[Halld-tagger] [EXTERNAL] photon beam working group

[Richard Jones]

Today would be the regular biweekly meeting of the Photon Beam working
group, but I think we can substitute the meeting today with the following
summary.

   - Mike Dugger continues to work through an issue with slow file
   transfers between Jefferson Lab and his server at ASU. Last week he and I
   sat for a work session to study this issue, and we found the issue is
   larger than just the specific endpoints in question. We found that transfer
   rates between centrally managed data transfer nodes at ASU and similar
   endpoints at UConn also showed the same bottleneck, limiting throughput to
   the range 5-10 MB / s using high-performance tools like globus online and
   perfsonar. At the same time, I was seeing rates > 500 MB/s from Jefferson
   Lab to UConn. This seems to be a recent phenomenon because Mike has done
   transfers in the past using commandline tools and seen much better rates.
   Mike is reaching out to ASU network experts to try and find out what is
   going on.


   - Andrew Schick has been occupied on a number of fronts, in preparation
   for the upcoming CPP run, but he continues to move forward with systematics
   studies of the analyzing power from his BH analysis. Since the last
   meeting, I generated new samples of muon pairs to complement the existing
   e+e- samples, and compare with the UMass generator for both 2H and 208Pb
   targets. These samples have been reduced (accept/reject, energy, angle
   cuts) and transferred to Jlab where Andrew has begun to process them
   through his simulation and analysis pipeline. Problems with the use of
   MCwrapper for these studies with the UConn generator were explained as due
   to the use of too old a version of hdgeant4. Andrew has yet to confirm that
   he is able to run his entire analysis through MCwrapper for both
   generators, by selecting a newer tagged version of hdgeant4, but I claim
   that this should be the case.


   - Early on in the SRC run, we have seen contradictory results between
   the projected photon beam focal spot at the collimator based on
   extrapolation from the harps and the measured transmission through the
   collimator. As soon as the run is over, we should have a discussion on what
   might be the cause of this, and how to improve our ability to determine the
   photon beam phase space distribution, so that we can reliably predict how
   changes in the beam tune are related to transmission through the collimator.


   - Plans are being finalized for a run in mid-December at the Saskatoon
   CLS to measure the crystal quality of the 3 new diamond samples obtained in
   2020 from Element Six.


-Richard Jones
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/halld-tagger/attachments/20211123/73036a99/attachment.html>