r75 - 26 Dec 2007 - 09:37:41 - HongMaYou are here: TWiki >  AtlasSoftware Web > BnlJamboreeDec2007

BnlJamboreeDec2007


Introduction

This Analysis Jamboree at the BNL ATLAS Analysis Support Center focuses on understanding how the analysis of initial data should be organized, and what roles the participants would like to play. We expect to discuss a few combined performance and early physics topics. This can be studied in the context of streaming data, if they become available in the time frame. The detailed agenda will be set up soon.

Dates and Agenda Outline

The dates for the Jamboree
Dec 17-19, 2007

The agenda for the Jamboree will be filled at here

The participants are welcome to stay longer at BNL if they find it beneficial, but please check the availability of personnel you intend to interact with. We hope that this environment would provide an added benefit for bringing collaborators of similar physics interests to work together, not just at the Analysis Week, but thereafter as well.

Location

The meeting will take place in BNL Physics Department, Building 510A, Room 2-84 for general meetings, and additional space for breakout sessions in Room 1-105 and Room 1-113 in the same building, and a coffee lounge in Room 1-100. Physics Bldg map

Please refer to the BNL Map for directions to lab and the physics department.

Registration

We require a registration for the Jamboree because space is limited, and we would like to know in advance the topics to be discussed. There is no registration fee, but please send e-mail to Hong Ma (hma@bnl.gov) with cc to Linda Feierabend (feierabe@bnl.gov), notifying your attendance. Please let us know the dates of your visit, whether or not you are a US citizen, and if you already hold a BNL guest appointment.

Attendees who already hold a valid BNL Guest Badge may come on site without notification to the main gate. Attendees who already have approved guest appointments, but not a Guest Badge will have to notify the BNL main gate of their arrival by checking in here. US citizens without a guest appointment are allowed onsite, but we must know the dates of your visit so that the main gate can be notified.

All others must apply for the BNL guest appointment by completing the Brookhaven's Guest Registration Form well in advance of the meeting. You will be notified when the guest appointment is approved. When you arrive at BNL with the approved guest appointment for the first time, you must stop at the security office by the main entrance where you will be issued a temporary pass, then proceed to the RHIC/AGS Users' Office to complete the check-in procedure.

NON-US citizens without BNL guest appointment should complete the BNL Guest Registration form as soon as possible, since it takes additional time for the approval process, especially for people from "sensitive" countries.

Please also note that obtaining a BNL computer account at USATLAS Tier-1 requires an approved guest appointment (irrespective of citizenship).

BNL Visitor information is http://www.bnl.gov/bnlweb/userindex.html .

Registered participants

  • Hong Ma, Srini Rajagopalan, Scott Snyder, Frank Paige, David Adams, Stephanie Majewski, George Redlinger, Abid Patwa, Serban Protopopescu. Michael Begel, Mark Baker, Bruce Gibbard; BNL
  • Kamal Benslama, Eduardo J. Ortega U.; Regina
  • Simona Rolli, Tufts
  • James Mueller, Pittsburgh
  • Bob Kehoe, SMU
  • Kyle Cranmer, Akira Shibata; NYU
  • Ariel Schwartzman; SLAC
  • Joey Huston, Brian Martin; Michigan State
  • Kevin Black, Corrinne Mills, Ben Smith; Harvard
  • Stephane Willocq, Carlo Dallapiccola, Emily Thompson, German Colon; UMass
  • Jim Cochran; Iowa State
  • Peter Loch; Arizona
  • Bob Orr; Toronto
  • Ayana Holloway; LBNL
  • Masayuki Kondo; UT Dallas
  • Jason Nielsen; UC Santa Cruz
  • Saj Alam; Albany
  • Vivek Jain, Nirmalya Parua ; Indiana
  • Keith Baker, Jennifer Hsu; Yale
  • Ryan Reece, Brig Williams; U Penn.
  • Tetteh Nee Addy, Hampton U
  • John Hobbs, Katy Tschann-Grimm, Ashfaq Ahmad, Rod Engelmann; Stony Brook
  • Jeremy Love, Robert Harrington; Boston Univ.
  • Jae Yu, Andy White; UTA
  • Chris Hays, Oxford
  • Rachid Ayad, Temple University

Dining

A guide to some of the options are listed on the DiningNearBNL page.

Housing

On-site housing in the dorms is available (approx. $32/night). Summer housing is always tight, so we encourage you to reserve early. There are also on-site guest houses available at approximately $100 per night. For details, see BNL housing webpage http://www.bnl.gov/staffservices/OnsiteHousing/OnsiteHousing.asp . On-site housing is in high demand in the summer, please make the reservation early.

For reservation, please contact the house office directly. On-site housing for non-US citizens can only be confirmed after the guest appointment is approved.

For off-site housing, we suggest

  • Hampton Inn
    2000 North Ocean Avenue
    Farmingville, NY
    Phone : +1-631-732-7300
    Directions : East on the Long Island Express, Exit 63 North
    Rates : BNL rate ($109/night )
    Daily rate includes: Daily Continental Breakfast, Wireless Connections, Microwave, Refrigerator, Coffee Makers in each guest room

  • Radisson Hotel (on the site of the former Best Western)
    1730 North Ocean Avenue
    Holtsville, NY
    Phone : +1-631-758-2900
    Directions : East on the Long Island Expressway, Exit 63 South
    Rates : BNL Rate ($109/night, breakfast included)

Computing Environment

We assume everyone will bring a laptop to work on, and everyone has a BNL computer account at the Tier-1.

Network connections, Wireless

All computers connected to BNL network need to be registered with the BNL ITD. When you first connect to the network, your web browser will be connected to a page to register your computer. Enter the appropriate information on the page (your Guest number is needed), re-start the network connection, you will then be connected.

The physics building has visitor wireless network. The name of the wireless access point is Corus2. A few wired connections (inside firewall) are available in rooms 2-84 and 1-105. You need to set http proxy in the web browser after registration, for both wireless and wired connections, see i http://www.bnl.gov/cybersecurity/www.asp for instructions.

Tier-1 computer account.

If you have not logged onto your BNL account lately, you should be aware that using ssh private key is required. Follow this link

We will ask for additional interactive nodes be made available during the Jamboree. They are included in the pool of the interactive machines, so using "rterm" is sufficient.

Printers (modified since the last Jamboree).

Working groups and analysis topics

egamma (Kamal)

Topics for the working session:
  • EM calibration with Z to ee
  • Identification of low pT electrons
  • Identification of High pT electrons
  • Identification of Very high pT electrons (Exotics signals)
  • Performance of eID in SM and Discovery processes
  • Strategies for early data: Efficiency, fake rate,...etc
  • Analysis Examples of Stream test data
  • Photon-ID (TBC)

Jets (Peter)

The most important topics to be addressed quickly with the first data are related to the hadronic signals in general and the systematic error of the jet energy scale inparticular:
  • stability of the hadronic signal
    • topological cluster and towers
    • classification, calibration and corrections in case of clusters
      • effect of noise
  • jet energy scale
    • jet level corrections
      • jet-by-jet: use signal shapes in calorimeters
    • photon/jet balance
    • di-jet balance

We will prepare some CBNTAA or similar to address the following specific analysis topics:

  • pT balance as function of cluster jet shape
    • needs jet constituents (topo clusters)
  • pT balance using photons and clusters only
    • no jets if missing Et projection method is used
  • best if cluster could be looked at uncalibrated, and then with increasing levels of corrections

Dataset from 13.0.30.2 production:

  • trig1_misal1_mc12.006379.PythiaPhotonJet_AsymJetFilter.recon.AOD.v13003002
  • trig1_misal1_mc12.006379.PythiaPhotonJet_AsymJetFilter.recon.ESD.v13003002

Muon (Stephane)

Analysis & discussion topics (what do we need to know to do physics with muons?):

  • Efficiency
    • Trigger efficiency
    • Combined muon efficiency (with and without selection cuts)
    • Extrapolation from Z peak to high pT
  • Fake rate
    • V0 sample
    • Other samples?
  • Momentum scale and resolution
    • Resolution
    • Magnetic field
    • Energy loss (Calo material)
    • Wrong-charge fraction

EtMiss/Tau (Kyle)

Look at MissingET and Tau CSC note drafts as a starting point. Focus on sections aimed at early data

  • What ideas in CSC notes look most promising and should have more attention
  • Constructive criticism for techniques proposed in CSC note
  • New ideas or missing techniques for early data
  • Event Cleaning and fakes for MissingET

Serban will give a talk on D0 experience on Tuesday morning, so some of the tau discussion will happen on Tuesday.

Notes:

Standard Model (Joey)

Analysis topics: *We will be concentrating on the construction of Standard Model benchmark cross sections from the early data, with concentration on W/Z + jets. Many of the more complicated SM analyses, such as top pair production, as well as BSM analyses will be based on these final states. In addition, analyses of W/Z + jets exercises the ability to reconstruct leptons, jets and missing transverse energy, as well as the determinination of the corrections for resolution and hadronization.

Top (Jim)

Use standard TV ntuples to develop a 1st pass analysis with estimates of signal & background (primarily W+jets & QCD) and then try to extract a ttbar signal from the stream data. Akira will provide standard top group TV ntuples and a root analysis script (Balint ?). May also try to get a ttbar csec measurement (need stream luminosity) and/or other measurements (Mt).

SUSY (George)

Since this jamboree is focused on early days of data-taking, I would like to suggest thinking about QCD backgrounds. QCD is the first background (not just for SUSY) physics process we will encounter when we turn on.

Strategies for data-driven estimations of ttbar and W+jets backgrounds to SUSY have been well studied in the CSC note exercise. QCD backgrounds have been less well studied. Yet, before the background estimations of ttbar and W+jets can proceed, we will need to get QCD under control to ensure that it is not contaminating the control samples for the study of other backgrounds.

QCD backgrounds are notorious for not being amenable to Monte Carlo studies. Tevatron experiments seem to be using a mix of data-driven methods and simply squashing the background to tiny levels to suppress the systematics. Some of these methods are being studied in ATLAS. It would be interesting to exchange ideas with those with Tevatron experience.

Defining a strategy for estimating the QCD background will be tied to the performance of the reconstruction. It would be interesting to understand what level of performance can be expected in the early days, and how we intend to measure this performance in the absence of MC truth. This could help us narrow down the possible background estimation strategies that need to be explored.

Stream Test (Ayana)

See details about streaming data in this page: StreamingTestForUsers.

For the Jamboree, the only streaming data available are those mixed in release 12.

  • RDO dataset:
    • streamtest.00XXXX.inclYYY.digit.RDO.v12000399
  • release 12 AODs
    • streamtest.00XXXX.inclYYY.merge.AOD.v12000605
  • release 13 AODs
    • user.HongMa.streamtest.inclYYY.recon.AOD.v13003003_V1 * A copy of AODs are stored in /usatlas/scratch2/dladams/data/streamtest

XXXX=run numbers, and YYY = Ele, Muo, Tau, Jet and Pho

Analysis examples for direct AOD analysis and ARA in rel 13:

  • Some direct AOD examples are given in/afs/usatlas.bnl.gov/users/holloway/StreamingTools.tgz. Quick instructions to use this package:
    • Check out PhysicsAnalysis? /AnalysisCommon/UserAnalysis (-r UserAnalysis? -00-10-12 for release 13.0.30)
    • Check out LumiBlock? /LumiBlockComps ( -r LumiBlockComps? -00-00-09 for release 13.0.30)
    • Untar the UPDATED file in your $TestArea, and compile. If you downloaded the tarball on Monday, please get the new version! It was missing two files.
    • For experts/impatient users, the most important files in this tarball are: LumiBlock/LumiBlockComps/python/LumiQuery.py (needed to work with rel13 luminosity block metadata) and PhysicsAnalysis/AnalysisCommon/UserAnalysis/UserAnalysis/StreamTrigUtil.h (this is helpful for understanding the trigger configuration).
  • Analysis example with AthenaROOTAccess? in rel 13: See /usatlas/u/hma/workarea/13.0.30-Anaysis/ARA/README for details.

Useful links for Physics Analysis

  • WebSearch: Can't find the wiki page you want? Search the ATLAS Wiki pages.
  • SoftwareSupport : ATLAS Software in the US
  • AnalysisWithRelease12Portal: A portal for information relevant to physics analysis with release 12.0.x
  • For beginners, we strongly recommend you get familiar with the ATLAS computing environment, and take analysis software tutorials

Photo and News

BNL Today


Major updates:
-- TWikiAdminGroup - 22 Jul 2018

About This Site

Please note that this site is a content mirror of the BNL US ATLAS TWiki. To edit the content of this page, click the Edit this page button at the top of the page and log in with your US ATLAS computing account name and password.


Attachments


jpg JamboreeDec2007Photo3.jpg (1495.0K) | HongMa, 20 Dec 2007 - 22:42 | Jamboree Photo
 
Powered by TWiki
This site is powered by the TWiki collaboration platformCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback