HEP computing in Korea

https://doi.org/10.1016/j.cpc.2007.02.063Get rights and content

Abstract

The current status of High Energy Physics (HEP) computing system and plan of HEP data grid in Korea are reported. The main infrastructure is located at the Center for High Energy Physics at Kyungpook National University for the experiment of Compact Muon Solenoid (CMS), Collider Detector at Fermilab (CDF), Belle and the Alpha Magnetic Spectrometer (AMS). Recently, we have been concentrating on computing in CMS experiment. The current status of regional data center in Korea and Service Challenge 4 (SC4) for CMS experiment are also reported.

Section snippets

Introduction: Korean HEP activities

Large-scale enterprise experiments in which Koreans are involved now and will be in the future are Belle and K2K at KEK in Japan (in progress), CDF at Fermilab in USA (in progress), AMS on the International Space Station (ISS) (data taking starts in 2007), CMS at CERN in Europe (data taking starts in 2007) and other experiments, such as PHENIX at BNL, ZEUS at DESY, D0 at Fermilab, as well as ALICE, OPERA, KIMS, STAR etc. International Linear Collider experiment might start in mid 2010s.

Korean HEP institutions

Belle is an experiment at the KEK B-factory. Its goal is to study the origin of CP violation in B-meson Decays [1]. Korean Belle collaboration consists of Korea University, Yonsei University, Ewha University, Seoul National University, Sungkyunkwan University, Chonnam National University, Kyungpook National University and Gyeongsang National University.

The CDF experimental collaboration is committed to studying high energy particle collisions at the world's highest energy particle accelerator.

Korean HEP computing and Grid R&D

The Center for High Energy Physics (CHEP) at Kyungpook National University (KNU) is the place where most of these experimental activities are centered. Naturally the computing resources and R&D activities are based on the CHEP activities. CHEP is building a supercomputing center for HEP to be equipped with

CPU> 1,000
Disk cache1,000 TB
Tape storage3,000 TB
Networking> 20 Gbps
Technologyin Grid
Timebefore 2007

Data grid R&D and collaborations in Korea are

Domestic grid WGsince 2001
HEP/Physics

CHEP resources

Recently, CHEP has participated in the LCG Service Challenge 4 (SC4) as a Tier-2 (T2) center. SC4 started in June and ended on September 30. The minimum requirement for a Tier-1 or for a Tier-2 center for CMS experiment is

Tier-1Tier-2
CPU2.5 MSi2K2.5 MSi2K
Disk1.2 PB200 TB
Mass storage2.8 PB
WANin: 7.2 GBbps>1 Gbps
out: 3.5 GBbps
CPU node IO BW1 Gbps1 Gbps

At the moment CHEP has 230 kSi2K CPU, 51 TB hard disk and 45 TB mass storage system. These are about 25% of total requirement of a CMS Tier-2

Summary

Korean HEP computing is now in the phase from base configuration to more intensified system. HEP data grid R&D and network tests have been performed in collaboration with international and domestic partners. Enough bandwidth with new technology such as GLORIAD 10 Gbps are ready and will be up to 40 Gbps in 2008. To meet the demands in Korean HEP some HEP data grid collaborations and HEP working groups for networks are formed. They are supported by governments (MIC, MOST) and various groups of

Acknowledgements

The efforts and supports of the people involved in Korean HEP computing are gratefully acknowledged: MOST, KOSEF, MIC, KISTI, NCA, KISDI, KT/KOREN-NOC, IBM-Korea, CIES, ANF, APAN, KREONET.

This work is partly supported by CHEP(SRC) and by Kyungpook National University Research Fund, 2004.

References (5)

  • M. Aguilar

    Physics Reports

    (2002)
There are more references available in the full text version of this article.

Cited by (0)

View full text