reana-2011-jet-inclusivecrosssection-analysis

Analysis structure

1. Input Data

In this example the input data is read from eospublic.cern.ch//eos/opendata in the first step of the workflow. The input data is in the format of CMS AOD root files. There are two root files, one containing ntuples of detected data and the other ntuples of simulated Monte Carlo data.

2. Analysis Code

This example uses the ROOT analysis framework with the custom user code located in the code directory.

• A pure ROOT6 container image is used for most steps, this analysis needs no specific tools. The five steps of the analysis are:

  1. FillHistos

  The first part of the code reads the input root files and fills appropriate histograms. The histograms get categorized according to (1) eta intervall and (2) the trigger that triggered the detection.

  2. NormalizeHistos

  This part of the analysis normalizes the histograms with the following command:

  for int i from 1 to hpt->GetNbinsX()+1
    double norm = hpt->GetBinWidth(i) * etawid * hlumi->GetBinContent(i);
    hpt->SetBinContent(i, hpt->GetBinContent(i) / norm);
    hpt->SetBinError(i, hpt->GetBinError(i) / norm);
  

  3. CombineHistos

  Combine the histograms that are detected with different triggers but within the same eta width. Every trigger is the most efficent within different pt ranges. Use the detected events within a certain pt range from the trigger who have the largest effectivness within that range.

  4. Theory

  Here the data gets fitted to a theory curve, a NLO curve. The curves can be found at

  http://www-ekp.physik.uni-karlsruhe.de/~rabbertz/fastNLO_LHC/InclusiveJets/
  -> fnl2342_cteq66_aspdf_full.root
  Numbering scheme explained in
  https://twiki.cern.ch/twiki/bin/view/CMS/CMSfastNLO
  2-point to 6-point theory uncertainty:
  h200X00->h300X09, h400X00->h300X08 
  

  5. Dagostini

  Last step is unfolding of the data. The method used is called d'Agostini ("Baysian" or Richardson-Lucy) unfolding and the code includes a response matrix generation from NLO theory and parameterized JER.

3. Compute environment

In order to be able to rerun the analysis even several years in the future, we need to "encapsulate the current compute environment", for example to freeze the ROOT version our analysis is using. We shall achieve this by preparing a docker container image for our analysis steps.

The analysis steps will run in a pure ROOT analysis environment. We can use an already existing container image, for example reana-env-root6 for these steps.

4. Analysis workflow

Before calculating the inlcuisve jet crossection a few necessary steps are performed. The flow of the workflow is visualized below. The workflow language used here is CWL. CWL is a specification for describing analysis workflows and tools in a way that makes them portable and scalable. With the help of CWL we are able to run the first three steps in parallel where after the steps get more entangled. This optimizes the workflow, in comparison to having everything run in a serial manner.

  +-----------+             +-------+
  | Fill Data |             |Fill MC|   
  +-----------+             +-------+  
       |                       |
       |                       |        
       v                       v  
  +---------------+       +------------+
  | Normalize Data|       |Normalize MC|   
  +---------------+       +------------+  
       |                       |
       |                       |        
       v                       v 
  +-------------+         +----------+
  | Combine Data|         |Combine MC|   
  +-------------+         +----------+  
       |                        |
       |     +------------+     |        
       +---> | Theory Data|<--- +
       |     +------------+     |
       |      +------------+    |        
       +--->  |  Theory MC |<---+
       |      +------------+    |
       |             |          |
       |             |          |
       +-------------+----------+ 
       |                        |
       v                        v
 +--------------+        +------------+
 | Dagostini Data|       |Dagostini MC|
 +--------------+        +------------+ 
       |
       |
       v
 +------------+
 | Draw plots |
 +------------+