1. FZU Computing Centre Jan Švec Institute of Physics of the AS CR, v.v.i. 29.8.2011

2. FZU Computing Centre Computing center is an independent part of the Department of Networking and Computing Techniques. Members of the team lead by J. Chudoba from the Department of Experimental Particle Physics: T. Kouba, J. Švec, J. Uhlířová, M. Eliáš, J. Kundrát. Operation is strongly supported by some members of the Department of detector development and data processing lead by M. Lokajíček: J. Horký, L. Fiala.

3. FZU Computing Centre HEP experiments  D0 experiment, Fermilab (USA)  LHC experiments ATLAS, ALICE (CERN)  STAR experiment, BNL (USA) Solid state physics Astroparticle physics  Pierre Auger Observatory (Argentina)

4. FZU Computing Centre

5. History: 2002 34 HP LP1000  2x 1.3 GHz Pentium 3  1 GB RAM First 1 TB of disks “Terminal room”

6. History: 2004 A real data center 200 kVA UPS  380 kVA diesel 2x 56 kW CRACs 67 HP DL140  3.06 GHz Prescotts 10 TB disk storage

7. History: 2006 - 2007 2006  36 HP BL35p  6 HP BL20p 2007  18 HP BL460c  12 HP BL465c

8. History: 2006 – 2007 (cont.) 3 HP DL360 for Xen  FC infrastructure HP EVA  70+ TB usable capacity, FATA drives  Disk images for Xen machines  Rest used as DPM  Warranty ended in Dec 2010 A new box cheaper than warranty extension

9. History: 2008 84x IBM iDataPlex node dx340  2x Xeon E5440 => 8 cores 20x Altix XE 310 twins (40 hosts)  2x Xeon E5420 => 8 cores 3x Overland Ultamus 4800 (48TB raw each) SAN Tape library NEO 8000 2x VIA based NFS storage First decommissioning of computing nodes  2002's HP LP1000r

10. History: 2009 65x IBM iDataPlex dx360M2 nodes  2x Xeon E5520 with HT => 16 Cores 9x Altix XE 340 twins (18 hosts)  2x Xeon E5520 without HT => 8 cores All water cooled 3x Nexsan SataBeast (84TB raw each) SAN 3x Atom-based storage nodes  Wrong idea, WD15EADS-00P8B0 are just weird

11. History: 2009 SGI Altix ICE 8200  Solid state physics  512 cores (128 x E5420 2.5G)  1TB RAM  Infiniband  6TB disk array (Infiniband)  Torque/Maui, OpenMPI

12. Cooling 2009 2009: New water cooling infrastructure  STULZ CLO 781A 2x 88 kW Additional ceiling for hot-cold air separation

13. History: 2010 26x IBM iDataPlex dx360M3 nodes  2x Xeon X5650 with HT => 24 Cores Bull Optima 1500 – HP EVA replacement 8x Supermicro SC847E16 + 847E16-RJBOD  DPM pool nodes  1,3PB Second decommissioning of computing nodes  2004's HP DL140  8 IBM dx340 swapped for dx360M3

14. Network facilities External connectivity delivered by CESNET, a GEANT stakeholder  10Gb to public Internet  1Gb to FZK (Karlsruhe)  10Gb connection demuxed to several dedicated 1Gb lines FNAL, BNL, ASGC, Czech Tier-3s,... 10G internal network  Few machines still on 1G switches → LACP

15. 2011: Current Procurements Estimates: – Worker nodes, approx. 5500 HEPSPEC06 – 800TB disk storage – Water cooled rack + water cooled back doors – Another NFS fileserver

16. Overall performance growth

17. Current numbers 275 WNs HEP, 76 WNs solid state physics 3000 cores HEP, 560 cores solid state physics Torque & Maui 2PB on disk servers (DPM or NFSv3)

18. Q & A Questions?