40m QIL Cryo_Lab CTN SUS_Lab TCS_Lab OMC_Lab CRIME_Lab FEA ENG_Labs OptContFac Mariner WBEEShop
  40m Log  Not logged in ELOG logo
Entry  Fri Sep 8 12:08:32 2017, Gabriele, Summary, LSC, Good reconstruction of PRMI degrees of freedom with deep learning full_real_data_2017_09_06b_100x4_4000_300_20_free_swinging_time_domain.png.pngfull_real_data_2017_09_06b_100x4_4000_300_20_free_swinging_free_swinging_histograms2d.png.pnggru_2017_08_13a_100x4_4000_300_20_free_swinging_simulation_reference.png.png
    Reply  Fri Oct 6 12:56:40 2017, gautam, Summary, LSC, RTCDS NN post_NN_test.png
       Reply  Tue Oct 17 17:53:25 2017, jamie, Summary, LSC, prep for tests of Gabriele's neural network cavity length reconstruction c1dnn.png
          Reply  Wed Oct 18 12:14:08 2017, jamie, Summary, LSC, prep for tests of Gabriele's neural network cavity length reconstruction 
             Reply  Thu Oct 19 15:42:03 2017, jamie, Summary, LSC, MICH/PRCL reconstruction neural network running on c1lsc NN.pngC1DNN_GDS.pngC1DNN_CPU_METER.png
                Reply  Tue Oct 24 20:14:21 2017, jamie, Summary, LSC, further testing of c1dnn integration; plugged in to DAQ NN.pngc1dnn_out.png
                   Reply  Wed Oct 25 09:32:14 2017, Gabriele, Summary, LSC, further testing of c1dnn integration; plugged in to DAQ 
                   Reply  Mon Nov 6 18:22:48 2017, jamie, Summary, LSC, current procedure for running c1dnn code 
                      Reply  Thu Nov 9 10:51:37 2017, gautam, Summary, LSC, current procedure for compiling and installing c1dnn code 
Message ID: 13400     Entry time: Tue Oct 24 20:14:21 2017     In reply to: 13395     Reply to this: 13401   13411
Author: jamie 
Type: Summary 
Category: LSC 
Subject: further testing of c1dnn integration; plugged in to DAQ 

In order to try to isolate CPU6 for the c1dnn neural network reconstruction model, I set the CPUAffinity in /etc/systemd/system.conf to "0" for the front end machines.  This sets the cpu affinity for the init process, so that init and all child processes are run on CPU0.  Unfortunately, this does not affect the kernel threads.  So after reboot all user space processes where on CPU0, but the kernel threads were still spread around.  Will continue trying to isolate the kernel as well...

In any event, this amount of isolation was still good enough to get the c1dnn user space model running fairly stably.  It's been running for the last hour without issue.

I added the c1dnn channel and testpoint files to the daqd master file, and restarted daqd_dc on fb1, so now the c1dnn channels and test points are available through dataviewer etc.  We were then able to observe the reconstructed signals:

We'll need to set the phase rotation of the demodulated RF PD signals (REFL11, REFL55, AS55, POP22) to match them with what the NN expects...

ELOG V3.1.3-