the latest intermediate result

22 Computing Element Storage Element IN2P3 (Re)Submit QCD job (executable + small inputs) Submit QCD job Retrieve output WMS Su2.x Retrieve status & (small) output files Input file(s) Output file(s) Overview of QCD Simulation Runs on the Grid Heartbeat Monitor Check status & Intermediate Result send intermediate result CheckpointServer Retrieve the latest Intermediate result

Distribution of QCD Jobs on the FKPPL VO

Two-color QCD Production  ~ 8.2 CPU years  360 runs * 200 step * 1 hours = hours  ~ 360 concurrent QCD runs on the FKPPL VO  As of now (Aug. 03), 51.70% has been done out of a total of 72,000 steps

Summary  FKPPL VO Grid  provides a production-level Grid infrastructure for scientists to carry out relatively large-scale simulation runs  Ganga/Application-level checkpointing  makes it straightforward to run a long-running jobs on the Grid on a large scale

Thank you