L3 Algorithms: status and plans for the near future

1
L3 Algorithms status and plans for the near
future

• All-D? Meeting 1st March 2002
• Dan Claes and Terry Wyatt.

2
Design
Curently

3
Level 3 Jargon

• Tool Does the real work
• Unpacks raw data, finds tracks, clusters
• Identifies physics objects (e, ?, ?, jet ,?, W,
  Z)
• Filter Applies (simple) cuts on objects
• e.g. pt(?) gt 10 GeV
• Filter Script Defines a L3 trigger condition
• Logical .AND. of one or more filters
• If all filters in a script are .TRUE. trigger is
  satisfied and event is recorded
• Mark and Pass
• Selects unbiassed sample of input events to be
  recorded
• Poor mans efficiency calculation/monitoring

4
L3TJet Tool

• author Volker Buescher (Mainz)
• Rejection of L1-accepts makes use of
• high precision calorimeter readout available at
  L3
• simple cone algorithm
• identify (and reject) low-ET events passing L1
  trigger
• sharpen the turn-on curve

• running online stably since early Sept01

5
e.g., Fraction of events passing L3FJet(1,15)
DATA runs 142871-142874
pT of leading offline JCCA jet (GeV)

• Main effect smearing turn-on is lack of primary
  vertex at L3
• L3 NADA needed (Marumi Kado investigating)
• QCD group currently adjusting each L3 jet pt
  cut to value at which the relevant L1 trigger
  reaches 100 efficiency.

6
L3TEle Electron Tool authors Volker
Buescher (Mainz) Ulla
Blumenschein (Mainz)

• Current filter
• simple 0.4 cone
• applying cuts on
• ET
• emfrac (gt0.9)
• providing a rejection of 3

PrVtx L3TNullVertex() CAL_CLUS4L3TCalCluster(calu
npCAL_UNP, vertexPrVtx, conesize0.4,
MinSeedEt.5) ELE_LOOSE L3TEle(CALCAL_CLUS4,EMFR
0.9)
7
p11 certification data runs 142871 - 142874
Fraction of events passing L3Fele(1,5) vs pT of
leading offline EM object (GeV) after pre- and
standard-selections
8
p11 certification data runs 142871 - 142874
9
Improvements 1) Reconstruct EM Object within
smaller (0.25) cone
L3 Efficiency vs offline EM object pT
after pre- and standard-selections
ELE_LOOSE L3TEle(CALCAL_CLUS4, EMFR0.9
ConeSize0.25)
Rejection
10
Improvements 2) Additional discrimination
available through transverse shower shape
cut (shape variables based on an energy-weighted
?r) ?,? energy weighted cluster axis
position

11
ELE_LOOSE L3TEle(CALCAL_CLUS4,EMFR0.9,
ConeSize0.25, Width_EM10.085,
Width_EM20.085, Width_EM30.045)
Z?ee sample
Dr, floor 1
Dr, floor 3
data sample
12
Events rejected by pt cut e.m. fraction shower
shape
13
Triggerlist Proposals (single electron
triggers) Lower L3 threshold for inclusive
unprescaled EM trigger
Replace CEM(1,10)L3FEle(1,15,emfr) with
CEM(1,10)L3EM(1,12,emfr,cone0.25,shape)
CEM(1,10)L3EM(1,15,emfr,cone0.25) Repl
ace CEM(2,5)L3EM(1,10,emfr) with
CEM(2,5)L3EM(1,12,emfr,cone0.25,shape)
CEM(2,5)L3EM(1,15,emfr,cone0.25) Replace
CEM(1,5)L3EM(1,5,emfr) with
CEM(1,5)L3EM(1,7,emfr,cone0.25,shape)
CEM(1,5)L3EM(1,10,emfr,cone0.25)
Drop prescale?
Rejection 10.3
14
L3FTauHadronic Level3 TauTool author Yann
Coadou (Upsala)
Running online since 17-Jan-2002 Bug fix to
filter installed 09-February p11.01.00
certification included runs over Z??? QCD
MC data run 144102 (gt100K events)
non-optimized filter currently in triggerlist,
gathering information for tuning under online
conditions
15
Z??? 0 minbias
16
QCD pTgt20 GeV
17
Raw Data Run 144102, 104k events
18
L3TMuon responsibles Martin Wegner (Aachen),
Martijn Mulders
(Fermilab) (and the L3Muon working group)

• Short term goal to filter with L3TMuon
• cutting on muo_local pT
• with tight / loose / a-stub-like quality
  criteria
• Mainly needed for single muon triggers
• (which currently have a L1 prescale of 25)

19
Test Samples

• p11.01.00 performance/certification
• Processed following samples
• Single 5 GeV Muon MC, ?lt4.2 5000
  events
• Single 5-100 GeV MC, ?lt 2.5 5000
  events
• Golden J/psi samples
  300 events
• Special run 14747 mu1ptaxaa only 19324
  events
• Recent run 14890
  22995 events
• No crashes!
• Rejection factors gt4 wrt L1 (loose muons) but
• p11.01.00 efficiency 50-60 for offline
  tight quality muons

Attributed to bugs fixed in p11.02
20
Muon finding efficiency in p11.02 version
Low efficiency for tight in central region
still under investigation
21

• Most important outstanding issues
• 1) Memory leak
• approx. 1.5 kB/event
• (probably in the muon segment reconstruction
  code)
• (probably affecting both offline and L3)

22
Memory leak
23

• Most important outstanding issues
• 2) L3TMuoUnpack (unpacking tool)
• Current version requires a configuration
  file.
• When a module is removed from readout, without a
  corresponding change to the cfg.dat file,
• crate data will be ignored creating a hole in
  L3Muon acceptance
• A dynamic unpacker independent of rapidly
  changing
• configuration files (a fix provided by
  Scott Snyder)
• yet to be implemented.
• However, we hope to run p11.02 muon code
  online
• initially a test run, in mark and pass mode.
• aim for active filtering asap!

24
CFT Unpacking
Principal author Robert Illingworth (IMPERIAL)

• Recent improvement (p11.02)
• replace global threshold with individual channel
  thresholds
• (as used now in offline reco)
• Tested on real data
• gt89,000 events (Run 146562 17-FEB)
• Updating thresholds and cable maps
• requires no code changes
• but care in archiving/version-tagging (general
  problem!)
• Daniel Whiteson is studying the improvement this
  offers
• global tracking.

25
Level 3 Global Track Finding author Daniel
Whiteson (BERKELEY)
a global (SMT plus CFT) track
finder Find axial CFT tracks Match stereo
clusters Extend into SMT
26
Improvement in tracking from new CFT unpacking
Beam-spot corrected dca
track phi
27
Improvement in tracking from new CFT unpacking
i.e., SMT matching is failing completely!
Q/pt
q/pt
Number of smt hits
Number of tracks
chisquared
28
CFT Tracking Algorithm - L3TCFTTracker
Principal author Ray Beuselink (IMPERIAL)
P11 tool certification Robert Illingworth and
Chris Barnes
CFT unpacked and tracked gt 27000 assorted MC
events (delivered to IC farm by SAM) No
crashes, no memory leaks 10,000 events with full
(tsim_l3) triggerlist
29
Tracking will not be perfect for a long time.
However, our aim is to try and get tracking
turned on reasonably soon. For example, add
another electron trigger? CEM(1,10)L3EM(1,12,emfr,
cone0.25,shape) CEM(1,10)L3EM(1,12,emfr,cone0.25
,matched_track) CEM(1,10)L3EM(1,15,emfr,cone0.25)
and similarly for muons? Or run a standalone
track trigger at L3 for single muon
triggers? N.B. If you dont like it, you can
always exclude it by using the L3 trigger bits.
30
L3 Primary Vertex needed soon!
1) Histogram technique using SMT hits
author Guilherme Lima (UERJ/Brazil)
Has yet to be tested on REAL DATA Opportunity for
new person to get involved!
2) L3TVertexFinder Track-based vertexing tool
author Ray Beuselinck (Imperial)
Recently upgraded to use either CFT or Global
candidate tracks as input Per Jonsson
(Imperial) testing N.B. Marseille group (Arnaud
Duperrin, Mossadek Talby, Eric Kajfasz) will be
actively involved in testing tracking,
vertexing.
31

• Level3 Requirements for certification of Code
• Fully tested on a Linux system.
• Works "out of the box" (No private mods to code
  or RCP files)
• No crashes/memory leaks on samples of order 100K
  events
• real data and Monte Carlo.
• Timing studies
• Performance studies/plots DOCUMENTED
• Efficiency, rejection, physics distributions
• these tests may run the tool singly, but
• Must run on data with triggerlist exercising all
  released tools
• The filter code must be tested, as well as
  associated tools.
• Must test persistency of physics_results
• Write out events.
• Read back in to check physics distributions.
• Verification run trigger simulator on real data
  and check that results agree with those obtained
  online.
• Shadow nodes (in the future) to test new code
  online

32
L3 Monitoring

• L3 filter statistics for each trigger available
  to shift crew via daq_monitor
• physics_results for each tool written out on
  each accepted event
• debug_info
• l3fanalyze program produces rootuple
• Each tool must provide methods to fill rootuple
• Used for offline checks of data quality
• Plan to run online as examine use root to
  fill monitoring histograms from rootuple
• Extra person needed to work on this!
• L3 monitoring needs to get a lot more routine!

33
Conclusions, outlook.

• Currently factor 5 rejection needed at L3
• Calorimeter-based filtering (jets, electrons,
  taus) only
• Next steps (p11.02 release)
• Improve electron rejection (transverse shower
  shape)
• Commission muons, tracking, primary vertex
• Get more systematic monitoring for L3
• When L2 turns on well still need factor 5
  rejection at L3, but will have to work harder to
  achieve it!

34
To find out more

• L3 Algorithms web-pages
• http//www-d0.fnal.gov/computing/algorithms/level3
  /home.html
• L3 Algorithms working group meetings
• take place every on-week, Wednesday 1400-1530
  in the Farside
• Talk to Dan Claes (dclaes_at_unlhep.unl.edu) or
  Terry Wyatt (twyatt_at_fnal.gov)
• about the opportunities to get involved!