Frontiers of research at CERN and the computing and data management - - PowerPoint PPT Presentation

SLIDE 1

Frontiers of research at CERN and the computing and data management needs

E.Elsen

EGI Conference 2016. Science Park, Amsterdam, Apr 6-8, 2016

SLIDE 2

SLIDE 3

Experimental Tools of the Research Programme at CERN

• LHC
• ongoing run 2 @ 13 TeV
• HL-LHC (>2025)
• Fixed target programme
• ISOLDE
• AD-programme

SLIDE 4

…and

• Year 2014
• 3.4 bn CPU hours
• 100 PBytes

storage

SLIDE 5

European Strategy

• LHC and its upgrade
• Future energy frontier machines
• e+e- linear collider in Japan
• (not detailed here)
• ν–physics
• …

Accelerating science and innovation

Societal benefits of European research in particle physics

SLIDE 6

Energy Frontier

• The discovery of the Higgs boson is the start of a major programme of work to

measure this particle’s properties with the highest possible precision for testing the validity of the Standard Model and to search for further new physics at the energy frontier. The LHC is in a unique position to pursue this programme. 
 
 Europe’s top priority should be the exploitation of the full potential of the LHC, including the high-luminosity upgrade of the machine and detectors with a view to collecting ten times more data than in the initial design, by around

• 2030. This upgrade programme will also provide further exciting opportunities

for the study of flavour physics and the quark-gluon plasma.

SLIDE 7

14

Status of LHC machine and injectors SPC Frédérick Bordry 14th December 2015

Peak Integrated

ATLAS CMS

5 x 1033 cm-1 s-1 Design 1034 cm-1 s-1

2015 LHC Luminosity (p-p)

SLIDE 8

A L I C E

SLIDE 9

L H C b

SLIDE 10

L H C b

SLIDE 11

Measurement of Standard Model cross sections

• Standard Model

cross sections, incl. 13 TeV data

• relevant cross

sections vary over 8 orders of magnitude

C M S

SLIDE 12

Inclusive Jets @ 13 TeV

C M S

SLIDE 13

B-Mesons

• 8 TeV Data
• FONLL predictions
• Good agreement at

13 TeV

C M S

SLIDE 14

ATLASFCONFF2015F069'

Measurement of the Higgs Cross Section

• Combined significance from γγ and 4l
• expected 3.4σ
• bserved 1.4σ

A T L A S

SLIDE 15

Cross Sections as a function of √s

A T L A S

]

TeV

[ s 4 6 8 10 12 14

[pb]

σ

5

10

6

10

7

10

8

10

9

10

10

10

11

10

12

10 10

1

10

2

10

3

10

4

10

5

10

inelastic

Pythia8

W → pp

FEWZ

* γ / Z → pp

FEWZ

t t → pp

top++ NNLO+NNLL

tq → pp

NLO+NNLL

H → pp

LHC-XS

ZZ → pp

MCFM

inelastic

, Nat. Commun. 2, 463 (2011)

• 1

b µ 7 TeV, 20 , ATLAS-CONF-2015-038

• 1

b µ 13 TeV, 63

W → pp

, PRD 85, 072004 (2012)

• 1

7 TeV, 36 pb , ATLAS-CONF-2015-039

• 1

13 TeV, 85 pb

* γ / Z → pp

, PRD 85, 072004 (2012)

• 1

7 TeV, 36 pb , ATLAS-CONF-2015-039

• 1

13 TeV, 85 pb

t t → pp

, Eur. Phys. J. C 74:3109 (2014)

• 1

7 TeV, 4.6 fb , Eur. Phys. J. C 74:3109 (2014)

• 1

8 TeV, 20.3 fb , ATLAS-CONF-2015-049

• 1

13 TeV, 78 pb

tq → pp

, PRD 90, 112006 (2014)

• 1

7 TeV, 4.6 fb , ATLAS-CONF-2014-007

• 1

8 TeV, 20.3 fb , ATLAS-CONF-2015-079

• 1

13 TeV, 3.2 fb

H → pp

, arXiv:1507.04548

• 1

7 TeV, 4.5 fb , arXiv:1507.04548

• 1

8 TeV, 20.3 fb , ATLAS-CONF-2015-069

• 1

13 TeV, 3.2 fb

ZZ → pp

, JHEP 03, 128 (2013)

• 1

7 TeV, 4.6 fb , ATLAS-CONF-2013-020

• 1

8 TeV, 20.3 fb , STDM-2015-13

• 1

13 TeV, 3.2 fb

Preliminary ATLAS

Prediction Measurement

20'

SLIDE 16

Search for Diboson resonances

• Run 1: CMS ~2σ excess near 1.8-2.0 TeV

• Repeat search at 13 TeV using most sensitive channels: lνJ, JJ

• Analysis categorised in dijet mass for optimal sensitivity to WW, WZ,

ZZ signals

• 13 TeV: no excess observed in the region of interest near 2 TeV
• More data needed to fully exclude Run 1 excess
• s
• n
• GBulkWW (lνJ+JJ)

GBulkZZ (JJ) W’WZ (lνJ+JJ)

• son resonances
• s
• e

n

• f
• s

Run 1

• lνJ+JJ

G ZZ (JJ) W’ lνJ+JJ

C M S

SLIDE 17

Search for Dilepton resonances

Search%for%Z’%in%dilepton%(LFC)%and%(LFV)%(in!eµ!decays)!

F Main'background'DY'is'taken'from'MC' F Top'and'diboson'extrapolated'at'very'high'masses'using'a'funcDonal'form' F Background'from'MC'except'for'MJ'in'dielectron'uses'Matrix'method'(based'on'electron'ID)'

Highest'diFelectron' mass'event'at'1.8'TeV'' Highest'diFmuon'mass' event'at'1.4'TeV'' Highest'eµ'mass'event' at'2.1'TeV'' 95%'CL'Limit'on'SSM' LFV'Z’'at'3.0'TeV'(2.5'TeV'

from'RunF1)'

95%'CL'Limit'on'SSM'Z’'at'3.4'TeV'(2.9'TeV'from'RunF1)' No'Excess'found'!'

ee%

39'

µµ µµ! eµ!

ATLASFCONFF2015F070' ATLASFCONFF2015F072'

A T L A S

SLIDE 18

Search for Diphoton resonances

mγγ=745 GeV C M S

SLIDE 19

Search for Diphoton resonances

• Status Moriond 2016
• March 2016

C M S

• Improved calibration yields 10% greater sensitivity
• Fit background directly to the data using parameterization:

f(mγγ) = ma+b log mγγ

γγ

Events / ( 20 GeV )

1 10

2

10

Data Fit model σ 1 ± σ 2 ±

EBEE

(GeV)

γ γ

m

400 600 800 1000 1200 1400 1600

stat

σ (data-fit)/

• 2

2

(13 TeV, 3.8T)

• 1

2.7 fb

CMS Preliminary

Events / ( 20 GeV )

1 10

2

10

Data Fit model σ 1 ± σ 2 ±

EBEB

(GeV)

γ γ

m

400 600 800 1000 1200 1400 1600

stat

σ (data-fit)/

• 2

2

(13 TeV, 3.8T)

• 1

2.7 fb

CMS Preliminary

Diphoton spectrum at 3.8 Tesla (13 TeV)

Barrel-Barrel Barrel-Endcap

SLIDE 20

include data at B=0

• Without the magnetic field, we need new algorithms for vertex

selection and photon identification

• Correct vertex assignment is ~60% at 0T (~90% at 3.8T)
• Comparable photon efficiency: 85% (90%) at 0T (3.8T) per γ in barrel

Events / ( 20 GeV )

1 10

2

10

Data Fit model σ 1 ± σ 2 ±

EBEB

(GeV)

γ γ

m

400 600 800 1000 1200 1400 1600

stat

σ (data-fit)/

• 2

2

(13 TeV, 0T)

• 1

0.6 fb

CMS Preliminary

Events / ( 20 GeV )

1 10

Data Fit model σ 1 ± σ 2 ±

EBEE

(GeV)

γ γ

m

400 600 800 1000 1200 1400 1600

stat

σ (data-fit)/

• 2

2

(13 TeV, 0T)

• 1

0.6 fb

CMS Preliminary

Diphoton spectrum at 0 Tesla (13 TeV)

Barrel-Barrel Barrel-Endcap

C M S

SLIDE 21

C M S

New Physics with light SM particles at CMS – JPC – Rutgers University – Sunday, March 20th, 2016

DIPHOTON RESONANCES

8

[EXO-16-018]

• Without the magnetic field, we need new algorithms for vertex

selection and photon identification

• Correct vertex assignment is ~60% at 0T (~90% at 3.8T)
• Comparable photon efficiency: 85% (90%) at 0T (3.8T) per γ in barrel

Events / ( 20 GeV )

1 10

2

10

Data Fit model σ 1 ± σ 2 ±

EBEB

(GeV)

γ γ

m

400 600 800 1000 1200 1400 1600

stat

σ (data-fit)/

• 2

2

(13 TeV, 0T)

• 1

0.6 fb

CMS Preliminary

Events / ( 20 GeV )

1 10

Data Fit model σ 1 ± σ 2 ±

EBEE

(GeV)

γ γ

m

400 600 800 1000 1200 1400 1600

stat

σ (data-fit)/

• 2

2

(13 TeV, 0T)

• 1

0.6 fb

CMS Preliminary

Diphoton spectrum at 0 Tesla (13 TeV)

Barrel-Barrel Barrel-Endcap

SLIDE 22

New Physics with light SM particles at CMS – JPC – Rutgers University – Sunday, March 20th, 2016

DIPHOTON RESONANCES

• Combined 8 TeV + 13 TeV results
• Largest excess is observed for 750 GeV, spin-0, narrow width
• local significance of 3.4σ, 1.6σ after look-elsewhere effect
• Dec ’15 result: largest excess at 760 GeV for Γ/M=1.4x10-2
• local significance of ~3σ, <1.7σ after look-elsewhere effect

10

[EXO-16-018]

(GeV)

S

m

2

10 × 5

3

10

3

10 × 2

3

10 × 3

) (fb) γ γ → S → (pp σ 95% C.L. limit

2 4 6 8 10 12

J=0

• 2

10 × = 1.4 m Γ Expected limit σ 1 ± σ 2 ± Observed limit

(8 TeV)

• 1

(13 TeV) + 19.7 fb

• 1

3.3 fb

CMS Preliminary

combined limits for 8 + 13 TeV

(GeV)

S

m

2

10 × 5

3

10

3

10 × 2

3

10 × 3

p

• 4

10

• 3

10

• 2

10

• 1

10

J=0

• 2

10 × = 1.4 m Γ Combined 8TeV 13TeV

σ 1 σ 2 σ 3

(8 TeV)

• 1

(13 TeV) + 19.7 fb

• 1

3.3 fb

CMS Preliminary

combined p-values for 8 + 13 TeV

C M S

SLIDE 23

A T L A S

Jan Stark for the ATLAS collaboration Moriond QCD -- March 19-26, 2016 10

Di-photons: search for spin-0 resonance

ATLAS-CONF-2016-xxx

Selection optimised for Higgs-like signal:

• two Photons

(tight identification)

• photons required

to be isolated

• Photon transverse

energies: ET(γ1) > 0.4 mγγ ET(γ2) > 0.3 mγγ (effectively depletes forward regions) Background modelled using fit to functional form

SLIDE 24

Jan Stark for the ATLAS collaboration Moriond QCD -- March 19-26, 2016 11

Di-photons: search for spin-0 resonance

Perform 2D p0 scan (as function of mass and width

• f the hypothetical resonance).

Largest deviation from background-only hypothesis: near 750 GeV width  45 GeV (i.e. 6%) Local significance: 3.9σ Global significance: 2.0σ Report limits on fiducial cross section as a function of mass hypothesis, for several width hypotheses. Example shown here: width of 6% ATLAS-CONF-2016-xxx

A T L A S

SLIDE 25

A T L A S

Jan Stark for the ATLAS collaboration Moriond QCD -- March 19-26, 2016 14

Compatibility with the s = 8 TeV data

ATLAS-CONF-2016-xxx

Spin 0 Spin 2

8 TeV data: 1.9σ deviation from B-only hypothesis at mX = 750 GeV, ΓX/mX = 6% Assuming common signal model; production cross-section scales like Parton luminosities gg s-channel = 4.7 qq s-channel = 2.7 Compatibility 8 TeV ↔ 13 TeV (gg hypothesis): 1.2σ Compatibility 8 TeV ↔ 13 TeV (qq hypothesis): 2.1σ 8 TeV data: no excess in the region of interest Compatibility 8 TeV ↔ 13 TeV (gg hypothesis): 2.7σ Compatibility 8 TeV ↔ 13 TeV (qq hypothesis): 3.6σ

SLIDE 26

LHC 2016 run

• LHC is operational
• beam commissioning has started Easter 2016
• Looking forward to integrate some 25 fb-1
• Large potential for discoveries in the coming runs

SLIDE 27

High Lumi LHC (HL-LHC)

• FP7 Design Study completed in 2015
• first 11 Tesla magnet
• Now underway as a project at CERN
• and recognised as a landmark on the ESFRI list 2016

β β

SLIDE 28

• F. Gianotti, EPS-HEP 2015, Vienna

6

The present and near/medium-term future: LHC and HL-LHC

30 fb-1 3000 fb-1

7-8 TeV 13-14 TeV

300 fb-1

Splices fixed Injectors upgrade New low-β* quads

LHC is highest-E, highest-L operational collider à full exploitation (√s ~ 14 TeV, 3000/fb) is mandatory:

q If new physics discovered in Run 2-3: à first detailed exploration of new physics with well understood machine and experiments q If no new physics in Run 2-3: à extend direct discovery potential by ~ 20-30% (up to m ~ 8 TeV)

In either case: measure H couplings to few percent (including 2nd generation: Hμμ)

Run 1 Run 2 Run 3 HL-LHC

F.Gianotti EPS 2015

SLIDE 29

Energy Frontier

• To stay at the forefront of particle physics, Europe needs to be in a position to

propose an ambitious post-LHC accelerator project at CERN by the time of the next Strategy update, when physics results from the LHC running at 14 TeV will be available. 
 
 CERN should undertake design studies for accelerator projects in a global context, with emphasis on proton-proton and electron- positron high-energy frontier machines. These design studies should be coupled to a vigorous accelerator R&D programme, including high-field magnets and high-gradient accelerating structures, in collaboration with national institutes, laboratories and universities worldwide.

SLIDE 30

Future Circular Collider FCC

• European Design Study
• ~100 TeV pp in a ~100 km ring

SLIDE 31

High-field magnets

• Key to high energies (both HE-LHC and FCC)
• Nb3Sn may lead to ~16 T magnets
• HL-LHC magnets provide a ~1.2 km test of the technology
• an insert of HTS may increase field to 20 T

SLIDE 32

Magnets R&D

16 T Dec 2015: 2 in 1 dipole

• f 1.8 m length reaches

nominal 11.3 T. Nb3Sn matrix LHC: nominal 8.3 T; exercise 9 T ? HL-LHC:

• 11 T dipoles in dispersion suppression collimators
• 12-13 T low-𝛾 quadrupoles ATLAS and CMS IR’s

SLIDE 33

Conceptual Design Report by end 2018

• pp-Collider (FCC-hh) – sets the boundary conditions
• 100 km ring, √s=100 TeV, L~2x1035
• HE-LHC is included (~30 TeV)
• e+e--Collider as a possible first step
• √s= 90 - 350 GeV, 


L~1.3x1034 at high E

• eh-Collider
• √s=3.5 TeV, L~1034

SLIDE 34

Site investigations @ CERN

• Studies are site independent. – FCC@CERN benefits from existing

infrastructure.

90-100 km ring fits geology

SLIDE 35

3-D Model and Injectors @ CERN

High-Energy Booster (HEB) is “refurbished” LHC – New power converters to achieve fast ramp (50 A/s) – Resulting filling time 30 mins

SLIDE 36

FCC-hh Parameters

Parameter FCC-hh SppC LHC HL LHC

collision energy cms [TeV] 100 71.2 14 dipole field [T] 16 20 8.3 # IP 2 main + 2 2 2 main + 2 bunch intensity [1011] 1 1 (0.2) 2 1.1 2.2 bunch spacing [ns] 25 25 (5) 25 25 25 luminosity/Ip [1034 cm-2s-1] 5 ~25 12 1 5 events/bunch crossing 170 ~850 (170) 400 27 135 stored energy/beam [GJ] 8.4 6.6 0.36 0.7 E-loss/turn synchrotron radiation/beam 5 MeV 3 MW 2 MeV 5.8 MW 7 keV 5.4 kW 7 keV 9.5 kW

SLIDE 37

Layout of FCC-hh and FCC-ee

• Closed orbit solution now available for both machines.

2 rings + 
 1 booster
 in FCC-hh tunnel FCC-hh FCC-ee FCC-ee compatible with FCC-pp tunnel layout

SLIDE 38

Circular Lepton Colliders

parameter

FCC-ee CepC LEP2

energy/beam [GeV] 45 120 175 120 105 bunches/beam 90000 770 78 50 4 beam current [mA] 1450 30 6.6 16.6 3 luminosity/IP x 1034 cm-2s-1 70 5 1.3 2.0 0.0012 energy loss/turn [GeV] 0.03 1.67 7.55 3.1 3.34 synchrotron power [MW] 100 103 22 RF voltage [GV] 0.08 3.0 10 6.9 3.5

FCC-ee

• 2 rings
• 2 IP with crab

waist CepC (China) – 1 ring with possible double ring sections

SLIDE 39

FCC detector concepts for 100 TeV

Driving requirements: BL2 ~10 x ATLAS/CMS for 10% muon momentum resolution at 10-20 TeV. Requires 1µm resolution

• large-bore, high-field solenoid
• return flux captured by twin solenoid
• Coverage with tracking and precise calorimetry up to

|𝜃|~5 for light particles

• forward dipole à la LHCb: B~10 Tm

SLIDE 40

Possible Timeline

Constr. Physics

LEP

Construction Physics Proto Design

LHC

Construction Physics Design

HL-LHC

Physics Construction Proto

1980 1985 1990 1995 2000 2005 2010 2015 2020 2025 2030 2035

20 years

Design

FCC

CDR by end 2018 for next ES update

SLIDE 41

Compact Linear Collider CLIC

• e+e- collider 1-3 TeV
• currently only option for the TeV

region

• exploring 380 GeV operation

(klystrons?)

• CDR 2013
• CTF3 has provided key results (end 2016)
• ready for a demonstrator

SLIDE 42

ν-physics

• Rapid progress in neutrino oscillation physics, with significant European

involvement, has established a strong scientific case for a long-baseline neutrino programme exploring CP violation and the mass hierarchy in the neutrino sector. CERN should develop a neutrino programme to pave the way for a substantial European role in future long-baseline experiments. 
 
 Europe should explore the possibility of major participation in leading long- baseline neutrino projects in the US and Japan.

SLIDE 43

LBNF / DUNE

SLIDE 44

LAr Technology

• LarTPC large scale active detectors
• few mm precision
• good energy resolution

SLIDE 45

Membrane cryostats
 GTT license

SLIDE 46

Neutrino Platform at CERN

These prototype detectors will generate a data stream comparable to that of ALICE in Heavy Iron Running

SLIDE 47

LHC Computing

• WLCG has been

performing extremely well

• so far experiments

have found ways

• f dealing with the

computing

• But have collected
• nly 1% of

luminosity to date

LHCC; 1st March 2016 Ian Bird; CERN 1

HI data: up to 450 TB/day during HI run (2012 max was ~220 TB/day) 2015: LHC: 26/36 PB 27 PB 15 PB 23 PB

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

Jan-10 Mar-10 May-10 Jul-10 Sep-10 Nov-10 Jan-11 Mar-11 May-11 Jul-11 Sep-11 Nov-11 Jan-12 Mar-12 May-12 Jul-12 Sep-12 Nov-12 Jan-13 Mar-13 May-13 Jul-13 Sep-13 Nov-13 Jan-14 Mar-14 May-14 Jul-14 Sep-14 Nov-14 Jan-15 Mar-15 May-15 Jul-15 Sep-15 Nov-15 Jan-16 Billion HS06-hours

CPU Delivered HS06-hours/month

alice atlas cms lhcb

2015

q

Pledged resources for 2016 matching requests

q

Installation on track for data taking

SLIDE 48

Development of CPU hours

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

Jan-10 Mar-10 May-10 Jul-10 Sep-10 Nov-10 Jan-11 Mar-11 May-11 Jul-11 Sep-11 Nov-11 Jan-12 Mar-12 May-12 Jul-12 Sep-12 Nov-12 Jan-13 Mar-13 May-13 Jul-13 Sep-13 Nov-13 Jan-14 Mar-14 May-14 Jul-14 Sep-14 Nov-14 Jan-15 Mar-15 May-15 Jul-15 Sep-15 Nov-15 Jan-16 Billion HS06-hours

C PU!Delivered!HS 06-hours / month

alic e atlas c ms lhc b

SLIDE 49

Example: LHCb Upgrade for 40 MHz readout

• Triggerless readout
• complete detector read at

40 MHz

• using fast switches to distribute

events to computing farm

Detector'front*end'electronics'

Eventbuilder'network'

Eventbuilder'PCs''(so8ware''LLT)' Even=ilter'Farm' ~'80'subfarms' UX85B' Point'8'surface'

subfarm' switch'

TFC'

500'

6'x'100'Gbit/s'

subfarm' switch' Online' storage'

Clock'&'fast'commands'

8800' VersaQle'Link''

throRle'from' PCIe40' Clock'&'fast' commands' 6'x'100'Gbit/s'

Removal of coarse trigger opens the research for rare phenomena in his rate charm quark events.

SLIDE 50

Computing development anticipated for LHCb readout

• Estimate of the

performance growth for LHCb

• comparison to

2010 High Level Trigger (HLT)

SLIDE 51

Universal trends in Science and beyond

• Data volume will increase dramatically
• Smaller and embedded sensors become affordable
• Signals sampled in time and space with high resolution
• Efficient extraction of information is important
• sophisticated algorithms emerge
• machine learning and more

SLIDE 52

European Cloud Initiative

• There is no technical obstacle to

establishing an Open European Science Cloud today

• Clear trend to make data of

publicly funded research publicly available

• Issues are governance and

persistence

SLIDE 53

• This position paper is a rallying call for

adoption of a strategic approach

• http://dx.doi.org/10.5281/zenodo.34264


November 2015, 26 pages

• Endorsed by the Director Generals of


all EIROforum members and accompanied by a statement of intent to enact this strategy

SLIDE 54

Example: HNSciCloud Joint Pre-Commercial Procurement

• Procurers: CERN, CNRS, DESY, EMBL-EBI, ESRF,
• IFAE, INFN, KIT, SURFSara, STFC
• Experts: Trust-IT & EGI.eu
• The group of procurers have committed
• >1.6M€ of procurement funds
• Manpower for testing/evaluation
• Use-cases with applications & data
• In-house IT resources
• To procure innovative IaaS level cloud services integrated into a hybrid

cloud model

• Commercial cloud services
• European e-Infrastructures
• Services will be made available to end-users from many research

communities

• Co-funded via H2020 (Jan’16-Jun’18)
• Grant Agreement 687614

Total procurement commitment >5M€

SLIDE 55

Science user groups to be supported by HNSciCloud

SLIDE 56

Outlook

• Computing needs will grow
• Access to publicly funded resources has to be eased
• Open Scientific Cloud may provide the sustainable solution to future needs