Welcome to the STAR Fast Offline Quality Assurance and Offline DST Production QA Shift home page. In Run 12 the TPC, EMC-barrel, EMC-SMD-eta grid, EMC-SMD-phi grid, EMC End Cap, End Cap SMD, TOF-MRPCs and partial GEM and MTD detector subsystems will be active. Run 12 is devoted to 200 and 500 GeV p-p spin and hopefully some tests of U+U with 15 GeV Au-Au and 200 GeV Cu-Au. Our job is to monitor the validity of the data, calibrations and event reconstruction from the full experiment for the Fast Offline event reconstruction. From a practical standpoint it is only possible for Offline QA to detect fairly large problems and to monitor only a few percent of the total Fast Offline productions. But even this level of QA is beneficial. In the following I will attempt to answer the most common questions about the shift work. Following this is a link to the relevent pages which you will need to become familiar with prior to your first week on shift in Run 12. Please note that new QA tools introduced last year for testing are now available for routine use. These new features enable all offline data files to be checked using automated algorithms. Information on this is included in the documentation pages for Run 12.
No programming skills, such as C++, are required. All the tasks are web based "point-and-click" activities. However, you should have valid RCF and AFS accounts since you may need direct access to the histogram postscript files or reference QA histograms. Note that there are new login security measures in effect at the RCF.
General knowledge of the STAR detector, reconstruction methods and calibration issues are expected since the purpose of these shifts is to spot problems with the hardware and event reconstruction, or with the calibrations. Expert level knowledge is not required however.
All persons are required to be at BNL for their first week of Offline QA shift service. This is motivated by the fact that many unforeseen problems will very likely arise during these shifts and quick, real time help, which is more likely to be available at BNL, is essential to insure daily shift productivity.
Subsequent Offline QA shifts may be done from non-BNL sites provided adequate web access to the Auto QA system can be demonstrated from the remote site.
The offline QA shift may be done any time throughout the day but it is expected that a serious effort will require at least 8 hours a day.
There are no special QA training schools; the web based documentation is intended to fulfill such needs. But if you have further questions please send email to Lanny Ray, ray@physics.utexas.edu.
The Fast Offline QA shift work involves the following tasks, all of which are web based:
Fetch the latest set of Fast offline production jobs using the Auto QA browser and compare a standard set of histograms to a reference. Due to changing run conditions these reference histograms will not be ready until some reasonable data have been collected in the first week or so of the new run. With the new QA tools comparison with reference histograms is much more transparent than it is has been in the past.
Fill in a web based report form listing the jobs which have been QA-ed and giving detailed comments as needed. Summarize observed changes in hardware/software performance for the Fast Offline data and report directly to the three Shift Leaders and Period Coordinator on duty. Note that there is a "play" mode which can be used to practice using the QA Shift report form.
Provide a data/run quality summary based on the Online Run Log information and comments and the QA work.
Notify the appropriate people about suspected problems with the hardware, calibrations, production, or reconstruction software.
Check the Online-to-Offline data base migration using the "Database Migration Monitor" link on the first page of the QA browser after you login. When data are being taken the first several tables should appear in green font. If no data have been acquired for a day or so then all the tables should be in red. If there are any red fonts in the first several tables labelled "RunLog_onl" while data are being taken then this may indicate a problem and you should notify starqa-hn explicitly.
If you are not already subscribed to the 'starqa-hn' and 'starprod-hn' hypernews forums then please do so since this is the principal means for communicating with the production crew. See the HyperNews home page or select "hypernews" from the left panel on the STAR Home page. Follow the instructions for subscribing.
Lanny Ray and Gene van Buren will be available to assist and supervise the shift crew to insure that meaningful QA work is getting done.
Welcome aboard!
Lanny Ray, The University of Texas at Austin
January 30, 2012