Документ взят из кэша поисковой машины. Адрес оригинального документа : http://xmm.vilspa.esa.es/calibration/documentation/epic_cal_meetings/200610/Kirsch_previewtool.pdf
Дата изменения: Wed May 14 13:06:11 2014
Дата индексирования: Wed Apr 13 13:37:05 2016
Кодировка:

Поисковые слова: п п п п п п
XMM-Newton Calibration (Pre-)view Tool
(overall coordination, early processing scripts, idl plotting)

M.G.F. Kirsch
1)

1)

(refined processing scripts and automated xspec data fitting)

M. Stuhlinger

& Elena Gonzales3)

(some RGS related routines and other ideas)

Andy Pollock1)

(automation of data processing, version control and interfaces to the ESAC grid)

Stephan Doerr

2)

Daniel Harbarth
1) 2)

2)

(front end web system)

European Space Agency University of Tuebingen 3) University of Madrid

Marcus G. F. Kirsch
XMM-Newton EPIC Page 1


starting situation

· Calibration status is updated once per year · User has no information between these updates what (s)he can expect · Cross cal document will have same frequency of updates

· Information flow is to slow and unclear ­ Should I wait for the new calibration to reprocess or not ?
Marcus G. F. Kirsch
XMM-Newton EPIC Page 2


idea - proposal

· Calibration (pre-)view tool
­ browse cross cal examples for various targets for
· actual SAS/Calibration · new upcoming versions --> Improving transparency of current calibration work

­ Giving expectation horizon ­ User will be able to judge if (near) future calibration will impact his analysis
Marcus G. F. Kirsch
XMM-Newton EPIC Page 3


implementation
XSA Automated fitting tool

web page
Version n Version n+1

data processing and spectral extraction Cross Cal data archive Cal scientist can tick to publish information Tick box to publish AJAX based web application

fully automated + version control
Upload of other interesting calibration progress

news web page
Comment to be made public

Comment to be made public

Marcus G. F. Kirsch
XMM-Newton EPIC Page 4


Tool released · http://xmm.vilspa.esa.es/cgi-doc/ept/preview.pl · Using ESAC grid:
­ 10 nodes so far, each node has 2 CPUs Intel(R) Xeon(TM) 3.00GHz with 2GB of memory and hyper-threading ­ That reduces processing time by factor ~15-20 -->15-25 observations/hour ­ further speed up possible by · parallel fitting · enhancing data-caching on the grid

· Current content:
­ 234 observations ­ 50 ready to process
Marcus G. F. Kirsch
XMM-Newton EPIC Page 5


features
· · · website as the front end of the tool easy access the cross calibration data archive comparing joint and individual fits and their parameters and fluxes for various targets and observations concerning different calibration versions Features for every FIT (one table entry):
­ pop up and enlarge the spectrum ­ display the associated logfile

­ display the parameter (extraction of some values of the log file)

·

Features for every version (a whole column):

­ plot parameter over all versions this executes an IDL script, which plots all n parameter values of all cameras versus the versions. A pop-up shows the n plots ­ plot flux over all version a similar IDL script plots the flux in different energy ranges versus the versions. For every energy range one plot ­ Observation Log Browser executes the Observation Log Browser for the given observation id.

·

Second level data products regarding statistical flux comparison (only dev Marcus G. F. version)

More

·

Features for all cameras and all versions (the whole table)

in

­ see flux for all cameras shows a pop-up which contains all flux for all cameras in a table

Martin

­ see parameters for all cameras shows a pop-up which contains all parameters for all cameras in a table

Stuhlingers

­ display the associated readme file

talk

Kirsch

XMM-Newton EPIC Page 6


rel flux in diff. energy bands

SAS 6.5

SAS 7.0

Marcus G. F. Kirsch
XMM-Newton EPIC Page 7


let us check your new calibration updates

Marcus G. F. Kirsch
XMM-Newton EPIC Page 8