Distributing storage of LHC data - in the nordic countries

Similar documents
Lessons Learned in the NorduGrid Federation

Operating the Distributed NDGF Tier-1

Interoperating AliEn and ARC for a distributed Tier1 in the Nordic countries.

A distributed tier-1. International Conference on Computing in High Energy and Nuclear Physics (CHEP 07) IOP Publishing. c 2008 IOP Publishing Ltd 1

UW-ATLAS Experiences with Condor

Austrian Federated WLCG Tier-2

ARC integration for CMS

N. Marusov, I. Semenov

Andrea Sciabà CERN, Switzerland

Overview of ATLAS PanDA Workload Management

EGEE and Interoperation

EISCAT_3D Support (E3DS) Project.

CernVM-FS beyond LHC computing

Monitoring ARC services with GangliARC

A short introduction to the Worldwide LHC Computing Grid. Maarten Litmaath (CERN)

Computing / The DESY Grid Center

Spanish Tier-2. Francisco Matorras (IFCA) Nicanor Colino (CIEMAT) F. Matorras N.Colino, Spain CMS T2,.6 March 2008"

Data services for LHC computing

dcache: sneaking up on NFS4.1

Status of KISTI Tier2 Center for ALICE

Scientific data management

Future trends in distributed infrastructures the Nordic Tier-1 example

Influence of Distributing a Tier-2 Data Storage on Physics Analysis

Online data storage service strategy for the CERN computer Centre G. Cancio, D. Duellmann, M. Lamanna, A. Pace CERN, Geneva, Switzerland

Clouds in High Energy Physics

The INFN Tier1. 1. INFN-CNAF, Italy

ATLAS NorduGrid related activities

A scalable storage element and its usage in HEP

Scientific data processing at global scale The LHC Computing Grid. fabio hernandez

Analysis of internal network requirements for the distributed Nordic Tier-1

Summary of the LHC Computing Review

AGIS: The ATLAS Grid Information System

Conference The Data Challenges of the LHC. Reda Tafirout, TRIUMF

Virtualization. A very short summary by Owen Synge

Clouds at other sites T2-type computing

dcache, activities Patrick Fuhrmann 14 April 2010 Wuppertal, DE 4. dcache Workshop dcache.org

Presentation of the LHCONE Architecture document

DESY. Andreas Gellrich DESY DESY,

Introduction to SRM. Riccardo Zappi 1

IEPSAS-Kosice: experiences in running LCG site

Towards sustainability: An interoperability outline for a Regional ARC based infrastructure in the WLCG and EGEE infrastructures

Grid Computing Activities at KIT

WLCG Transfers Dashboard: a Unified Monitoring Tool for Heterogeneous Data Transfers.

Challenges and Evolution of the LHC Production Grid. April 13, 2011 Ian Fisk

From raw data to new fundamental particles: The data management lifecycle at the Large Hadron Collider

Considerations for a grid-based Physics Analysis Facility. Dietrich Liko

Philippe Charpentier PH Department CERN, Geneva

NCP Computing Infrastructure & T2-PK-NCP Site Update. Saqib Haleem National Centre for Physics (NCP), Pakistan

The Grid: Processing the Data from the World s Largest Scientific Machine

The Legnaro-Padova distributed Tier-2: challenges and results

Geant4 on Azure using Docker containers

IllustraCve Example of Distributed Analysis in ATLAS Spanish Tier2 and Tier3

Grid Computing at the IIHE

The ATLAS Tier-3 in Geneva and the Trigger Development Facility

Application of Virtualization Technologies & CernVM. Benedikt Hegner CERN

ISTITUTO NAZIONALE DI FISICA NUCLEARE

The Software Defined Online Storage System at the GridKa WLCG Tier-1 Center

PoS(ACAT2010)039. First sights on a non-grid end-user analysis model on Grid Infrastructure. Roberto Santinelli. Fabrizio Furano.

Data Management for the World s Largest Machine

The LHC Computing Grid

What s new in HTCondor? What s coming? HTCondor Week 2018 Madison, WI -- May 22, 2018

HEP replica management

Evolution of the ATLAS PanDA Workload Management System for Exascale Computational Science

dcache as open-source project showcase for education Tigran Mkrtchyan for dcache team CHEP2018, Sofia,

CC-IN2P3: A High Performance Data Center for Research

UK Tier-2 site evolution for ATLAS. Alastair Dewhurst

ATLAS Nightly Build System Upgrade

Visita delegazione ditte italiane

Tier-2 structure in Poland. R. Gokieli Institute for Nuclear Studies, Warsaw M. Witek Institute of Nuclear Physics, Cracow

Physics Computing at CERN. Helge Meinhard CERN, IT Department OpenLab Student Lecture 27 July 2010

The Grid. Processing the Data from the World s Largest Scientific Machine II Brazilian LHC Computing Workshop

New strategies of the LHC experiments to meet the computing requirements of the HL-LHC era

ALICE Grid Activities in US

The PanDA System in the ATLAS Experiment

The JINR Tier1 Site Simulation for Research and Development Purposes

Experiences with the new ATLAS Distributed Data Management System

The ATLAS EventIndex: Full chain deployment and first operation

Preparing for High-Luminosity LHC. Bob Jones CERN Bob.Jones <at> cern.ch

I Tier-3 di CMS-Italia: stato e prospettive. Hassen Riahi Claudio Grandi Workshop CCR GRID 2011

Scaling Internet TV Content Delivery ALEX GUTARIN DIRECTOR OF ENGINEERING, NETFLIX

Worldwide Production Distributed Data Management at the LHC. Brian Bockelman MSST 2010, 4 May 2010

CERN: LSF and HTCondor Batch Services

Compact Muon Solenoid: Cyberinfrastructure Solutions. Ken Bloom UNL Cyberinfrastructure Workshop -- August 15, 2005

1. Introduction. Outline

150 million sensors deliver data. 40 million times per second

Grids and Security. Ian Neilson Grid Deployment Group CERN. TF-CSIRT London 27 Jan

Past. Inputs: Various ATLAS ADC weekly s Next: TIM wkshop in Glasgow, 6-10/06

Physics Computing at CERN. Helge Meinhard CERN, IT Department OpenLab Student Lecture 21 July 2011

The LHC Computing Grid

Ivane Javakhishvili Tbilisi State University High Energy Physics Institute HEPI TSU

The High-Level Dataset-based Data Transfer System in BESDIRAC

ATLAS Tier-3 UniGe

CERN and Scientific Computing

Grid Data Management

The LHC Computing Grid. Slides mostly by: Dr Ian Bird LCG Project Leader 18 March 2008

Introduction to Programming and Computing for Scientists

The Global Grid and the Local Analysis

LHCb Computing Status. Andrei Tsaregorodtsev CPPM

Installation of CMSSW in the Grid DESY Computing Seminar May 17th, 2010 Wolf Behrenhoff, Christoph Wissing

PROOF-Condor integration for ATLAS

Transcription:

Distributing storage of LHC data - in the nordic countries Gerd Behrmann INTEGRATE ASG Lund, May 11th, 2016

Agenda WLCG: A world wide computing grid for the LHC NDGF: The Nordic Tier 1 dcache: Distributed data storage for e-science 2

Graphics extremetech.com 3

Graphics CERN 4

Graphics CERN 5

Graphics CERN 6

7

World wide LHC computing grid (WLCG) 8

Graphics CERN 9

585,530 cores 536 PB disk 540 PB tape Graphics CERN 9

585,530 cores 536 PB disk 540 PB tape Used to be strictly hierarchical with clearly defined data flows and roles for what is computed where (the MONARCH model). Becoming more dynamic, with remote data access, caching, federation. Still a clear distinction between Tier 1 and Tier 2 in terms of Service level Data retention Contact with experiments Regional support role Graphics CERN 9

Graphics GEANT 10

Agenda WLCG: A world wide computing grid for the LHC NDGF: The Nordic Tier 1 dcache: Distributed data storage for e-science 11

A Tier 1 in the Nordic Countries Computing Element (batch system with remote access) Storage Element (a locally and remotely accessible file system) A tape system A File Transfer Service (transfers files between your tier 1 and other sites) 24x7 monitoring, with 4 hour response time for alarms. A liaison for each supported experiment. Enough man power to join lots of periodic meetings. Enough man power to provide expert support for your region. LHC runtime + 15 years commitment. 12

NT1 aka the Nordic Data Grid Facility A distributed Tier 1-4 countries, 6 sites (7 originally) [ + Slovenia]. Build on the foundations of NorduGrid. First production scale instance in 2006. Since 2012 as part of the Nordic e- Infrastructure Collaboration. Physical resources at existing super computing centers - run by local staff. A virtual team of people to fulfill Tier 1 obligations and to maintain the illusion of a single site. 13

What s in a Name? NorduGrid Loose collaboration of Nordic HEP community. Founded in 2001 as Scandinavian Grid. Formed around the Advanced Resource Connector (ARC) software, developed by the NorduGrid community. Now over 80 sites in 19 countries. Nordic Data Grid Facility (NDGF) The Nordic Tier 1 site founded in 2006, layered on top of a slice of NorduGrid. Nordic e-infrastructure Collaboration (NeIC) Nordic collaboration founded in 2012 to facilitate development of IT tools and services in areas of importance to Nordic researchers. Nordic Tier 1 (NT1) An area of NeIC maintaining and operating NDGF. 14

CE CE CE CE Finishing Batch farm Other Tier 2s SE Preparing ARC Data taking Job Cache & session data Other Tier 1s AcT LHC PANDA 15

The queue after preparing and before execution isolates the job from the SE. Allows downtime on the SE without draining batch farms or having idle compute nodes. SE can be designed for density and cache/session disk for IOPS. CE CE CE CE Finishing Batch farm SE ARC Other Tier 2s Preparing Data taking Job Cache & session data Other Tier 1s AcT LHC PANDA 15

NDGF Storage System stats 94 million files on disk 6 million files on tape 3.5 PB on tape 7.5 PB on disk 16

Agenda WLCG: A world wide computing grid for the LHC NDGF: The Nordic Tier 1 dcache: Distributed data storage for e-science 17

dcache is... software for providing scalable, managed storage for huge amounts of data. deployed at research institutes throughout the world and used by a diverse collection of user communities. supported through the dcache.org collaboration, which provides: regular feature releases that are maintained with subsequent bugfix releases. Support and advice through a variety of channels. dcache is open source (AGPL 3). 18

Graphics dcache.org 19

Graphics dcache.org 20

Graphics dcache.org 21

Physical content (object store) Logical organization (database) End user access mechanism (protocol and endpoint) 22

Pool Physical content (object store) Logical organization (database) End user access mechanism (protocol and endpoint) 22

Pool Name space Physical content (object store) Logical organization (database) End user access mechanism (protocol and endpoint) 22

Pool Name space Physical content (object store) Logical organization (database) End user access mechanism (protocol and endpoint) FTP HTTP NFS 22

Pool Name space Physical content (object store) Request scheduling Monitoring Authorization Administration Message passing Logical organization (database) End user access mechanism (protocol and endpoint) FTP HTTP NFS 22

Graphics dcache.org 23

24

Graphics dcache.org 25