Grid for CBM Kilian Schwarz, GSI

What is Grid ? ● Sharing of distributed resources within one Virtual Organisations !!!!

Europa:267 Institute, 4603 User Sonstige:208 Institute, 1632 User LHC Wissenschaftler weltweit

Start of CBM Grid ● There are considerations to start a CBM Grid ● Task: distributed MC production ● Potential sites: 3 (Bergen, Dubna, GSI) ● After positive experiences the Grid can be enlarged to more sites and tasks, like distributed analysis

requirements * Globus-style X509 user certificates issued for CBM by GermanGrid CA * How to get a certificate ? at GSI: >. globuslogin > grid-cert-request –cn “ ” certificate request file and private key will be stored in $HOME/.globus The request file has to be signed (openssl) by the CA responsible person and mailed to GermanGrid CA The certificate will be mailed back via

GermanGrid CA How to get a certificate in detail: See

requirements: CBM VO Server (one per VO) additional sites: - Bergen, Dubna additional users: - to be added

Globus/LCG – creation of grid-mapfile necessary for each site ● E.g. with gLite-security tools: - adjust $GLITE_LOCATION/etc/glite- mkgridmap.conf add: “group ldap://glite001.gsi.de:8389/o=cbm,dc=de,dc= de” - Create grid-mapfile $GLITE_LOCATION/sbin/glite-mkgridmap – output=/etc/grid-security/grid-mapfile

user creation on each site (support of CBM VO) Each site has to create cbm-user-IDs onto which the Grid-users will be mapped: EGEE/LCG: a certain number of POOL accounts, e.g. cbmvo00 – cbmvo10 Globus & AliEn: one production user: via this userID the jobs will be submitted. E.g. cbmprod

CBM software environment ● To be able to send real CBM jobs to the Grid, the participating sites have to * Install the CBM software and prepare the environment * Or the job has to bring it’s own environment (static links)

Agreement on common Grid middleware basically, the possibilities are: - Globus - NorduGrid - LCG-2 - AliEn - gLite (EGEE) - gLite (AliEn)

LHC Computing Grid Project Fundamental Goal of the LCG To help the experiments’ computing projects Phase 1 – prepare and deploy the environment for LHC computing Phase 2 – acquire, build and operate the LHC computing service SC2 – Software & Computing Committee  SC2 includes the four experiments, Tier 1 Regional Centres  SC2 identifies common solutions and sets requirements for the project PEB – Project Execution Board  PEB manages the implementation  organising projects, work packages  coordinating between the Regional Centres

EDG Middleware Architecture Collective Services Information & Monitoring Replica Manager Grid Scheduler Local ApplicationLocal Database Underlying Grid Services Computing Element Services Authorization Authenticatio n and Accounting Replica Catalog Storage Element Services SQL Database Services Fabric services Configuration Management Node Installation & Management Monitoring and Fault Tolerance Resource Management Fabric Storage Management Grid Fabric Local Computing Grid Grid Application Layer Data Management Job Management Metadata Management Service Index APPLICATIONS GLOBUS CondorG (via VDT) M / W

Dubna (JINR): LCG-2 site

Dubna (JINR): LCG-2 site LCG-test mostly successful

JINR (LCG-2 site: job-submit )

Timeline oAfter only 2 years of development, we have deployed a distributed computing environment which meets the needs of Alice experiment Simulation & Reconstruction Event mixing Analysis oUsing Open Source components (representing 99% of the code), internet standards (SOAP,XML, PKI…) and scripting language (perl) was the key element that alllowed quick prototyping and very fast development cycles First production (distributed simulation) 10% DC (analysis) P. Buncic, CERN

Building AliEn P. Saiz, CERN

AliEn Grid (ALICE VO): ● 77 configured sites worldwide

DC Monitoring: ● Monalisa:

lxts05.gsi.de: AliEn client (PANDA VO)

JINR and Bergen: AliEn sites

Grids and Open Standards Increased functionality, standardization Time Custom solutions Open Grid Services Arch GGF: OGSI, … (+ OASIS, W3C) Multiple implementations, including Globus Toolkit Web services Globus Toolkit Defacto standards GGF: GridFTP, GSI X.509, LDAP, FTP, … App- specific Services

Architecture Guiding Principles ● Lightweight (existing) services – Easily and quickly deployable – Use existing services where possible as basis for re-engineering ● Interoperability – Allow for multiple implementations ● Resilience and Fault Tolerance ● Co-existence with deployed infrastructure – Run as an application (e.g. on LCG-2; Grid3) – Reduce requirements on site components ● Basically globus and SRM – Co-existence (and convergence) with LCG-2 and Grid3 are essential for the EGEE Grid service ● Service oriented approach – WSRF still being standardized – No mature WSRF implementations exist to date, no clear picture about the impact of WSRF hence: start with plain WS ● WSRF compliance is not an immediate goal, but we follow the WSRF evolution ● WS-I compliance is important

Approach ● Exploit experience and components from existing projects – AliEn, VDT, EDG, LCG, and others ● Design team works out architecture and design – Architecture: – Design: ● Components are initially deployed on a prototype infrastructure – Small scale (CERN & Univ. Wisconsin) – Get user feedback on service semantics and interfaces ● After internal integration and testing components are delivered to SA1 and deployed on the pre-production service EDGVDT... LCG EGEE...AliEn

gLite (AliEn) * From now on used by ALICE for globally distributed analysis in connection with PROOF (at GSI:  PROOF at GSI )

gLite (EGEE) * Will replace LCG-2.X in near? future, but nobody has real experience with it

summary (middlewares) ● LCG-2: GSI and Dubna - pro: large distribution, support - contra: difficult to set up, no distributed analysis ● AliEn: GSI, Dubna, Bergen - pro: in production since contra: unsecure future, no support Globus 2: GSI, Dubna, Bergen? - pro/contra: simple, but functioning (no RB, no FC, no support) gLite/GT4: new on the market - pro/contra: nobody has production experience (gLite)

lxg01-05.gsi.de ● LCG test installation, visible in LCG – preproduction testbed ● Trying to port LCG to Debian Linux