ALICE BP Meeting G.G. Barnaföldi et al, Budapest, 10th August 2015

Transcription

1 2 nd ALICE BP Meeting 2015 G.G. Barnaföldi et al, Budapest, 10th August 2015

2 Directions for 2015 Financing OTKA NK will run out by next year (CERN DAQ, analysis, GRID) Less resources: minimal travel, salaries (1+0,5+0,5), 0 for equipments Varga D. Lendület (Detector R&D) Other resources are weak, but... please apply for e.g. Wigner KIS INFRA (6 MHUF) Clean Room development (1-2 MHUF?) GRID (10 MHUF) ALICE Old directions: HMPID (analysis), DAQ, TPC (R&D), New directions: PID analysis, TPC (analysis, R&D), DAQ R&D GROUPS DAQ TPC R&D HMPID GRID T2 HF (Weak point is still the manpower...) [E Dávid, J. Imrek, G. Kiss, T. Kiss, Tuan M. Nguyen] [D Varga, G Hamar, Gy Bencédi, G. Kiss, Gy. Bencze, L. Boldizsár] [L. Oláh, S. Pochybová, E. Futó] [SzM. Harangozó, G. Bíró, E Futó, P. Pósfay, A. Lowe, GGB] [R Vértesi] 08/10/15 G.G. Barnaföldi, Wigner RCP, HAS 2

3 The manpower: PhD (9) S. Pochybová (Quark/Gluon analysis, HMPID analysis, HMPID shifts) Gy. Bencédi, (TPC/HMPID PID analysis, UNAM Mexico) L. Oláh (HMPID ageinganalysis) G. Hamar (ALICE/REGARD, TPC) postphd G. Kiss (REGARD, DAQ) Sz.M. Harangozó (GRID Management) L. Boldizsár (TPC, HMPID shifts) G. Bíró (GRID management) P. Pósfay (GRID management) MSc/BSc (2) M. Tuan Nguyen (DAQ) Zs. Szendi (TPC)? Sz. Karsai PostDoc/Staff (2+4+4) A. Lowe (HMPID & Analysis) D. Varga (REGARD, TPC) R. Vértesi (HF Jets) G. Volpe (HMPID, PID until ) E. Futó (Shifts) T. Tölyhi J. Imrek E. Dávid 08/10/15 G.G. Barnaföldi, Wigner RCP, HAS 3

4 Directions for 2015 Financing OTKA NK will run out by next year (CERN DAQ, analysis, GRID) Less resources: 2 M for travel, salaries, 0,5M for equipments Varga D. Lendület (Detector R&D) Other resources are weak, but... please apply for e.g. Wigner KIS INFRA (6MHUF) Clean Room development (1-2 MHUF) GRID ALICE Old directions: HMPID (analysis), DAQ, TPC (R&D), New directions: PID analysis, TPC (analysis, R&D), DAQ R&D, HF GROUPS DAQ TPC R&D HMPID GRID T2 HF 08/10/15 (Weak point is the manpower... still) [E Dávid, J. Imrek, G. Kiss, T. Kiss, Tuan M. Nguen] [D Varga, G Hamar, Gy Bencédi, G. Kiss, Gy. Bencze, L. Boldizsár] [L. Oláh, S. Pochybová, E. Futó] [SzM. Harangozó, G. Bíró, E Futó, P. Pósfay, A. Lowe, GGB] [R Vértesi] G.G. Barnaföldi, Wigner RCP, HAS 4

5 Planned tasks and PAPERs PhD & PostDoc (see more in next presentations) S. Pochybová: ALICE PID Paper, AN (NOW), PhD Gy. Bencédi: correlation paper, HpT4LHC, Fiz Szemle G. Hamar: (see presentation) L. Oláh: Muontomograph, AN/HMPID Performance Paper G. Kiss: DAQ papers A. Lowe: AN? Sz. Karsai: ALICE Bochure and Cartoon (CERN/ alice.wigner.mta.hu) OTHER G. Biró / Nagy Máté Ferenc (MC@GPU) G. Kalmár FF paper D. Varga & G. Hamar (TPC GEM test method) 08/10/15 G.G. Barnaföldi, Wigner RCP, HAS 5

6 Planned Trips Conferences Quark Matter 2015 (BGy, PL, PG, BGG) other resources High-pT For the LHC, BNL, USA (???) OTKA? Need to pick up the line: Zagreb Budapest Meetup Zimányi Winter School 2015 (All) Data LHC (A. Lowe) GRIDKa School, Karlsruhe (B.G?) Planned trips Beam Test with Bergen group ( / ) UNAM-CERN E-PLANet (BGy, Mexico) Wuhan (HSzM, PG, PL, BGG) 08/10/15 G.G. Barnaföldi, Wigner RCP, HAS 6

7 Visibility Seminars New Students, New members: Wigner Seminar Eötvös Univeristy Close-to-PhD, Seminar at Eötvös University PR activities T. Kiss, E. Dávid, T. Tölyi: DAQ Labor has been established BGG: Atomcsill (2. féléb BGG et al: November Tudomány napja ALICE TPC PR.. 08/10/15 G.G. Barnaföldi, Wigner RCP, HAS 7

8 Summaries by the Groups... 08/10/15 G.G. Barnaföldi, Wigner RCP, HAS 8

9 DAQ Group 08/10/15 G.G. Barnaföldi, Wigner RCP, HAS 9

10 Recent events and activities (2015) Group expansion: József Imrek - a new colleague joined our group as a designing engineer in April Tuan Máté Nguyen - he has been accepted as a technical student in CERN ALICE DAQ Group from September Recent purchases/investments: LHCb MiniDAQ StratixV FPGA dev kit for ALICE CRU firmware development including 10 GbE NIC and optical break-out cables New 3-channel power supply unit Schroff MicroTCA crate for operating Advanced Mezzanine Cards (AMC) 2 new desktop PC for FPGA firmware development Preparing for the renovation of the laboratory: Scrapping, packing

11 Recent events and activities (2015) Group activity related to ALICE LS1: ALICE TPC Readout Control Unit (RCU2) upgrade - FPGA firmware development SGMII Ethernet Link firmware modules development (E. Dávid) Group activity related to ALICE LS2: ALICE Common Readout Unit (CRU) Development Project management, coordination (T. Kiss) Firmware development coordination (E. Dávid) Parts of firmware development (E. David) ALICE FLP nodes I/O and memory performance tests (J. Imrek) Regular CRU weekly meetings (every 10:30) (T. Kiss, E. Dávid) Organizing of 1st CRU Workshop at CERN (1-2 July, T. Kiss, E. Dávid, J. Imrek, T. M. Nguyen)

12 Further plans (2015) Ongoing developments: ALICE CRU project management (T. Kiss), firmware development coordination (E. Dávid), and firmware development (E. David) ALICE FLP nodes I/O and memory performance tests (J. Imrek) T. M. Nguyen will join to the CRU development in CERN from September developing linux drivers for the new CRU cards Travel plans (from Wigner / OTKA budget): September/October, CRU Meetings (T. Kiss, E. Dávid) November/December, ALICE week/mini-week, 2nd CRU Workshop (T. Kiss, E. Dávid, J. Imrek) 5-6 filghts to Geneva in total. Flights to be booked in August, to reduce costs! Planned purchases (from CRU development budget at CERN): Two Altera Arria10 FPGA Engineering Sample v2 for the PCIe40 prototypes 1 or 2 Altera Arria10 FPGA Development Board for CRU prototype development CERN Versatile Link Demo Board (VLDB) for GBT link development

13 ALICE Tier 2 & Local CAF 08/10/15 G.G. Barnaföldi, Wigner RCP, HAS 13

14 Szilveszter Harangozó Research progress HIJING v3.0 ktpqcd v2.1 Core software dev. Calculation for LHC energies Pythia8 XML libraries GSL integration Function validation Model development Shadowing upgrade Jet quenching QCD cascade model pp, pa charged π,p,k spectra at 7 and 5.02 TeV (Gyu) PHENIX and ALICE energy scan for 22.4, 62.4, 200, 2760 GeV Program upgrades HIJING

15 ALICE Tier2 and CAF Tier2 massive storage upgrade 136TB 316TB CPU stack upgrade 6.5k 13k HS06 decommission of grid130 CAF alice09 machine alice06, 08 rebooted MOVING TO SEPARATE RACK IN AUGUST Decommissioned machince may be placed to CAF

16 Bíró Gábor Achievements Language exam, MSc Talks: Zimányi 2014, Statistical Physics Day 2015, CiiT 2015 (Bitola, Macedonia), GPU Day 2015 Papers: 2015 J. Phys.: Conf. Ser , CiiT Proceedings, others in progress Activities Analysing identified hadron spectra with Tsallis Pareto distributions in the soft+hard approach ALICE Grid Tier 2 Plans Start a successful PhD :) Travel: September 2015 Karlsruhe, GridKa School (80kHUF) 08/10/15 G.G. Barnaföldi, Wigner RCP, HAS 16

17 REGARD & TPC R&D 08/10/15 G.G. Barnaföldi, Wigner RCP, HAS 17

18 G. Hamar : Short Summary on Works and their Connection to the ALICE-Budapest Group ALICE Budapest Meeging, 10. Aug Budapest PhD: done :) [formal paper expected in winter] Postdoc application: INFN Trieste (Italy) : accepted, start: 18. Aug Topic: Micropattern detectors, COMPASS Do not want to stop the activities in the REGARD, and ALICE BP groups. 08/10/15 G.G. Barnaföldi, Wigner RCP, HAS 18

19 Recent work topics: Leopard project (AIDA 2020), GEM QA, MuonTomography, MWPC, PET,... [GEM QA is for the ALICE GEM TPC Upgrade] Conferencies: Muographers2015; RD51CM; Erice ISSP Coming: MPGD2015 (October) Papers: Two Proceedings of TIPP2014 on CCC, and Leopard In preparation: TCPD paper, Trieste Leopard paper [CCC and TCPD were developed for the VHMPID] 08/10/15 G.G. Barnaföldi, Wigner RCP, HAS 19

20 GEM gain mapping for QA purposes in framework of ALICE TPC UG Dezső Varga for the Budapest group GEM gain maps, 4mm by 4mm pixel size (4 min. time for 10cm by 10cm) Consistency checks (rotating, flipping) Maps at various gains: relation / predictivity Maps in different gases: relation / predictivity Hot spots? Scheme as a QA device 20

21 Prelude: demonstration of hole geometry and GEM gain relation Recent paper by the Helsinki group, demonstrating gain predictability by hole geometry! T. Hildén, E. Brücken et al, Optical quality assurance of GEM foils 21

22 GEM gain mapping As a Quality Assurance device: gain map for detector performance assessment rejecting foils with excessive non-uniformity cross-check the prediction from optical scanning As an R&D issue: how to obtain the best prediction of gain map from hole geometries how to relate different GEM voltages, working gases, transfer/drift fields etc. 22

23 GEM from the Helsinki group we are very grateful for the collaboration GEM contacts 23

24 Chamber outline: GEM + high gain MWPC Guard rim for well defined geometry and field 24

25 Detector assembly MWPC bare with GEM fixing screws 25

26 Installing GEM GEM contacts 26

27 Installing cathode 27

28 The complete setup Source 40cm above GEM contacts (Fe55) DAQ board ADC (12 bit) R-Pi computer (DAQ developed within RD51 CFP) 28

29 Check GEM / detector / readout orientation GEM contacts 29

30 MWPC uniformity (GEM off) Well equalized gain of the CCC good reference! GEM contacts 30

31 GEM gain map at gain of 6 UGEM=300V USW=1000V GEM contacts Re-done after 15 min, ratio 31

32 GEM gain map at gain of 30 High gain similar map as lower gain, nonuniformity increased Re-done after 15 min, UGEM=400V USW=800V ratio 32

33 Relation between the different voltages: power law (linear) Non-uniformity goes with GEM gain slope! Predictable at any voltage Gain ratio, 400V vs 300V Gain at 400V Gain at 300V (Done in flipped configuration) 33

34 Rotation exercise (shown in detector coordinate system) 300V GEM voltage 400V GEM voltage (nicely rotates) 34

35 Comparing with non-rotated Apparent compatibility, needs to be quantified ORIGINAL ROTATED 35

36 Flipping (upside down) exercise (shown in detector coordinate system) 300V GEM voltage 400V GEM voltage (nicely fips left/right) 36

37 Comparing with non-flipped Apparent compatibility, needs to be quantified ORIGINAL FLIPPED 37

38 Going to TPC gas: Ne:CO2:N2 90:10:5 300V GEM voltage 400V GEM voltage (similar as in ArCO2) 38

39 GEM voltage dependence again established Again correlated, different slope Predictable at any voltage Gain map ratio 400V vs 300V Gain at 400V Gain at 300V 39

40 Difference between different gases: power law / linear Well established correlation at 300V and 400V 300V Gain map ratio Correlation Gain in NeCO2N2 Correlation Gain in NeCO2N2 Gain map ratio 400V Gain in ArCO2 Gain in ArCO2 Predictable between ArCO2 and NeCO2N2! 40

41 GEM scanning observables TG26 maps from Timo and Erik (thanks!) Inner Outer BOTTOM Outer TOP Misalignment 41

42 Comparing to present version of gain prediction (Helsinki results) PREDICTION MEASURED Prediction prescription is now under work and being improved by Helsinki colleagues! 42

43 As a device: key parameters with the existing setup Now 4kHz data taking: 4min. for 1M events, that is: 1 hour for IROC (50cm) size Point-by-point statistical error of 0.3% Above 90% good event efficiency 43

44 Next steps towards real size: OROC and IROC foils for the ALICE TPC Step 1: 50cm version, should match an IROC Step 2: OROC version, matched with transportable GEM-s of full size 44

45 50cm version: based on existing design, existing components Will it work: YES, part of an existing cosmic detector showing 99% efficiency and 4mm position sensitivity Key details, such as GEM installation, gas box integrity, top cover is to be clarified 45

46 Can we find hot spots? Strategy: High GEM voltage/gain (e.g. 420V, 50) Go with SW from 750V to 950V (!) No source (only cosmics) Sensitivity to equivalent of a few electrons (to be checked). Local corona discharges should appear as low signal counts 46

47 No source, high signal (cosmic) map Uniform cosmic, local structures UGEM=400V USW =950V POSITION MAP Signal region:

48 No source, low signal region Local hot spots clearly appear UGEM=400V USW =950V POSITION MAP Signal region: e 48

49 Higher GEM voltage Hot spots persist (note 5 times data statistics!) UGEM=420V USW =950V POSITION MAP 30e Signal region: RD51 14 ecollaboration Meeting, 20/03/

50 Hot spot appearance in rotated configuration Some (?) of the hot spots follow as expected ORIGINAL ROTATED 50

51 Observations Hot spots (few electron signals creating local noise at few Hz) clearly observable Does it have implication on QA: can such a hot spot trigger sparking? can one characterize GEM-s by appearance voltage of hot spots? may identified hot spots be correlated by optical? 51

52 Conclusions Detector proposed for fast GEM gain mapping Rotation, flipping exercise: detector performance OK Voltage dependence: seems predictable with power law / linear Gas filling ArCO2 vs NeCO2N2 dependence: seems predictable with power law / linear Full size versions for ALICE IROC/OROC foils are being designed as a QA device 52

53 ANALYSIS GROUP 08/10/15 G.G. Barnaföldi, Wigner RCP, HAS 53

54 A two-slide summary of activities and plans Róbert Vértesi ALICE-Budapest meeting /10/15 G.G. Barnaföldi, Wigner RCP, HAS 54

55 RHIC PHENIX Debrecen University / Wigner RCP, Budapest Calorimetry, neutral pions and direct photons Femtoscopy: pion simulations In-medium meson modification via femtoscopy STAR - Nuclear Physics Institute, Prague/Řež 1. Bulk and Correlations PWG Freeze-out dynamics via kaon femtoscopy in 200 GeV Au+Au SQM2013 (RV for STAR), Nucl. Phys. A 910, 331 (2013) PAs: P. Chaloupka, P. Chung, R. Lednický, M. Šumbera, R.V. Published in Phys. Rev. C 88, (2013) 2. Work in the Heavy Flavor PWG Heavy Flavor production in STAR, ICHEP 2015 (Accepted in NUPHBP) Upsilons production in 193 GeV U+U collisions Paper in progress, PA: R.V. Presented on PANIC2014, Desy Conf. Proc.; and HP /10/15 G.G. Barnaföldi, Wigner RCP, HAS 55

56 LHC/ALICE Get to know the system (just got the CERN account) Get involved in PWG-HF activities Workgroup? HFE : Use knowledge for Calorimetry and open HF; not very novel but need manpower for quick results HFCJ : Expertise of the Bp (and NPI) group on jets; way less explored field Plan to go to CERN for the ALICE RHIC/STAR Hard Probes proceedings, deadline 18th September Finish the paper (current status: accepted by PWG, GPC being formed) Wish for the future: show final result on a conference 08/10/15 G.G. Barnaföldi, Wigner RCP, HAS 56

57 Sona Pochybova Activities: Particle ratios in jet and bulk Contribution with a cross-check to statistical identification with TOF Paper in discussion HPT 2014: Proceedings:arXiv: [hep-ex] Plans Finish PhD Possible travel: 08/10/15 January 2016 CERN, on-call HMPID G.G. Barnaföldi, Wigner RCP, HAS 57

58 Gyula Bencédi Paper (expected to be on arxiv before QM 15') Multiplicity dependence of pion, kaon and proton production at large transverse momentum in p-pb collisions at sqrt(s_{nn}) = 5.02 TeV Upcoming conference QM 15' Contribution:??? maybe poster (on 13TeV) not 100% (most of the submitted abstracts in PWGLF were downgraded to poster presentation) Thesis Conclude in /10/15 G.G. Barnaföldi, Wigner RCP, HAS 58

59 A. Lowe: ALICE activities: Machine learning for particle physics Devise prototype quark/gluon jet tagger using machine learning Data-driven variable selection for model building First analysis of this type performed in R (instead of ROOT) Automated methods for searching feature space for new discriminants Found new discriminant variables for quark/gluon tagging Proof of concept demonstration that R can be used for fast prototyping of HEP analysis Collaboration with SZTAKI Data Mining and Search Group, HAS András Benczúr (head of department) and students Joint research project to co-investigate machine learning for q/g tagging Permission granted by ALICE Management to share MC data More information can be found on these slides 08/10/15 G.G. Barnaföldi, Wigner RCP, HAS 59

60 ALICE activities: paper plans Data-driven feature selection for quark/gluon jet tagging at ALICE Computer-vision inspired discriminants for quark/gluon jet tagging Could be a single paper, but this might be a topic best not removed from the context of building a jet tagger (danger: selection bias!) Preliminary results from feature selection suggest that a family of new discriminants (Hu image moments) could be at least as good or better than those already proposed for quark/gluon jet discrimination Prototype quark/gluon jet tagger for ALICE Still working on improving performance Primary bottleneck: lack of computing muscle Need more CPU cores & RAM cloud computing ($) or local resources? Joint paper with András Benczúr's group? Need to discuss with ALICE how non-members can be attributed credit 08/10/15 G.G. Barnaföldi, Wigner RCP, HAS 60

61 Other activities Outreach "Machine learning for particle physics using R", Budapest R Users Group Meetup, BME, 18 February 2015 "Big Fast Data in High-Energy Particle Physics", Budapest Data Forum, Hotel Mercure Buda, 3 June 2015 Same day as start of LHC Run 2! Talk evaluation: 4.22 out of 5 (top speaker score was 4.75) Co-organisation of Data LHC2015 workshop at CERN, 9-13 November Workshop to help foster long-term connections between the data science and particle physics communities Will be attending Registration now open! 08/10/15 G.G. Barnaföldi, Wigner RCP, HAS 61

62 10 August 2015 Andrew Lowe, ALICE group meeting 62