Infrastructure for testing accelerators and new

Slides:



Advertisements
Similar presentations
The Development of Mellanox - NVIDIA GPUDirect over InfiniBand A New Model for GPU to GPU Communications Gilad Shainer.
Advertisements

ATLAS Tier-3 in Geneva Szymon Gadomski, Uni GE at CSCS, November 2009 S. Gadomski, ”ATLAS T3 in Geneva", CSCS meeting, Nov 091 the Geneva ATLAS Tier-3.
PRAKTICKÝ ÚVOD DO SUPERPOČÍTAČE ANSELM Infrastruktura, přístup a podpora uživatelů David Hrbáč
LinkSCEEM-2: A computational resource for the development of Computational Sciences in the Eastern Mediterranean Mostafa Zoubi SESAME SESAME – LinkSCEEM.
S. Gadomski, "ATLAS computing in Geneva", journee de reflexion, 14 Sept ATLAS computing in Geneva Szymon Gadomski description of the hardware the.
SUMS Storage Requirement 250 TB fixed disk cache 130 TB annual increment for permanently on- line data 100 TB work area (not controlled by SUMS) 2 PB near-line.
Academic and Research Technology (A&RT)
IFIN-HH LHCB GRID Activities Eduard Pauna Radu Stoica.
High Performance Computing (HPC) at Center for Information Communication and Technology in UTM.
Real Parallel Computers. Modular data centers Background Information Recent trends in the marketplace of high performance computing Strohmaier, Dongarra,
HPCC Mid-Morning Break Dirk Colbry, Ph.D. Research Specialist Institute for Cyber Enabled Discovery Introduction to the new GPU (GFX) cluster.
HPC at IISER Pune Neet Deo System Administrator
GPU Programming with CUDA – Accelerated Architectures Mike Griffiths
Parallelization with the Matlab® Distributed Computing Server CBI cluster December 3, Matlab Parallelization with the Matlab Distributed.
Bob Thome, Senior Director of Product Management, Oracle SIMPLIFYING YOUR HIGH AVAILABILITY DATABASE.
IPlant Collaborative Tools and Services Workshop iPlant Collaborative Tools and Services Workshop Collaborating with iPlant.
Computing Labs CL5 / CL6 Multi-/Many-Core Programming with Intel Xeon Phi Coprocessors Rogério Iope São Paulo State University (UNESP)
S&T IT Research Support 11 March, 2011 ITCC. Fast Facts Team of 4 positions 3 positions filled Focus on technical support of researchers Not “IT” for.
Looking Ahead: A New PSU Research Cloud Architecture Chuck Gilbert - Systems Architect and Systems Team Lead Research CI Coordinating Committee Meeting.
14 Aug 08DOE Review John Huth ATLAS Computing at Harvard John Huth.
The LHCb Italian Tier-2 Domenico Galli, Bologna INFN CSN1 Roma,
The DCS lab. Computer infrastructure Peter Chochula.
1 Lattice QCD Clusters Amitoj Singh Fermi National Accelerator Laboratory.
WLCG Overview Board, September 3 rd 2010 P. Mato, P.Buncic Use of multi-core and virtualization technologies.
CD FY09 Tactical Plan Status FY09 Tactical Plan Status Report for Neutrino Program (MINOS, MINERvA, General) Margaret Votava April 21, 2009 Tactical plan.
Università di Perugia Enabling Grids for E-sciencE Status of and requirements for Computational Chemistry NA4 – SA1 Meeting – 6 th April.
Workshop on Advanced Computing for Accelerators Day 3 Roger Barlow.
PDSF and the Alvarez Clusters Presented by Shane Canon, NERSC/PDSF
Computing Issues for the ATLAS SWT2. What is SWT2? SWT2 is the U.S. ATLAS Southwestern Tier 2 Consortium UTA is lead institution, along with University.
HEPMARK2 Consiglio di Sezione 9 Luglio 2012 Michele Michelotto - Padova.
CNAF Database Service Barbara Martelli CNAF-INFN Elisabetta Vilucchi CNAF-INFN Simone Dalla Fina INFN-Padua.
CERN IT Department CH-1211 Genève 23 Switzerland t SL(C) 5 Migration at CERN CHEP 2009, Prague Ulrich SCHWICKERATH Ricardo SILVA CERN, IT-FIO-FS.
FESR Consorzio COMETA Giuseppe Andronico INFN Sez. CT & Consorzio COMETA Workshop Grids vs. Clouds Beijing, Consorzio.
PADME Kick-Off Meeting – LNF, April 20-21, DAQ Data Rate - Preliminary estimate Tentative setup: all channels read with Fast ADC 1024 samples, 12.
Daniele Cesini - INFN CNAF. INFN-CNAF 20 maggio 2014 CNAF 2 CNAF hosts the Italian Tier1 computing centre for the LHC experiments ATLAS, CMS, ALICE and.
Farming Andrea Chierici CNAF Review Current situation.
Introduction to Data Analysis with R on HPC Texas Advanced Computing Center Feb
APE group Many-core platforms and HEP experiments computing XVII SuperB Workshop and Kick-off Meeting Elba, May 29-June 1,
CNAF - 24 September 2004 EGEE SA-1 SPACI Activity Italo Epicoco.
Sobolev(+Node 6, 7) Showcase +K20m GPU Accelerator.
+ The INFN COSA project Michele Michelotto
Brief introduction about “Grid at LNS”
Disruptive Storage Workshop Lustre Hardware Primer
NIIF HPC services for research and education
NFV Compute Acceleration APIs and Evaluation
HPC Roadshow Overview of HPC systems and software available within the LinkSCEEM project.
DCS Status and Amanda News
HPC usage and software packages
Data Analytics and CERN IT Hadoop Service
Grid site as a tool for data processing and data analysis
Daniele Cesini – INFN-CNAF - 19/09/2017
SuperB and its computing requirements
LinkSCEEM-2: A computational resource for the development of Computational Sciences in the Eastern Mediterranean Mostafa Zoubi SESAME Outreach SESAME,
Moroccan Grid Infrastructure MaGrid
Heterogeneous Computation Team HybriLIT
Yaodong CHENG Computing Center, IHEP, CAS 2016 Fall HEPiX Workshop
Luca dell’Agnello INFN-CNAF
The INFN TIER1 Regional Centre
Stallo: First impressions
Статус ГРИД-кластера ИЯФ СО РАН.
Preparation for the Di-Jet Tsukuba
Stato progetto COSA
Proposal for the LHCb Italian Tier-2
This work is supported by projects Research infrastructure CERN (CERN-CZ, LM ) and OP RDE CERN Computing (CZ /0.0/0.0/1 6013/ ) from.
Scientific Computing At Jefferson Lab
Introduce yourself Presented by
Footer.
Luca dell’Agnello, Daniele Cesini – INFN-CNAF CCR - 23/05/2017
This work is supported by projects Research infrastructure CERN (CERN-CZ, LM ) and OP RDE CERN Computing (CZ /0.0/0.0/1 6013/ ) from.
Workflow and HPC erhtjhtyhy Doug Benjamin Argonne National Lab.
Presentation transcript:

Infrastructure for testing accelerators and new technologies @CNAF Daniele Cesini, Gaetano Maron INFN-CNAF

Accelerators at CNAF Small cluster Infiniband interconnected independent from T1 Funded by INFN-CNAF, INFN-BO, UniBo, former COKA project 27 Worker Nodes CPU: 872 HT cores 608 HT cores E5-2640v2 48 HT cores X5650 48 HT cores E5-2620v2 168 HT core E5-3683v3 15 GPUs: 8 Tesla K40 7 Tesla K20 2x(4GRID K1) 3 MICs: 2 x Xeon Phi 5100 1 x Xeon Phi 3100 Dedicated STORAGE 2 disks server 60 TB shared disk space 4 TB shared home Disk server 60+4 TB 1Gb/s 10Gb/s 2x10Gb/s Worker Nodes IB ETH CPU GPU MIC TOT TFLOPS (DP - PEAK) 6 19 3 28 CCR Workshop – La Biodola - 19/05/2016

Clusters Users Simulation of acceleration and plasma physics 90% of the usage Bologna and CERN groups Preparatory runs before launching in supercomputers Astrophysics simulations Tests by COSMO_WNEXT (PLANK in particular) Collaboration between CNR, UniBo, INFN Radio-protection simulations at particle accelerators INFN-UniBO collaboration Currently 90% of the production usage is on the CPUs, not accelerators Testing and performance evaluation of hw and sw Daniele Cesini – INFN-CNAF CCR Workshop – La Biodola - 19/05/2016

Testing hw and sw Recent testing activities GPU and Phi performance Application porting to GPU Intel vs gcc compilers evaluation GPU in the cloud GPU and DOCKER LHCb event building sw Cannot be done in production Interference with other jobs The batch system add complexity A constant number of nodes is out of the cluster, reducing the total power Frequent administrator interventions to install needed libraries or operating system and to restore the nodes back to production Daniele Cesini – INFN-CNAF CCR Workshop – La Biodola - 19/05/2016

Low power cluster The COSA project clusters (see next talk…) 4xINTEL N3700 16xARMv7 2xARMv8 4xINTEL AVOTON C-2750 4xINTEL XEOND-1540 Daniele Cesini – INFN-CNAF CCR Workshop – La Biodola - 19/05/2016

Other testbeds Testbed for event building sw Few nodes low latency interconnected Intel and Mellanox fabrics Traditional and low power architectures Event building for LHCb tested Testbed for the new Intel KNL Intel XEON-PHI with Omnipath (We hope) Daniele Cesini – INFN-CNAF CCR Workshop – La Biodola - 19/05/2016

An infrastructure for testing new technologies @CNAF Reconfiguration of the current testbeds to create a single infrastructure dedicated only to testing purposes Accelerators (GPU, KNC, KNL) – How many? Low Power CPUs Low latency interconnection for event builder sw how many nodes? New CPU and storage systems Services Access to hw and sw to interested INFN users Tools to book in advance and access the systems System administration support Daniele Cesini – INFN-CNAF CCR Workshop – La Biodola - 19/05/2016

Research Agreement with the CERN Open Lab We are signing a Research Agreement with the CERN Open Lab Tests at CERN OpenLab will be possible Participation to Open Lab projects Intel "Code Modernization" project as soon as the RA will be signed http://openlab.web.cern.ch/technical-area/computing- platforms-offline Access to Intel sw and hw products located at CERN 30k euro per year for training (@CERN) Daniele Cesini – INFN-CNAF CCR Workshop – La Biodola - 19/05/2016

Collaboration with the ISSS infrastructure The “Nuove tecnologie” infrastructure would be the hardware counterpart of the ISSS project Daniele Cesini – INFN-CNAF CCR Workshop – La Biodola - 19/05/2016

Funding Manpower from CNAF HW & SW i.e. fraction of the user support team open to any possible collaboration HW & SW CSN V Any other CSN interested (i.e requests from CSN2/VIRGO) Direct funding from interested experiments Letter of Intent to the GE from interested experiments on specific topics CCR on specific topics Already contributing to sw license (i.e. Intel compiler) Daniele Cesini – INFN-CNAF CCR Workshop – La Biodola - 19/05/2016

Summary At CNAF about 20 accelerators are available in a cluster, low latency interconnected Interested users can request access for testing purposes and/or production We are restructuring all our testbeds in an infrastructure open to projects and collaborations with other departments and experiments We are signing a Research Agreement with CERN OpenLab Participation to OpenLab project Intel Code Modernization Daniele Cesini – INFN-CNAF CCR Workshop – La Biodola - 19/05/2016