coherent WaveBurst Review Meeting 05 Mar 2009 09:00 Pacific / 12:00 Eastern
Agenda: Thursday 05 Mar 12:00 Eastern
- LFS segment lists page [web].
Dial-in number: 1 800 704 9896 International dial-in: +1 404 920 6472 Conference code: 5374 2349 #
AttendancePatrick (minutes), Marco, Gabriele, Romain, Igor, Sergey, Virginia, Michele, Stas.
LFS segment lists pageIgor walked us through this page.
- "Original" segments section: Some of these segments are "inclusive", others "exclusive". For example, the H1H2 segments contains times that V1 was also on. That's why the job segments list has a much smaller duration.
- "Final" segments section: These start from the single-IFO lists, without subtracting livetime from any other detectors.
- segs.py: single-IFO segments without subtracting other detectors. Only good for 4X network.
- For other networks see segs?.py. E.g., in segs1.py subtract L1, V1 livetime from both H1 and H2, then make H1H2 list. IY: Do it in this order (subtraction first, then H1H2 intersection) to get correct segments list for time lags as well.
- Comment: Codes lack comments entirely, have obscure names.
- Applying final segments: Set threshold, open box, then manually check any surviving triggers for overlap with a veto flag. IY: Done this way because there are usually only a handful of interesting zero-lag events so it is not worthwhile automating the procedure. Open box after final segments are defined, but before applying them. If any events (usually 0 or 1) then apply final segments. Automation only needed for simulations and background (many triggers).
- Cat 1 and 2 applied in trigger production.
- Action item:: Section below "Final category 1,2,3 ..." is not up-to-date. E.g., some links point to year-1 files. IY to update.
- Simulations: Use condor. Divide work into many separate jobs (millions of triggers).
- IY: May be the case that some times that are bad in the original cat 2 list could be good in the final cat 2 list. PJS: Only important for livetime. Could in principle overcount livetime if use only final cat 123 instead of union(original cat 123, final cat 123).
- Final cat 123 lists, including vetoes: Had email thread a few weeks ago (Yakushin, Cadonati, Blackburn, others) about what where the final segment lists and how to apply them (leading to "Final category 1,2,3" table). This was needed to get the final simulation trigger lists for Cadonati for the combined UL calculation.
- Action item: IY to generate found/missed injection tables for RG's follow-up study.