Difference: BrendanHomepage (1 vs. 14)

Revision 142011-04-21 - PeterWinter

Line: 1 to 1
 
META TOPICPARENT name="BrendanKiburg"
Changed:
<
<
-- BrendanKiburg - 30 Aug 2006

Current Tasks - 15 Jan 2008

general entrance counter related

  • Make sure musc_lo time is available in the main Event Tree
  • musclo && mupc
  • mupc && NOT musc_hi
  • in the raw analysis, extract the electron efficiency of the mupc planes by dividing ( musc_lo && mupc_x && mupc_y ) / (mupc_x && musc_lo)
  • also, (musc_lo && mupc_xycluster) / (musc_lo)
  • also, (musc_lo && mupc_xycluster && ! musc_hi) / (mupc_xycluster && musc_hi)
  • look at mu+to increase the electron fraction
  • also get the musc and mupc muon efficiencies by looking at similar parameters for the mta analysi
  • line up timing offset for musc_lo, musc_hi mupc counters, or define the windows
  • once the relative timing of the musc_lo and musc_hi are well understood, fit the relative phases of the musc_hi and musc_lo signals to see if the electrons and muons are indeed out of phase
  • investigate the musc_lo && mupc time distribution between 0 and 200 ns, where the coincidence seems to exceed the raw musc_lo signal

WFD

  • combine musc_lo information with the wfd -- this requires that all data quality cuts (sparks etc) also cut the wfd data, for proper normalization

mupc clustering

  • Upgrade the TCluster class to include all wires involved, not just the first one (then check for this in the auto/cross correlation plots)
  • mupc Clustering uses an updating deadtime when killing the wire

Trend plot related

  • Needed to include "time.h" on tungsten in BeamlineExtract.cpp
  • "trend_" modifies the entire expession : pathdir/filename.root , whereas it should be sandwhiched somwhere near filename.root
  • when the trendplot tree was turned on, the main tree was written, but not filled
  • Peter winter's email indicates that I need to get a bunch of the trend plot variables from the MHistory module that is not yet implemented

Coding

  • Add -n flag to the mta analysis to run through n events

Other general requests/suggestions from Peter

  • Make 2d plots for the entrance counters so we can relate mupc-t_kick and musclo - tkick
  • Make a tertiary tree that "skims" entrance counter information with 0.1% of regular events as well as all pileup signals
  • Show plots in terms of the kicker time rather than true autocorrelation due to the chopped nature of our beam
  • For main production, i also need to decrease the spark limit in Spark/TSparkTimes.h from 2200 back down to 200 (this was changed for Tim's Fe - Xray runs)
  • Also need to make sure that the "Save All Muons" flag in the odb is set to no for general production

Ongoing interests

  • Temporal distortions of lifetime spectra
  • MuStop definition
  • Kicker - Entrance counter dependencies

BKArchivedTaskList

>
>
Please go to https://muon.npl.washington.edu/twiki/bin/view/Main/BrendanHomepage
 \ No newline at end of file

Revision 132008-04-02 - BrendanKiburg

Line: 1 to 1
 
META TOPICPARENT name="BrendanKiburg"
-- BrendanKiburg - 30 Aug 2006

Current Tasks - 15 Jan 2008

Changed:
<
<
>
>

general entrance counter related

  • Make sure musc_lo time is available in the main Event Tree
  • musclo && mupc
  • mupc && NOT musc_hi
  • in the raw analysis, extract the electron efficiency of the mupc planes by dividing ( musc_lo && mupc_x && mupc_y ) / (mupc_x && musc_lo)
  • also, (musc_lo && mupc_xycluster) / (musc_lo)
  • also, (musc_lo && mupc_xycluster && ! musc_hi) / (mupc_xycluster && musc_hi)
  • look at mu+to increase the electron fraction
  • also get the musc and mupc muon efficiencies by looking at similar parameters for the mta analysi
  • line up timing offset for musc_lo, musc_hi mupc counters, or define the windows
  • once the relative timing of the musc_lo and musc_hi are well understood, fit the relative phases of the musc_hi and musc_lo signals to see if the electrons and muons are indeed out of phase
  • investigate the musc_lo && mupc time distribution between 0 and 200 ns, where the coincidence seems to exceed the raw musc_lo signal
 
Added:
>
>

WFD

  • combine musc_lo information with the wfd -- this requires that all data quality cuts (sparks etc) also cut the wfd data, for proper normalization

mupc clustering

  • Upgrade the TCluster class to include all wires involved, not just the first one (then check for this in the auto/cross correlation plots)
  • mupc Clustering uses an updating deadtime when killing the wire

Trend plot related

  • Needed to include "time.h" on tungsten in BeamlineExtract.cpp
  • "trend_" modifies the entire expession : pathdir/filename.root , whereas it should be sandwhiched somwhere near filename.root
  • when the trendplot tree was turned on, the main tree was written, but not filled
  • Peter winter's email indicates that I need to get a bunch of the trend plot variables from the MHistory module that is not yet implemented

Coding

  • Add -n flag to the mta analysis to run through n events

Other general requests/suggestions from Peter

  • Make 2d plots for the entrance counters so we can relate mupc-t_kick and musclo - tkick
  • Make a tertiary tree that "skims" entrance counter information with 0.1% of regular events as well as all pileup signals
  • Show plots in terms of the kicker time rather than true autocorrelation due to the chopped nature of our beam
  • For main production, i also need to decrease the spark limit in Spark/TSparkTimes.h from 2200 back down to 200 (this was changed for Tim's Fe - Xray runs)
  • Also need to make sure that the "Save All Muons" flag in the odb is set to no for general production
 

Ongoing interests

  • Temporal distortions of lifetime spectra
  • MuStop definition

Revision 122008-01-15 - BrendanKiburg

Line: 1 to 1
 
META TOPICPARENT name="BrendanKiburg"
-- BrendanKiburg - 30 Aug 2006
Changed:
<
<

Current Tasks - 4 Apr 2007

  • Quantifying background component due to cosmics / other sources
  • Generating UIUC numbers for the water impurity data for CHUPS paper
  • Generating run-by-run plots - Plotting any interesting histogram from mta vs. run# using a general directory structure script
  • Production Pass -
>
>

Current Tasks - 15 Jan 2008

 

Ongoing interests

Changed:
<
<
  • Capture time distribution fits
>
>
  • Temporal distortions of lifetime spectra
  • MuStop definition
  • Kicker - Entrance counter dependencies
 

BKArchivedTaskList

\ No newline at end of file

Revision 112007-04-03 - BrendanKiburg

Line: 1 to 1
 
META TOPICPARENT name="BrendanKiburg"
-- BrendanKiburg - 30 Aug 2006
Changed:
<
<

Current Tasks

>
>

Current Tasks - 4 Apr 2007

 
  • Quantifying background component due to cosmics / other sources
Changed:
<
<
>
>
  • Generating UIUC numbers for the water impurity data for CHUPS paper
  • Generating run-by-run plots - Plotting any interesting histogram from mta vs. run# using a general directory structure script
  • Production Pass -
 

Ongoing interests

  • Capture time distribution fits

Revision 102007-03-13 - BrendanKiburg

Line: 1 to 1
 
META TOPICPARENT name="BrendanKiburg"
-- BrendanKiburg - 30 Aug 2006

Current Tasks

Changed:
<
<
  • Small b-cut deuterium distributions
  • Deuterium linearity
  • Capture time distribution fits
>
>
  • Quantifying background component due to cosmics / other sources
 

Ongoing interests

Changed:
<
<
  • Creating cvs modules (or alternative CVS repository)
  • ROOT default settings
>
>
  • Capture time distribution fits
 

BKArchivedTaskList

\ No newline at end of file

Revision 92007-02-09 - BrendanKiburg

Line: 1 to 1
 
META TOPICPARENT name="BrendanKiburg"
-- BrendanKiburg - 30 Aug 2006

Current Tasks

Added:
>
>
  • Small b-cut deuterium distributions
  • Deuterium linearity
 
  • Capture time distribution fits
  • Production Pass
  • Generating run-by-run plots
Deleted:
<
<

BKArchivedTaskList

 

Ongoing interests

  • Creating cvs modules (or alternative CVS repository)
  • ROOT default settings
\ No newline at end of file
Added:
>
>

BKArchivedTaskList

 \ No newline at end of file

Revision 82007-02-08 - BrendanKiburg

Line: 1 to 1
 
META TOPICPARENT name="BrendanKiburg"
-- BrendanKiburg - 30 Aug 2006

Current Tasks

  • Capture time distribution fits
Changed:
<
<
  • Production Pass
>
>
 
  • Generating run-by-run plots

Revision 72007-02-08 - BrendanKiburg

Line: 1 to 1
 
META TOPICPARENT name="BrendanKiburg"
-- BrendanKiburg - 30 Aug 2006

Current Tasks

Changed:
<
<
  • Data Transfer
  • Security
  • Impurity analysis
  • Analysis structure
>
>
  • Capture time distribution fits
  • Production Pass
  • Generating run-by-run plots
 
Deleted:
<
<

Task list / progress 10 Oct 2006

Progress:
  • Updated mucap databases with CHUPS
  • Generated plots for meeting from analysis
  • Began improving the overall job submission structure
  • Meeting
Tasks:
  • Continue impurity analysis
    • Electron spectrum fitting with full kinetics
    • UIUC EVH time spectrum
  • Security issues with local machines and with PSI
  • Contact NCSA regarding job submission scheme
  • Use info from analysis meeting to make detailed checks in analysis plots
  • Complete data transfer when system is secured
 
Changed:
<
<

Task list / progress 27 Sep 2006

Progress:
  • Extracted basic UCB online info into a master root file for use with CHUPS slow control plots
  • Generated slow control plots for impurity data (with CHUPS)
  • Generated October meeting TWiki page
  • Adjusted the analyzer settings to match Steve's production MODULES without the skimmer
  • Started Production data pass
  • Ran analysis passes over several impurity data sets
Tasks:
  • Produce production data sets for subsequent analysis DONE
  • Produce plots for basic impurity lifetime analyses prior to Oct meeting DONE
  • improve analysis scripts (see 14 Sep tasklist), more automation
  • re-implement muSClo flag to kicker data
  • Finish up the data transfer and clean up "loose ends" (bad transfers etc.)
  • Begin storing tree data on NCSA as well
  • Consult with NCSA regarding optimal analysis job submission

Task list / progress 14 Sep 2006

Progress:
  • Some primitive results from pass 1 Analysis collected from NCSA
  • Built script to produce slow control information for the impurity calibration runs and submitted to twiki
  • Worked with Gabe/NCSA to get mucap1 in a firewall bypass to NCSA. The data transfer rates are currently hovering around 15 MB/sec when things are going well

Tasks [Analysis/Production]:
  • Extract UCB online analysis information from the /online_hists/ directory on cheetah to be able to plot on the slow control pages DONE
  • Generate October meeting TWiki page DONE
  • Pare down production modules to match steve: odb file, UIUC_ANALYZER module_switches, ANALYZER module_switches, MODULES file DONE
  • Run over some more data sets DONE
  • re-implement muSClo flag to kicker data
  • Improve analysis scripts
    • Have a single editable file which launches all of the jobs: generate lists based on database, grab files based on lists
    • Have a single production launcher to build the directory structure and launch the jobs in parallel for files in the lists with reasonably short wall times
    • Consult with NCSA concerning optimal job submission
  • Begin generating data sets for the production data - TWiki page
>
>

BKArchivedTaskList

 
Deleted:
<
<

Task list / progress 02 Sep 2006

  • Pass 1 Analysis submitted to tungsten cluster
  • 1/2 of data were missing, jobs ended quickly
  • some problems getting all data on scratch
  • craft script to launch get mss script followed by bsub upon completiong of mss ; -- to preserve
 \ No newline at end of file
Added:
>
>

Ongoing interests

  • Creating cvs modules (or alternative CVS repository)
  • ROOT default settings

Revision 62006-10-10 - BrendanKiburg

Line: 1 to 1
 
META TOPICPARENT name="BrendanKiburg"
-- BrendanKiburg - 30 Aug 2006

Current Tasks

Changed:
<
<
  • Database organization at NPL
  • Pass 0 using data from "online" analysis
  • Pass 1 using data at NCSA
  • Data Transfer from NPL to NCSA
  • October Meeting preparation
    • Plots
    • Reports
    • Wireless
    • Coffee / Bagels, donuts etc.
>
>
  • Data Transfer
  • Security
  • Impurity analysis
  • Analysis structure

Task list / progress 10 Oct 2006

Progress:
  • Updated mucap databases with CHUPS
  • Generated plots for meeting from analysis
  • Began improving the overall job submission structure
  • Meeting
Tasks:
  • Continue impurity analysis
    • Electron spectrum fitting with full kinetics
    • UIUC EVH time spectrum
  • Security issues with local machines and with PSI
  • Contact NCSA regarding job submission scheme
  • Use info from analysis meeting to make detailed checks in analysis plots
  • Complete data transfer when system is secured
 

Task list / progress 27 Sep 2006

Progress:
  • Extracted basic UCB online info into a master root file for use with CHUPS slow control plots
Line: 21 to 32
 
  • Ran analysis passes over several impurity data sets
Tasks:
  • Produce production data sets for subsequent analysis DONE
Changed:
<
<
  • Produce plots for basic impurity lifetime analyses prior to Oct meeting
>
>
  • Produce plots for basic impurity lifetime analyses prior to Oct meeting DONE
 
  • improve analysis scripts (see 14 Sep tasklist), more automation
  • re-implement muSClo flag to kicker data
  • Finish up the data transfer and clean up "loose ends" (bad transfers etc.)

Revision 52006-09-28 - BrendanKiburg

Line: 1 to 1
 
META TOPICPARENT name="BrendanKiburg"
-- BrendanKiburg - 30 Aug 2006

Current Tasks

Line: 17 to 17
 
  • Generated slow control plots for impurity data (with CHUPS)
  • Generated October meeting TWiki page
  • Adjusted the analyzer settings to match Steve's production MODULES without the skimmer
Changed:
<
<
  • Started Production data generation
>
>
  • Started Production data pass
 
  • Ran analysis passes over several impurity data sets
Tasks:
Added:
>
>
  • Produce production data sets for subsequent analysis DONE
 
  • Produce plots for basic impurity lifetime analyses prior to Oct meeting
  • improve analysis scripts (see 14 Sep tasklist), more automation
  • re-implement muSClo flag to kicker data
  • Finish up the data transfer and clean up "loose ends" (bad transfers etc.)
  • Begin storing tree data on NCSA as well
Added:
>
>
  • Consult with NCSA regarding optimal analysis job submission
 

Task list / progress 14 Sep 2006

Progress:
  • Some primitive results from pass 1 Analysis collected from NCSA
  • Built script to produce slow control information for the impurity calibration runs and submitted to twiki
  • Worked with Gabe/NCSA to get mucap1 in a firewall bypass to NCSA. The data transfer rates are currently hovering around 15 MB/sec when things are going well

Tasks [Analysis/Production]:
Changed:
<
<
  • Extract UCB online analysis information from the /online_hists/ directory on cheetah to be able to plot on the slow control pages
  • Generate October meeting TWiki page
  • Pare down production modules to match steve: odb file, UIUC_ANALYZER module_switches, ANALYZER module_switches, MODULES file
  • Run over some more data sets
>
>
  • Extract UCB online analysis information from the /online_hists/ directory on cheetah to be able to plot on the slow control pages DONE
  • Generate October meeting TWiki page DONE
  • Pare down production modules to match steve: odb file, UIUC_ANALYZER module_switches, ANALYZER module_switches, MODULES file DONE
  • Run over some more data sets DONE
 
  • re-implement muSClo flag to kicker data
  • Improve analysis scripts
    • Have a single editable file which launches all of the jobs: generate lists based on database, grab files based on lists

Revision 42006-09-27 - BrendanKiburg

Line: 1 to 1
 
META TOPICPARENT name="BrendanKiburg"
Deleted:
<
<
 -- BrendanKiburg - 30 Aug 2006

Current Tasks

  • Database organization at NPL
Line: 8 to 6
 
  • Pass 0 using data from "online" analysis
  • Pass 1 using data at NCSA
  • Data Transfer from NPL to NCSA
Changed:
<
<
>
>
  • October Meeting preparation
    • Plots
    • Reports
    • Wireless
    • Coffee / Bagels, donuts etc.

Task list / progress 27 Sep 2006

Progress:
  • Extracted basic UCB online info into a master root file for use with CHUPS slow control plots
  • Generated slow control plots for impurity data (with CHUPS)
  • Generated October meeting TWiki page
  • Adjusted the analyzer settings to match Steve's production MODULES without the skimmer
  • Started Production data generation
  • Ran analysis passes over several impurity data sets
Tasks:
  • Produce plots for basic impurity lifetime analyses prior to Oct meeting
  • improve analysis scripts (see 14 Sep tasklist), more automation
  • re-implement muSClo flag to kicker data
  • Finish up the data transfer and clean up "loose ends" (bad transfers etc.)
  • Begin storing tree data on NCSA as well
 

Task list / progress 14 Sep 2006

Progress:
  • Some primitive results from pass 1 Analysis collected from NCSA
Line: 24 to 40
 
    • Have a single editable file which launches all of the jobs: generate lists based on database, grab files based on lists
    • Have a single production launcher to build the directory structure and launch the jobs in parallel for files in the lists with reasonably short wall times
    • Consult with NCSA concerning optimal job submission
Deleted:
<
<
  • Generate October meeting TWiki page
 
  • Begin generating data sets for the production data - TWiki page

Task list / progress 02 Sep 2006

Revision 32006-09-14 - BrendanKiburg

Line: 1 to 1
 
META TOPICPARENT name="BrendanKiburg"
Line: 10 to 9
 
  • Pass 1 using data at NCSA
  • Data Transfer from NPL to NCSA
Added:
>
>

Task list / progress 14 Sep 2006

Progress:
  • Some primitive results from pass 1 Analysis collected from NCSA
  • Built script to produce slow control information for the impurity calibration runs and submitted to twiki
  • Worked with Gabe/NCSA to get mucap1 in a firewall bypass to NCSA. The data transfer rates are currently hovering around 15 MB/sec when things are going well

Tasks [Analysis/Production]:
  • Extract UCB online analysis information from the /online_hists/ directory on cheetah to be able to plot on the slow control pages
  • Generate October meeting TWiki page
  • Pare down production modules to match steve: odb file, UIUC_ANALYZER module_switches, ANALYZER module_switches, MODULES file
  • Run over some more data sets
  • re-implement muSClo flag to kicker data
  • Improve analysis scripts
    • Have a single editable file which launches all of the jobs: generate lists based on database, grab files based on lists
    • Have a single production launcher to build the directory structure and launch the jobs in parallel for files in the lists with reasonably short wall times
    • Consult with NCSA concerning optimal job submission
  • Generate October meeting TWiki page
  • Begin generating data sets for the production data - TWiki page

 

Task list / progress 02 Sep 2006

  • Pass 1 Analysis submitted to tungsten cluster
  • 1/2 of data were missing, jobs ended quickly

Revision 22006-09-03 - BrendanKiburg

Line: 1 to 1
 
META TOPICPARENT name="BrendanKiburg"
Line: 10 to 9
 
  • Pass 0 using data from "online" analysis
  • Pass 1 using data at NCSA
  • Data Transfer from NPL to NCSA
Added:
>
>

Task list / progress 02 Sep 2006

  • Pass 1 Analysis submitted to tungsten cluster
  • 1/2 of data were missing, jobs ended quickly
  • some problems getting all data on scratch
  • craft script to launch get mss script followed by bsub upon completiong of mss ; -- to preserve
 \ No newline at end of file

Revision 12006-08-30 - BrendanKiburg

Line: 1 to 1
Added:
>
>
META TOPICPARENT name="BrendanKiburg"

-- BrendanKiburg - 30 Aug 2006

Current Tasks

  • Database organization at NPL
  • Pass 0 using data from "online" analysis
  • Pass 1 using data at NCSA
  • Data Transfer from NPL to NCSA
 
This site is powered by the TWiki collaboration platform Powered by PerlCopyright © 2008-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback