Difference between revisions of "Data Analysis"

From BESIII Ferrara Group Wiki
Jump to: navigation, search
Line 2: Line 2:
 
Data are created by Alberto's software -> GUFI with the following format:
 
Data are created by Alberto's software -> GUFI with the following format:
  
- A folder labelled "day_time"
+
- A folder labelled "RUN_$runNumber"
  
- Inside there are two kind of file: a logbook and a file for each enabled ROC
+
- Inside there are three kind of file: logbook of the run, logbook of the acquisition and a file for each enabled ROC
  
 
-Format of the logbook ??
 
-Format of the logbook ??
  
 
-Format of the data ??
 
-Format of the data ??
 +
 +
The data are saved on the acquisition computer.
 +
 +
For each folder in /home/cgemlab2/data then a copy is created on srv-lab
 +
 +
In srv-lab the data are stored in /home/ihep_data/data
 +
 +
In the data folder there are several sub-folders for each step:
 +
 +
raw_dat -> output from ROC
 +
 +
raw_root -> decoded data in two steps: event.root and ana.root
  
 
== '''Data analysis:''' ==
 
== '''Data analysis:''' ==

Revision as of 04:22, 30 May 2019

Data creation:[edit | edit source]

Data are created by Alberto's software -> GUFI with the following format:

- A folder labelled "RUN_$runNumber"

- Inside there are three kind of file: logbook of the run, logbook of the acquisition and a file for each enabled ROC

-Format of the logbook ??

-Format of the data ??

The data are saved on the acquisition computer.

For each folder in /home/cgemlab2/data then a copy is created on srv-lab

In srv-lab the data are stored in /home/ihep_data/data

In the data folder there are several sub-folders for each step:

raw_dat -> output from ROC

raw_root -> decoded data in two steps: event.root and ana.root

Data analysis:[edit | edit source]

The data are threated in two step:

1- Wenlong code, renamed "TIGER Event Reconstruction" that convert the data collected by each ROC to a root file

2- GRAAL to digitize and clusterize the hit from each channel/strip and then performe the tracking

TIGER Event Reconstruction[edit | edit source]

The creation of the root file follow several steps:

1- [Script in python] Deconding the data from the ROC and it extract the variables into a root file: ecoarse, efine, tcoarse, tfine, strip ID, chip ID, FEB ID, ROC ID, frameword, altro?

2- Unify the root files from each ROC

3- [ana.C] Calibrate charge and time as a function of the FEB/channel

4- [event.C] Sort the entries as a function of the time, measure the trigger time and create the event

GRAAL[edit | edit source]

The reconstruction and the analysis in GRAAL is divided in three steps:

1- [$graal/src/rec] Prenalysis on the hits and clusterization

2- [$graal/src/evt] Tracking and good cluster selection

3- [$graal/src/ana] Generation of the PDF with the informations of charge, time and occupancy as a function of the FEB