Page History
...
- single-core runs over 380*80 events
- 80-core job is running on 380 events per core
- use ds.break_after(380*80) in stead of termination of each. rank after. reaching 380 events.
Better start-stop time definition:
- skip 1-st event, because of time consumption overhead due to loading of calib constants
- start time is preserved for the 1-st rank gettein 2-nd event
- stop time for the last finishing rank
Summary
Total processing time , (sec, ) of 380*80 eventsRatio is fare from 80...=30400 events for 1 or 80 cores or 254*120=30480 for 120 cores
number of cores | old algorithm | ←Ratio t1/t80tN | new algorithm v2 | ←Ratio t1/t80tN | new algorithm v2 with loop | ←Ratio t1/t80tN | ||
---|---|---|---|---|---|---|---|---|
1better start-stop | 1110239.5 | 27.1 | 8425 | 48.0 51.5 | 1 | 8570 | 50.2 57.8 | 1 |
80 | 281 | 1 | 175.41 | 170.6 | 50.2 | |||
Ratio t1/t80 | 39.5 | 48.01 | ||||||
80 - better start-stop time def, times for 3 jobs: average: | 418.3 408.4 401.0 409.2 | 157.4 165.1 168.4 163.6 | 1 | 148.6 148.2 147.7 148.2 | ||||
Ratio t1/t80 better start-stop | 27.1 | 51.5 | 57.8 of 80 | |||||
120 times for 3 jobs: average: | 252.3 250.4 266.7 256.5 | 131.2 132.0 132.0 131.7 | 120.4 120.5 121.6 120.8 | |||||
Ratio t1/t120 | 43.3 | 64.0 | 70.9 of 120 |
2024-03-08 time consumption by the calib_jungfrau_v2 algorithm
Description
Code of calib_jungfrau_v2 is interlaces with timestamps. The list of timestamps is returned along with calibrated array.
Code Block | ||||
---|---|---|---|---|
| ||||
def calib_jungfrau_v2(det, evt, cmpars=(7,3,200,10), **kwa):
"""
v2 - improving performance, reduce time and memory consumption, use peds-offset constants
Returns calibrated jungfrau data
- gets constants
- gets raw data
- evaluates (code - pedestal - offset)
- applys common mode correction if turned on
- apply gain factor
Parameters
- det (psana.Detector) - Detector object
- evt (psana.Event) - Event object
- cmpars (tuple) - common mode parameters
- cmpars[0] - algorithm # 7-for jungfrau
- cmpars[1] - control bit-word 1-in rows, 2-in columns
- cmpars[2] - maximal applied correction
- **kwa - used here and passed to det.mask_v2 or det.mask_comb
- nda_raw - if not None, substitutes evt.raw()
- mbits - DEPRECATED parameter of the det.mask_comb(...)
- mask - user defined mask passed as optional parameter
"""
t00 = time()
src = det.source # - src (psana.Source) - Source object
nda_raw = kwa.get('nda_raw', None)
loop_segs = kwa.get('loop_segs', False)
arr = det.raw(evt) if nda_raw is None else nda_raw # shape:(<npanels>, 512, 1024) dtype:uint16
if arr is None: return None
t01 = time()
detname = string_from_source(det.source)
t02 = time()
odc = cache.detcache_for_detname(detname)
t03 = time()
first_entry = odc is None
if first_entry:
odc = cache.add_detcache(det, evt)
odc.cmps = det.common_mode(evt) if cmpars is None else cmpars
odc.mask = det.mask_total(evt, **kwa)
poff = odc.poff # 4d pedestals + offset shape:(3, 1, 512, 1024) dtype:float32
gfac = odc.gfac # 4d gain factors evaluated form gains
mask = odc.mask
outa = odc.outa
cmps = odc.cmps
t04 = time()
if first_entry:
logger.info('\n ====================== det.name: %s' % det.name\
+'\n detname from source: %s' % string_from_source(det.source)\
+info_ndarr(arr, '\n calib_jungfrau arr ')\
+info_ndarr(poff, '\n calib_jungfrau peds+off')\
+info_ndarr(gfac, '\n calib_jungfrau gfac')\
+info_ndarr(mask, '\n calib_jungfrau mask')\
+info_ndarr(outa, '\n calib_jungfrau outa')\
+info_ndarr(cmps, '\n calib_jungfrau common mode parameters ')
+'\n loop over segments: %s' % loop_segs)
if loop_segs:
nsegs = arr.shape[0]
shseg = arr.shape[-2:] # (512, 1024)
for i in range(nsegs):
arr1 = arr[i,:]
mask1 = mask[i,:]
gfac1 = gfac[:,i,:,:]
poff1 = poff[:,i,:,:]
arr1.shape = (1,) + shseg
mask1.shape = (1,) + shseg
gfac1.shape = (3,1,) + shseg
poff1.shape = (3,1,) + shseg
out1, times = calib_jungfrau_single_panel(arr1, gfac1, poff1, mask1, cmps)
outa[i,:] = out1[0,:]
resp = outa
else:
resp, times = calib_jungfrau_single_panel(arr, gfac, poff, mask, cmps)
t17 = time()
times = (t00, t01, t02, t03, t04) + times + (t17,)
return resp, np.array(times, dtype=np.float64)
def calib_jungfrau_single_panel(arr, gfac, poff, mask, cmps):
""" Arrays should have a single panel shape, example for 8-panel detector
arr: shape:(8, 512, 1024) size:4194304 dtype:uint16 [2906 2945 2813 2861 3093...]
poff: shape:(3, 8, 512, 1024) size:12582912 dtype:float32 [2922.283 2938.098 2827.207 2855.296 3080.415...]
gfac: shape:(3, 8, 512, 1024) size:12582912 dtype:float32 [0.02490437 0.02543429 0.02541406 0.02539831 0.02544083...]
mask: shape:(8, 512, 1024) size:4194304 dtype:uint8 [1 1 1 1 1...]
cmps: shape:(16,) size:16 dtype:float64 [ 7. 1. 100. 0. 0....]
"""
t05 = time()
# Define bool arrays of ranges
gr0 = arr < BW1 # 490 us
gr1 =(arr >= BW1) & (arr<BW2) # 714 us
gr2 = arr >= BW3 # 400 us
t06 = time()
factor = np.select((gr0, gr1, gr2), (gfac[0,:], gfac[1,:], gfac[2,:]), default=1) # 2msec
t07 = time()
pedoff = np.select((gr0, gr1, gr2), (poff[0,:], poff[1,:], poff[2,:]), default=0)
t08 = time()
# Subtract offset-corrected pedestals
arrf = np.array(arr & MSK, dtype=np.float32)
t09 = time()
arrf -= pedoff
t10 = time()
if cmps is not None:
mode, cormax = int(cmps[1]), cmps[2]
npixmin = cmps[3] if len(cmps)>3 else 10
if mode>0:
#arr1 = store.arr1
#grhg = np.select((gr0, gr1), (arr1, arr1), default=0)
logger.debug(info_ndarr(gr0, 'gain group0'))
logger.debug(info_ndarr(mask, 'mask'))
t0_sec_cm = time()
t11 = time()
gmask = np.bitwise_and(gr0, mask) if mask is not None else gr0
t12 = time()
#sh = (nsegs, 512, 1024)
hrows = 256 #512/2
for s in range(arrf.shape[0]):
t13 = time()
if mode & 4: # in banks: (512/2,1024/16) = (256,64) pixels # 100 ms
common_mode_2d_hsplit_nbanks(arrf[s,:hrows,:], mask=gmask[s,:hrows,:], nbanks=16, cormax=cormax, npix_min=npixmin)
common_mode_2d_hsplit_nbanks(arrf[s,hrows:,:], mask=gmask[s,hrows:,:], nbanks=16, cormax=cormax, npix_min=npixmin)
t14 = time()
if mode & 1: # in rows per bank: 1024/16 = 64 pixels # 275 ms
common_mode_rows_hsplit_nbanks(arrf[s,], mask=gmask[s,], nbanks=16, cormax=cormax, npix_min=npixmin)
t15 = time()
if mode & 2: # in cols per bank: 512/2 = 256 pixels # 290 ms
common_mode_cols(arrf[s,:hrows,:], mask=gmask[s,:hrows,:], cormax=cormax, npix_min=npixmin)
common_mode_cols(arrf[s,hrows:,:], mask=gmask[s,hrows:,:], cormax=cormax, npix_min=npixmin)
logger.debug('TIME: common-mode correction time = %.6f sec' % (time()-t0_sec_cm))
t16 = time()
arrf = arrf * factor if mask is None else arrf * factor * mask # gain correction
return arrf, (t05, t06, t07, t08, t09, t10, t11, t12, t13, t14, t15, t16)
|
Results
time differences between timestamps are shown in msec, as dt[i] = t[i]-t[i-1]
Code Block | ||||
---|---|---|---|---|
| ||||
ana-4.0.59-py3 [dubrovin@sdfmilan204:~/LCLS/con-py3]$ mpirun -n 1 python Detector/examples/test-scaling-mpi.py 6
execute command: ['Detector/examples/test-scaling-mpi.py', '6']
rank:000 cpu_num:000 size:01
[I] L0398: test-scaling-mpi.py
====================== det.name: CxiDs1.0:Jungfrau.0
detname from source: CxiDs1.0:Jungfrau.0
calib_jungfrau arr shape:(8, 512, 1024) size:4194304 dtype:uint16 [2906 2945 2813 2861 3093...]
calib_jungfrau peds+off shape:(3, 8, 512, 1024) size:12582912 dtype:float32 [2922.283 2938.098 2827.207 2855.296 3080.415...]
calib_jungfrau gfac shape:(3, 8, 512, 1024) size:12582912 dtype:float32 [0.02490437 0.02543429 0.02541406 0.02539831 0.02544083...]
calib_jungfrau mask shape:(8, 512, 1024) size:4194304 dtype:uint8 [1 1 1 1 1...]
calib_jungfrau outa shape:(8, 512, 1024) size:4194304 dtype:float32 [0. 0. 0. 0. 0....]ndarray from tuple:
calib_jungfrau common mode parameters shape:(4,) size:4 dtype:int64 [ 7 7 200 10]
loop over segments: False
rank:000 cpu_num:000 nevt:0000 time:10.075316
[I] L0607: test-scaling-mpi.py rank:000 job 2-nd evt time:1710175768.956751 saveed in file: figs/mpi-job-2nd-evt-time.txt
rank:000 cpu_num:000 nevt:0010 time:0.578157
rank:000 cpu_num:000 nevt:0020 time:0.541390
rank:000 cpu_num:000 nevt:0030 time:0.518435
rank:000 cpu_num:000 nevt:0040 time:0.575237
rank:000 cpu_num:000 nevt:0050 time:0.577683
rank:000 cpu_num:000 nevt:0060 time:0.580110
rank:000 cpu_num:000 nevt:0070 time:0.579042
rank:000 cpu_num:000 nevt:0080 time:0.577012
rank:000 cpu_num:000 nevt:0090 time:0.577242
rank:000 cpu_num:000 nevt:0100 time:0.575845
[I] L0640: test-scaling-mpi.py Summary for rank:000 job 2-nd evt time:1710175768.956751 time total (sec):64.189028
...
QStandardPaths: XDG_RUNTIME_DIR not set, defaulting to '/tmp/runtime-dubrovin'
hostname:sdfmilan204 rank:000 cpu:000 cmt:8p-v6 proc time (sec) mean: 0.5668 +/- 0.0125 rms: 0.0186 +/- 0.0088
plot_figs_v2
fnprefix: figs/fig-mpi-data-8p-v6-sdfmilan204-ncores01 title: sdfmilan204 rank 000 of 001 cpu_num 000
...
000 1709951462.307 0.896 0.019 0.008 0.003 0.001 1.485 5.141 4.795 1.873 1.081 0.389 0.712 552.819 11.156 31.449 35.934 3.994
001 1709951462.963 0.883 0.015 0.003 0.002 0.001 1.478 5.195 4.963 1.878 1.216 0.269 0.686 553.074 11.234 31.461 36.158 3.678
002 1709951463.620 0.736 0.013 0.003 0.001 0.000 1.325 5.175 4.901 1.848 1.186 0.268 0.700 556.101 10.969 31.402 35.841 2.868
003 1709951464.278 0.672 0.014 0.003 0.001 0.001 1.491 5.275 4.833 1.874 1.156 0.271 0.695 555.187 11.171 31.454 36.028 3.687
004 1709951464.935 0.672 0.015 0.003 0.001 0.001 1.407 5.255 4.980 1.860 1.188 0.265 0.681 556.062 11.451 31.451 35.880 4.575
005 1709951465.599 1.828 0.014 0.004 0.001 0.000 2.680 5.197 5.440 3.114 1.231 0.275 1.372 552.135 11.523 31.402 35.908 3.678
006 1709951466.259 0.675 0.016 0.003 0.001 0.001 1.370 5.214 4.860 1.832 1.193 0.278 0.691 552.219 11.224 31.394 36.250 3.223
007 1709951466.914 0.873 0.022 0.008 0.002 0.001 1.367 5.303 5.026 1.864 1.209 0.363 0.718 551.416 11.165 31.385 36.066 4.027
008 1709951467.569 0.842 0.014 0.003 0.001 0.001 1.389 5.209 4.796 1.854 1.119 0.264 0.699 551.404 11.150 31.370 36.021 2.912
009 1709951468.223 0.838 0.014 0.003 0.001 0.001 1.381 5.234 4.893 1.874 1.191 0.264 0.690 554.368 11.403 31.414 35.849 5.744
010 1709951468.886 1.976 0.014 0.003 0.001 0.001 3.203 5.134 5.291 3.050 1.211 0.310 0.857 556.652 11.083 31.411 35.851 2.866
011 1709951469.550 0.858 0.016 0.003 0.001 0.001 1.345 5.208 4.876 1.868 1.172 0.266 0.656 554.571 11.404 31.372 35.806 3.667
012 1709951470.207 0.782 0.015 0.003 0.001 0.000 1.321 5.140 4.890 1.849 1.178 0.276 0.671 552.930 11.304 31.376 35.908 3.159
013 1709951470.863 0.902 0.021 0.009 0.002 0.001 1.363 5.316 5.044 1.849 1.206 0.368 0.675 551.805 11.463 31.375 35.844 3.764
014 1709951471.518 0.880 0.014 0.003 0.001 0.001 1.352 5.144 4.876 1.856 1.161 0.269 0.696 551.738 11.189 31.328 36.213 2.871
015 1709951472.172 0.675 0.015 0.003 0.001 0.001 1.365 5.187 4.965 1.848 1.222 0.276 0.686 554.584 11.385 31.463 35.826 5.604
016 1709951472.835 1.795 0.014 0.003 0.001 0.001 3.175 5.209 5.346 3.035 1.211 0.305 1.116 553.065 11.606 31.393 35.818 2.867
017 1709951473.495 0.679 0.014 0.003 0.001 0.001 1.370 5.263 4.933 1.799 1.271 0.279 0.690 555.862 11.446 31.403 36.174 3.659
018 1709951474.154 0.793 0.013 0.003 0.001 0.000 1.343 5.124 4.980 1.845 1.195 0.269 0.686 553.584 11.318 31.364 35.820 2.865
019 1709951474.810 0.783 0.014 0.004 0.001 0.001 1.342 5.148 4.929 1.868 1.203 0.270 0.695 554.589 11.329 31.414 36.334 3.758
020 1709951475.469 0.759 0.021 0.010 0.002 0.001 1.384 5.295 5.002 1.847 1.193 0.370 0.680 552.323 10.995 31.373 36.190 3.232
021 1709951476.124 0.789 0.014 0.003 0.001 0.001 1.334 5.149 4.863 1.873 1.159 0.262 0.694 551.174 11.586 31.429 36.093 3.683
022 1709951476.778 0.678 0.017 0.003 0.001 0.001 1.357 5.206 4.942 1.861 1.139 0.263 0.607 552.973 11.487 31.450 36.206 2.861
023 1709951477.434 0.829 0.014 0.004 0.001 0.001 1.348 5.168 4.786 1.909 1.113 0.262 0.706 556.439 11.474 31.396 35.960 5.616
024 1709951478.098 1.973 0.014 0.003 0.001 0.001 3.199 5.158 5.316 3.041 1.199 0.301 0.712 552.752 11.263 31.464 35.887 2.860
025 1709951478.758 0.863 0.014 0.003 0.001 0.001 1.356 5.180 4.890 1.876 1.167 0.269 0.661 554.007 11.324 31.501 35.965 3.695
026 1709951479.415 0.732 0.014 0.004 0.001 0.001 1.335 5.185 4.794 1.839 1.143 0.273 0.693 556.165 11.355 31.592 35.691 2.909
027 1709951480.073 0.819 0.014 0.003 0.001 0.001 1.350 5.151 4.818 1.882 1.267 0.249 0.675 554.184 11.207 31.441 37.435 3.780
028 1709951480.732 0.796 0.021 0.009 0.002 0.001 1.341 5.367 5.031 1.855 1.180 0.367 0.683 552.998 11.480 31.440 36.413 3.181
029 1709951481.389 0.770 0.014 0.003 0.001 0.001 1.347 5.169 4.633 1.886 1.028 0.263 0.737 551.741 11.409 31.451 36.130 3.669
030 1709951482.044 0.798 0.014 0.002 0.001 0.001 1.347 5.197 4.551 1.838 0.988 0.286 0.742 552.585 11.286 31.398 36.214 2.938
031 1709951482.705 0.780 0.013 0.003 0.001 0.001 1.319 5.225 4.844 1.896 1.138 0.273 0.686 551.471 11.422 31.405 36.418 3.673
032 1709951483.360 0.744 0.015 0.003 0.001 0.001 1.364 5.174 5.049 1.848 1.197 0.271 0.690 555.717 11.567 31.450 36.004 4.597
033 1709951484.024 1.865 0.014 0.003 0.001 0.001 3.177 5.151 5.408 3.073 1.192 0.302 0.859 551.751 11.420 31.518 36.076 3.935
034 1709951484.684 0.763 0.013 0.003 0.001 0.001 1.366 5.156 4.923 1.836 1.180 0.270 0.654 553.017 11.105 31.419 39.409 2.934
035 1709951485.343 0.744 0.013 0.007 0.001 0.001 1.339 5.207 4.864 1.884 1.127 0.268 0.715 552.273 11.389 31.421 36.091 3.676
036 1709951485.998 0.783 0.015 0.002 0.001 0.000 1.353 5.147 4.905 1.826 1.190 0.268 0.676 553.218 11.577 31.411 36.138 2.858
037 1709951486.654 0.825 0.014 0.003 0.001 0.001 1.353 5.150 4.901 1.883 1.196 0.278 0.677 553.971 11.355 31.408 36.152 3.771
038 1709951487.312 0.751 0.023 0.009 0.002 0.001 1.380 5.332 5.054 1.862 1.162 0.368 0.659 552.360 11.561 31.421 35.896 3.161
039 1709951487.967 0.674 0.013 0.004 0.001 0.000 1.400 5.185 4.996 1.900 1.302 0.254 0.668 551.565 11.267 31.442 36.180 3.667
040 1709951488.622 0.834 0.014 0.003 0.001 0.001 1.353 5.120 4.946 1.814 1.219 0.290 0.681 552.905 11.643 31.436 35.873 2.932
041 1709951489.277 0.678 0.014 0.003 0.001 0.001 1.372 5.282 4.882 1.873 1.193 0.274 0.710 552.397 11.231 31.636 35.957 3.679
042 1709951489.933 0.833 0.014 0.003 0.001 0.001 1.350 5.175 4.920 1.853 1.181 0.268 0.666 556.858 11.618 31.441 36.003 2.847
043 1709951490.593 0.840 0.014 0.003 0.001 0.001 1.358 5.231 4.746 1.896 1.068 0.262 0.703 555.706 11.318 31.440 35.805 5.592
044 1709951491.256 1.952 0.014 0.003 0.002 0.005 3.188 5.193 5.317 3.031 1.187 0.306 0.767 552.667 11.568 31.327 35.856 2.917
045 1709951491.918 0.880 0.014 0.004 0.001 0.001 1.378 5.135 4.664 1.875 1.036 0.266 0.731 551.976 11.368 31.469 36.103 3.685
046 1709951492.573 0.842 0.014 0.003 0.001 0.001 1.362 5.220 4.928 1.821 1.194 0.281 0.673 553.601 11.359 31.435 35.888 2.857
047 1709951493.229 0.771 0.013 0.003 0.001 0.001 1.374 5.173 4.786 1.860 1.136 0.269 0.694 552.346 11.322 31.404 35.964 3.676
048 1709951493.884 0.738 0.015 0.003 0.001 0.001 1.352 5.152 4.954 1.839 1.420 0.273 0.638 552.455 11.343 31.405 35.867 2.919
049 1709951494.539 0.735 0.014 0.003 0.001 0.001 1.350 5.213 4.919 1.858 1.203 0.267 0.684 555.511 11.320 31.425 36.101 3.734
050 1709951495.199 0.792 0.021 0.010 0.002 0.001 1.350 5.295 4.967 1.875 1.159 0.362 0.695 555.633 11.365 31.427 35.871 3.206
051 1709951495.858 0.725 0.014 0.003 0.001 0.001 1.355 5.179 4.966 1.855 1.193 0.267 0.663 552.571 11.290 31.442 36.051 3.706
052 1709951496.513 0.732 0.014 0.003 0.001 0.001 1.351 5.148 4.894 1.835 1.204 0.268 0.691 553.878 11.168 31.355 35.997 2.867
053 1709951497.169 0.884 0.015 0.003 0.001 0.000 1.359 5.130 4.803 1.873 1.126 0.268 0.679 552.045 11.216 31.336 35.805 3.684
054 1709951497.824 0.789 0.014 0.004 0.001 0.001 1.342 5.153 4.969 1.838 1.198 0.277 0.669 552.754 11.168 31.354 35.906 2.863
055 1709951498.479 0.699 0.014 0.003 0.001 0.001 1.379 5.230 4.929 1.879 1.177 0.276 0.626 551.721 11.107 31.259 35.795 3.709
056 1709951499.133 0.754 0.013 0.003 0.001 0.001 1.346 5.166 4.931 1.846 1.201 0.276 0.641 553.559 11.420 31.302 36.263 4.534
057 1709951499.795 1.892 0.014 0.004 0.001 0.001 3.194 5.144 5.213 3.070 1.102 0.313 0.883 552.342 11.358 31.356 35.834 3.947
058 1709951500.455 0.805 0.013 0.003 0.002 0.001 1.321 5.237 4.941 1.871 1.189 0.272 0.689 553.453 11.569 31.407 35.818 2.918
059 1709951501.111 0.843 0.015 0.003 0.001 0.001 1.343 5.250 4.936 1.879 1.151 0.264 0.641 555.424 11.339 32.551 35.822 3.683
060 1709951501.775 0.839 0.013 0.003 0.001 0.001 1.346 5.242 4.981 1.867 1.173 0.274 0.686 553.788 11.635 31.384 35.812 2.853
061 1709951502.432 0.756 0.013 0.003 0.001 0.001 1.360 5.247 4.971 1.871 1.191 0.264 0.652 551.001 11.677 31.382 36.142 3.688
062 1709951503.087 0.831 0.015 0.004 0.001 0.001 1.340 5.271 4.916 1.837 1.315 0.244 0.620 551.165 11.508 31.385 35.914 2.868
063 1709951503.741 0.812 0.015 0.003 0.002 0.000 1.352 5.172 4.941 2.346 1.167 0.271 0.684 553.406 11.084 31.473 35.871 3.790
064 1709951504.398 0.852 0.020 0.010 0.002 0.001 1.330 5.253 4.844 1.855 1.074 0.364 0.710 551.497 11.369 31.465 35.813 3.201
065 1709951505.052 0.858 0.013 0.003 0.001 0.001 1.350 5.165 4.972 1.875 1.328 0.241 0.646 550.681 11.250 31.382 35.781 3.694
066 1709951505.706 0.758 0.015 0.003 0.001 0.001 1.349 5.157 4.908 1.836 1.194 0.274 0.675 551.458 11.241 32.197 36.151 2.866
067 1709951506.361 0.741 0.014 0.004 0.001 0.001 1.340 5.150 4.874 1.872 1.164 0.268 0.691 553.973 11.016 31.438 36.056 3.660
068 1709951507.018 0.807 0.015 0.003 0.001 0.001 1.329 5.279 4.790 1.841 1.065 0.263 0.660 553.581 11.266 31.376 35.834 2.949
069 1709951507.674 0.787 0.014 0.003 0.001 0.000 1.334 5.126 4.892 1.881 1.157 0.268 0.637 554.552 11.103 31.501 36.175 3.676
070 1709951508.331 0.734 0.014 0.003 0.001 0.001 1.357 5.187 4.900 1.847 1.182 0.290 0.680 552.080 11.193 31.472 35.965 2.924
071 1709951508.986 0.918 0.014 0.003 0.001 0.001 1.399 5.111 4.948 1.878 1.180 0.273 0.678 552.339 11.160 31.449 38.325 5.545
072 1709951509.648 1.889 0.016 0.003 0.001 0.001 3.175 5.211 5.360 3.035 1.554 0.392 1.726 552.022 11.255 31.424 36.042 2.904
073 1709951510.309 0.870 0.015 0.003 0.001 0.001 1.353 5.160 4.832 1.892 1.129 0.276 0.626 551.263 11.202 31.387 36.047 3.684
074 1709951510.963 0.827 0.013 0.003 0.001 0.001 1.338 5.287 4.996 1.806 1.233 0.277 0.686 552.124 11.490 31.376 35.786 2.860
075 1709951511.618 0.840 0.014 0.003 0.001 0.001 1.317 5.194 4.713 1.892 1.059 0.266 0.633 556.157 11.050 31.419 35.863 3.686
076 1709951512.276 0.793 0.013 0.002 0.001 0.001 1.337 5.218 4.824 1.847 1.163 0.273 0.627 553.038 11.441 31.399 36.186 2.922
077 1709951512.932 0.860 0.013 0.003 0.002 0.001 1.341 5.239 4.809 1.879 1.132 0.264 0.633 553.340 11.380 31.325 35.777 3.681
078 1709951513.588 0.806 0.013 0.003 0.001 0.001 1.337 5.150 4.858 1.859 1.164 0.271 0.625 552.434 11.174 31.342 35.875 2.861
079 1709951514.242 0.770 0.014 0.003 0.001 0.001 1.328 5.261 4.904 1.884 1.201 0.278 0.675 552.390 11.277 31.314 36.119 3.900
080 1709951514.898 0.819 0.013 0.003 0.001 0.001 1.318 5.177 4.884 1.860 1.179 0.265 0.685 552.436 11.178 31.398 36.157 3.146
081 1709951515.553 0.855 0.020 0.009 0.002 0.001 1.342 5.287 5.025 1.875 1.196 0.365 0.692 551.800 11.389 31.425 36.143 3.788
082 1709951516.208 0.848 0.013 0.003 0.001 0.001 1.342 5.195 4.701 1.853 1.053 0.286 0.722 552.828 11.556 31.430 36.204 2.869
083 1709951516.864 0.887 0.014 0.003 0.001 0.001 1.360 5.119 4.796 1.879 0.997 0.262 0.692 556.199 11.292 31.449 36.011 3.677
084 1709951517.523 0.803 0.013 0.003 0.001 0.001 1.320 5.174 4.836 1.863 1.142 0.268 0.697 554.992 11.651 31.435 36.045 2.867
085 1709951518.180 0.872 0.014 0.003 0.001 0.001 1.356 5.264 4.562 1.873 1.000 0.256 0.732 552.314 11.364 31.386 36.011 3.661
086 1709951518.835 0.752 0.014 0.003 0.001 0.000 1.347 5.221 4.898 1.843 1.160 0.260 0.624 552.631 11.247 31.341 35.831 2.864
087 1709951519.490 0.789 0.013 0.003 0.002 0.000 1.338 5.193 4.737 1.894 1.049 0.262 0.695 552.360 11.667 31.430 36.007 3.717
088 1709951520.145 0.778 0.015 0.003 0.001 0.001 1.374 5.197 4.914 1.870 1.151 0.273 0.693 551.128 11.247 31.365 36.042 2.918
089 1709951520.799 0.780 0.014 0.003 0.001 0.001 1.332 5.203 4.920 1.879 1.057 0.261 0.717 552.486 11.064 31.368 35.817 3.692
090 1709951521.454 0.809 0.014 0.003 0.001 0.001 1.336 5.174 4.876 1.850 1.136 0.291 0.638 552.596 11.177 31.420 35.792 4.503
091 1709951522.114 1.924 0.014 0.003 0.001 0.001 3.204 5.130 5.308 3.101 1.187 0.305 1.203 552.445 15.121 31.392 36.129 3.931
092 1709951522.779 0.776 0.014 0.003 0.001 0.001 1.350 5.167 4.806 1.852 1.130 0.266 0.650 552.923 11.227 31.414 35.848 2.917
093 1709951523.434 0.863 0.015 0.003 0.001 0.001 1.354 5.155 4.900 1.870 1.303 0.237 0.600 553.494 11.353 31.414 35.815 3.677
094 1709951524.090 0.699 0.015 0.003 0.001 0.001 1.344 5.153 4.925 1.835 1.213 0.278 0.668 550.517 11.584 31.425 35.820 2.866
095 1709951524.744 0.896 0.013 0.003 0.002 0.001 1.376 5.210 5.044 1.849 1.206 0.274 0.690 551.603 11.192 31.399 36.161 3.706
096 1709951525.399 0.849 0.013 0.003 0.001 0.001 1.333 5.183 4.539 1.844 0.974 0.251 0.723 552.818 11.143 31.449 35.790 2.860
097 1709951526.054 0.858 0.014 0.003 0.001 0.001 1.331 5.178 4.903 1.882 1.147 0.266 0.648 552.017 11.579 31.375 35.923 3.667
098 1709951526.709 0.861 0.013 0.003 0.001 0.001 1.337 5.155 4.961 1.824 1.199 0.274 0.676 551.453 11.124 31.403 35.890 2.867
099 1709951527.363 0.871 0.015 0.003 0.001 0.001 1.380 5.244 4.905 1.872 1.201 0.279 0.670 551.980 11.592 31.406 36.138 3.706
evt t00(sec) ms: t01 t02 t03 t04 t05 t06 t07 t08 t09 t10 t11 t12 t13 t14 t15 t16 t17
job #1 median: 0.808 0.014 0.003 0.001 0.001 1.352 5.186 4.907 1.868 1.180 0.271 0.686 552.710 11.327 31.412 35.965 3.667
other jobs
job #2 median: 0.820 0.013 0.003 0.001 0.001 1.324 5.312 4.984 1.796 1.146 0.231 0.681 551.307 11.269 31.356 35.863 2.910
job #3 median: 0.773 0.014 0.003 0.001 0.001 1.355 5.325 4.983 1.720 1.158 0.267 0.677 548.701 11.053 31.174 35.818 3.001
job #4 median: 0.664 0.015 0.003 0.002 0.001 1.408 5.339 5.003 1.726 1.194 0.272 0.685 548.775 11.112 31.168 35.857 3.215
job #5 median: 0.804 0.014 0.003 0.001 0.001 1.315 5.293 4.972 1.736 1.153 0.274 0.674 551.204 11.219 31.334 35.955 2.873 |
Code Block | ||||
---|---|---|---|---|
| ||||
ana-4.0.59-py3 [dubrovin@sdfmilan204:~/LCLS/con-py3]$ mpirun -n 1 python Detector/examples/test-scaling-mpi.py 7
rank:000 cpu_num:000 size:01
[I] L0398: test-scaling-mpi.py
====================== det.name: CxiDs1.0:Jungfrau.0
detname from source: CxiDs1.0:Jungfrau.0
calib_jungfrau arr shape:(8, 512, 1024) size:4194304 dtype:uint16 [2906 2945 2813 2861 3093...]
calib_jungfrau peds+off shape:(3, 8, 512, 1024) size:12582912 dtype:float32 [2922.283 2938.098 2827.207 2855.296 3080.415...]
calib_jungfrau gfac shape:(3, 8, 512, 1024) size:12582912 dtype:float32 [0.02490437 0.02543429 0.02541406 0.02539831 0.02544083...]
calib_jungfrau mask shape:(8, 512, 1024) size:4194304 dtype:uint8 [1 1 1 1 1...]
calib_jungfrau outa shape:(8, 512, 1024) size:4194304 dtype:float32 [0. 0. 0. 0. 0....]ndarray from tuple:
calib_jungfrau common mode parameters shape:(4,) size:4 dtype:int64 [ 7 7 200 10]
loop over segments: True
rank:000 cpu_num:000 nevt:0000 time:10.117775
[I] L0604: test-scaling-mpi.py rank:000 job 2-nd evt time:1710176925.177748 saveed in file: figs/mpi-job-2nd-evt-time.txt
rank:000 cpu_num:000 nevt:0010 time:0.586269
rank:000 cpu_num:000 nevt:0020 time:0.581187
rank:000 cpu_num:000 nevt:0030 time:0.579508
rank:000 cpu_num:000 nevt:0040 time:0.566666
rank:000 cpu_num:000 nevt:0050 time:0.580894
rank:000 cpu_num:000 nevt:0060 time:0.580464
rank:000 cpu_num:000 nevt:0070 time:0.579505
rank:000 cpu_num:000 nevt:0080 time:0.582394
rank:000 cpu_num:000 nevt:0090 time:0.579650
rank:000 cpu_num:000 nevt:0100 time:0.580779
[I] L0637: test-scaling-mpi.py Summary for rank:000 job 2-nd evt time:1710176925.177748 time total (sec):65.284561
rank:000 times: shape:(100,) size:100 dtype:float64
...
QStandardPaths: XDG_RUNTIME_DIR not set, defaulting to '/tmp/runtime-dubrovin'
hostname:sdfmilan204 rank:000 cpu:000 cmt:8p-v7 proc time (sec) mean: 0.5819 +/- 0.0051 rms: 0.0066 +/- 0.0036
plot_figs_v2
fnprefix: figs/fig-mpi-data-8p-v7-sdfmilan204-ncores01 title: sdfmilan204 rank 000 of 001 cpu_num 000
arrts[msec]: shape:(100, 18) size:1800 dtype:float64
...
000 1710176925.179 0.666 0.017 0.008 0.002 569.880 0.195 0.680 0.623 0.117 0.084 0.127 0.052 0.001 11.087 31.610 36.165 0.397
001 1710176925.835 0.638 0.018 0.003 0.002 571.524 0.194 0.707 0.637 0.119 0.083 0.132 0.054 0.001 11.149 31.636 35.907 0.396
002 1710176926.492 0.639 0.019 0.003 0.002 568.908 0.196 0.649 0.592 0.127 0.094 0.140 0.057 0.001 10.950 31.537 35.854 0.398
003 1710176927.147 0.633 0.019 0.003 0.002 568.563 0.201 0.649 0.601 0.117 0.083 0.130 0.056 0.001 10.769 31.551 36.225 0.401
004 1710176927.801 0.644 0.019 0.003 0.002 574.708 0.356 0.669 0.618 0.119 0.083 0.133 0.140 0.001 11.196 31.468 36.211 0.675
005 1710176928.465 1.722 0.018 0.003 0.001 568.464 0.195 0.674 0.592 0.117 0.083 0.132 0.056 0.001 11.315 31.503 35.888 0.401
006 1710176929.121 0.640 0.018 0.003 0.002 568.027 0.195 0.692 0.658 0.118 0.084 0.128 0.062 0.001 10.968 31.407 36.123 0.394
007 1710176929.775 0.638 0.023 0.012 0.002 569.959 0.192 0.603 0.678 0.118 0.083 0.130 0.056 0.001 10.919 31.392 35.939 0.395
008 1710176930.430 0.640 0.018 0.003 0.002 570.235 0.196 0.625 0.637 0.118 0.083 0.126 0.055 0.001 10.902 31.453 36.084 0.405
009 1710176931.086 0.637 0.018 0.003 0.001 572.507 0.353 0.658 0.603 0.118 0.084 0.133 0.140 0.002 11.013 31.432 37.417 0.677
010 1710176931.749 1.720 0.017 0.003 0.002 569.025 0.193 0.633 0.605 0.117 0.084 0.133 0.057 0.001 10.891 31.381 36.222 0.401
011 1710176932.405 0.635 0.017 0.003 0.001 567.340 0.194 0.634 0.612 0.125 0.083 0.127 0.056 0.001 11.194 31.417 35.845 0.399
012 1710176933.058 0.635 0.017 0.002 0.002 567.859 0.196 0.678 0.635 0.117 0.084 0.126 0.057 0.001 11.080 31.371 36.001 0.394
013 1710176933.712 0.636 0.024 0.013 0.002 566.197 0.194 0.657 0.603 0.117 0.083 0.133 0.058 0.001 11.233 31.371 36.247 0.403
014 1710176934.364 0.643 0.022 0.003 0.002 566.952 0.197 0.664 0.612 0.117 0.083 0.129 0.057 0.001 10.935 31.378 35.875 0.394
015 1710176935.016 0.635 0.018 0.003 0.002 569.362 0.510 0.626 0.861 0.127 0.083 0.154 0.136 0.001 11.003 31.729 36.294 0.671
016 1710176935.676 1.736 0.025 0.004 0.002 570.226 0.195 0.636 0.596 0.118 0.083 0.127 0.062 0.001 11.322 31.442 35.914 0.399
017 1710176936.337 0.637 0.018 0.002 0.002 567.089 0.192 0.650 0.592 0.300 0.083 0.129 0.057 0.001 11.218 31.405 35.914 0.393
018 1710176936.990 0.640 0.017 0.004 0.002 568.821 0.194 0.618 0.598 0.117 0.084 0.125 0.057 0.001 11.068 31.435 36.364 0.397
019 1710176937.645 0.642 0.017 0.003 0.002 567.523 0.196 0.665 0.636 0.119 0.083 0.127 0.057 0.001 11.117 31.384 35.998 0.395
020 1710176938.299 0.638 0.023 0.012 0.002 566.782 0.197 0.680 0.628 0.118 0.084 0.124 0.056 0.001 10.767 31.521 35.868 0.398
021 1710176938.951 0.636 0.018 0.003 0.001 565.609 0.193 0.610 0.607 0.118 0.084 0.126 0.057 0.002 11.339 31.319 36.009 0.394
022 1710176939.602 0.638 0.017 0.003 0.001 567.499 0.209 0.604 0.621 0.117 0.084 0.127 0.057 0.001 11.328 31.393 36.040 0.396
023 1710176940.256 0.635 0.019 0.004 0.002 573.105 0.353 0.679 0.609 0.118 0.091 0.127 0.141 0.001 11.096 31.372 36.344 0.681
024 1710176940.918 1.740 0.016 0.003 0.001 572.266 0.195 0.651 0.597 0.116 0.083 0.133 0.057 0.001 11.027 31.364 36.005 0.396
025 1710176941.577 0.631 0.018 0.003 0.001 567.538 0.192 0.631 0.591 0.125 0.084 0.125 0.057 0.001 11.003 31.337 35.894 0.396
026 1710176942.230 0.636 0.018 0.003 0.002 568.037 0.195 0.621 0.670 0.117 0.083 0.133 0.055 0.001 11.123 31.391 36.233 0.401
027 1710176942.884 0.636 0.018 0.003 0.001 567.989 0.193 0.635 0.677 0.127 0.084 0.126 0.056 0.001 10.962 31.390 35.895 0.398
028 1710176943.538 0.641 0.025 0.012 0.003 566.633 0.196 0.628 0.597 0.118 0.084 0.126 0.056 0.001 11.214 31.428 35.818 0.403
029 1710176944.190 0.635 0.017 0.003 0.002 565.870 0.192 0.648 0.610 0.117 0.089 0.124 0.057 0.001 11.144 31.362 35.860 0.395
030 1710176944.842 0.634 0.017 0.004 0.002 567.096 0.202 0.658 0.613 0.117 0.083 0.123 0.056 0.001 11.078 31.397 35.869 0.394
031 1710176945.494 0.629 0.018 0.003 0.001 566.641 0.195 0.649 0.607 0.118 0.083 0.133 0.057 0.001 11.119 31.346 36.274 0.401
032 1710176946.147 0.638 0.018 0.003 0.001 576.210 0.353 0.639 0.625 0.118 0.083 0.127 0.138 0.001 11.161 31.395 35.792 0.682
033 1710176946.812 1.721 0.018 0.003 0.002 567.965 0.192 0.614 0.622 0.117 0.083 0.131 0.057 0.001 11.197 31.366 35.955 0.394
034 1710176947.467 0.634 0.017 0.003 0.002 567.574 0.195 0.615 0.596 0.117 0.091 0.127 0.055 0.001 10.901 31.393 35.964 0.395
035 1710176948.120 0.635 0.017 0.003 0.001 540.194 0.195 0.662 0.509 0.100 0.070 0.107 0.050 0.001 10.365 28.301 34.414 0.394
036 1710176948.740 0.638 0.018 0.003 0.001 552.474 0.171 0.576 0.513 0.101 0.070 0.107 0.051 0.001 10.327 28.882 32.906 0.363
037 1710176949.372 0.607 0.017 0.003 0.002 547.162 0.194 0.664 0.616 0.118 0.083 0.136 0.065 0.001 10.803 31.229 35.949 0.402
038 1710176950.004 0.650 0.025 0.013 0.002 539.100 0.195 0.609 0.637 0.117 0.084 0.126 0.056 0.001 11.068 31.485 33.966 0.349
039 1710176950.627 0.606 0.016 0.002 0.002 554.036 0.169 0.542 0.518 0.105 0.071 0.120 0.065 0.001 10.410 29.793 32.797 0.351
040 1710176951.261 0.642 0.017 0.003 0.001 554.199 0.195 0.613 0.619 0.118 0.083 0.126 0.056 0.001 11.444 31.373 36.057 0.397
041 1710176951.902 0.631 0.024 0.003 0.002 570.335 0.194 0.641 0.595 0.118 0.083 0.135 0.057 0.001 10.968 31.413 35.979 0.393
042 1710176952.557 0.638 0.016 0.003 0.002 568.253 0.205 0.659 0.617 0.117 0.083 0.125 0.055 0.002 11.354 31.421 36.379 0.403
043 1710176953.212 0.636 0.017 0.003 0.002 570.688 0.349 0.694 0.612 0.127 0.084 0.125 0.137 0.001 10.998 31.364 35.868 0.677
044 1710176953.872 1.725 0.018 0.003 0.002 568.433 0.195 0.644 0.601 0.117 0.084 0.125 0.055 0.001 11.340 31.365 35.909 0.393
045 1710176954.528 0.631 0.018 0.003 0.001 566.346 0.194 0.655 0.607 0.117 0.083 0.128 0.061 0.001 11.143 31.315 35.884 0.393
046 1710176955.179 0.638 0.017 0.003 0.002 566.295 0.196 0.653 0.603 0.118 0.084 0.124 0.056 0.001 11.112 31.403 36.420 0.394
047 1710176955.832 0.632 0.017 0.004 0.002 565.378 0.193 0.613 0.597 0.117 0.088 0.128 0.056 0.001 11.151 31.322 36.042 0.398
048 1710176956.483 0.637 0.017 0.003 0.001 566.095 0.196 0.617 0.639 0.116 0.084 0.125 0.056 0.001 11.119 31.394 35.955 0.397
049 1710176957.135 0.633 0.018 0.003 0.002 567.317 0.193 0.696 0.609 0.118 0.089 0.128 0.057 0.001 11.030 31.297 35.939 0.395
050 1710176957.788 0.638 0.024 0.013 0.002 570.480 0.194 0.650 0.613 0.124 0.083 0.126 0.056 0.001 11.094 31.349 36.069 0.401
051 1710176958.444 0.634 0.017 0.003 0.001 567.719 0.195 0.686 0.591 0.117 0.083 0.132 0.057 0.001 11.028 31.395 35.928 0.394
052 1710176959.097 0.643 0.017 0.003 0.002 566.287 0.201 0.640 0.597 0.118 0.084 0.126 0.057 0.001 10.899 31.312 35.987 0.405
053 1710176959.749 0.637 0.017 0.003 0.002 567.780 0.196 0.647 0.611 0.118 0.084 0.131 0.057 0.002 10.977 31.435 35.997 0.395
054 1710176960.402 0.642 0.018 0.003 0.002 566.576 0.196 0.659 0.598 0.118 0.083 0.125 0.063 0.001 10.988 31.387 35.980 0.394
055 1710176961.055 0.627 0.017 0.003 0.002 566.366 0.194 0.659 0.613 0.118 0.084 0.135 0.056 0.001 10.914 31.310 36.169 0.396
056 1710176961.707 0.636 0.019 0.003 0.001 569.009 0.503 0.641 0.903 0.118 0.091 0.156 0.135 0.001 11.073 31.615 36.288 0.661
057 1710176962.366 1.732 0.019 0.002 0.002 570.110 0.196 0.658 0.647 0.118 0.083 0.134 0.057 0.001 11.127 35.682 35.929 0.401
058 1710176963.027 0.644 0.018 0.003 0.002 567.595 0.194 0.651 0.635 0.117 0.083 0.127 0.056 0.001 11.326 31.369 36.038 0.396
059 1710176963.681 0.628 0.019 0.002 0.002 566.851 0.194 0.634 0.625 0.118 0.084 0.128 0.055 0.001 11.123 31.468 35.948 0.393
060 1710176964.333 0.642 0.017 0.003 0.002 566.545 0.205 0.656 0.594 0.117 0.083 0.129 0.055 0.001 11.380 31.452 36.488 0.396
061 1710176964.987 0.630 0.018 0.003 0.002 567.141 0.231 0.631 0.615 0.117 0.084 0.125 0.055 0.001 11.449 31.284 35.898 0.393
062 1710176965.640 0.635 0.017 0.003 0.001 566.420 0.194 0.602 0.623 0.118 0.083 0.121 0.057 0.006 11.244 31.312 36.203 0.401
063 1710176966.292 0.636 0.018 0.003 0.001 566.424 0.196 0.646 0.604 0.119 0.084 0.127 0.056 0.001 10.810 31.245 36.966 0.411
064 1710176966.945 0.630 0.024 0.013 0.002 566.426 0.194 0.675 0.612 0.118 0.094 0.128 0.057 0.001 11.130 31.371 36.999 0.401
065 1710176967.598 0.634 0.018 0.003 0.002 569.700 0.204 0.633 0.603 0.117 0.083 0.125 0.056 0.002 11.036 31.356 36.108 0.394
066 1710176968.253 0.640 0.018 0.003 0.002 567.129 0.195 0.607 0.688 0.124 0.084 0.124 0.056 0.001 11.009 31.371 36.027 0.401
067 1710176968.906 0.633 0.018 0.003 0.002 565.612 0.203 0.671 0.599 0.116 0.083 0.125 0.056 0.001 10.705 31.379 36.182 0.396
068 1710176969.557 0.642 0.017 0.003 0.001 568.389 0.201 0.625 0.654 0.119 0.084 0.131 0.057 0.001 11.058 31.396 36.111 0.395
069 1710176970.212 0.628 0.018 0.002 0.002 566.199 0.195 0.646 0.595 0.117 0.083 0.126 0.061 0.001 10.833 31.327 36.125 0.395
070 1710176970.863 0.642 0.017 0.003 0.002 566.881 0.197 0.664 0.591 0.117 0.090 0.124 0.057 0.001 10.981 31.328 35.953 0.392
071 1710176971.516 0.628 0.017 0.003 0.002 566.241 0.194 0.659 0.633 0.118 0.084 0.123 0.062 0.001 10.926 31.365 36.098 0.394
072 1710176972.167 0.639 0.017 0.003 0.002 569.547 0.356 0.654 0.613 0.118 0.084 0.131 0.141 0.001 11.005 31.306 35.999 0.685
073 1710176972.826 1.742 0.017 0.003 0.001 568.023 0.194 0.629 0.602 0.117 0.084 0.123 0.056 0.002 10.927 31.313 35.976 0.394
074 1710176973.480 0.641 0.017 0.003 0.001 569.769 0.195 0.636 0.678 0.117 0.083 0.124 0.061 0.001 11.282 31.428 35.903 0.399
075 1710176974.136 0.636 0.017 0.003 0.002 566.796 0.194 0.653 0.627 0.117 0.084 0.124 0.073 0.002 10.784 31.387 36.141 0.393
076 1710176974.788 0.641 0.018 0.003 0.002 566.806 0.196 0.656 0.595 0.118 0.083 0.131 0.056 0.001 11.171 31.349 35.968 0.396
077 1710176975.441 0.630 0.023 0.003 0.002 567.456 0.194 0.654 0.627 0.125 0.085 0.126 0.056 0.001 11.124 31.371 36.198 0.397
078 1710176976.094 0.643 0.017 0.003 0.002 566.938 0.195 0.613 0.596 0.117 0.083 0.124 0.056 0.001 10.970 31.337 36.105 0.393
079 1710176976.747 0.631 0.017 0.003 0.002 568.831 0.194 0.609 0.648 0.118 0.083 0.125 0.055 0.001 11.077 31.336 35.977 0.397
080 1710176977.401 0.638 0.018 0.003 0.001 564.980 0.195 0.685 0.601 0.117 0.084 0.122 0.063 0.001 10.931 31.312 36.033 0.396
081 1710176978.052 0.630 0.018 0.003 0.002 566.748 0.194 0.676 0.622 0.118 0.083 0.125 0.057 0.002 11.122 31.409 36.197 0.394
082 1710176978.704 0.641 0.026 0.012 0.002 566.579 0.197 0.665 0.650 0.118 0.083 0.125 0.061 0.001 11.342 31.363 40.708 0.396
083 1710176979.362 0.630 0.017 0.003 0.002 567.053 0.203 0.644 0.600 0.118 0.083 0.126 0.056 0.001 10.994 31.350 36.342 0.396
084 1710176980.014 0.648 0.018 0.003 0.001 567.508 0.201 0.628 0.676 0.116 0.083 0.126 0.055 0.001 11.326 31.353 35.999 0.395
085 1710176980.668 0.633 0.016 0.003 0.002 566.638 0.194 0.640 0.597 0.118 0.083 0.127 0.054 0.001 11.118 31.345 35.914 0.400
086 1710176981.320 0.646 0.017 0.003 0.002 566.983 0.195 0.702 0.590 0.121 0.087 0.124 0.056 0.001 11.220 31.319 35.979 0.398
087 1710176981.973 0.634 0.017 0.003 0.001 567.545 0.195 0.602 0.595 0.118 0.084 0.124 0.057 0.001 11.417 31.365 36.243 0.405
088 1710176982.626 0.640 0.017 0.003 0.002 565.968 0.195 0.637 0.599 0.119 0.083 0.123 0.056 0.001 10.989 31.346 35.936 0.401
089 1710176983.277 0.628 0.015 0.003 0.002 566.326 0.196 0.645 0.602 0.118 0.083 0.126 0.057 0.001 10.849 31.537 36.117 0.393
090 1710176983.929 0.637 0.016 0.003 0.002 569.852 0.194 0.683 0.611 0.117 0.084 0.126 0.055 0.001 10.798 31.367 35.928 0.401
091 1710176984.584 0.631 0.017 0.002 0.001 571.955 0.352 0.689 0.596 0.117 0.084 0.131 0.139 0.001 11.187 31.335 36.047 0.682
092 1710176985.244 1.727 0.016 0.004 0.001 566.733 0.194 0.612 0.616 0.117 0.090 0.124 0.055 0.001 10.952 31.308 36.219 0.400
093 1710176985.898 0.628 0.018 0.004 0.002 566.060 0.200 0.602 0.627 0.117 0.083 0.125 0.056 0.002 11.088 32.352 36.037 0.394
094 1710176986.550 0.635 0.017 0.003 0.002 566.814 0.195 0.640 0.622 0.117 0.083 0.123 0.056 0.001 11.324 31.351 36.345 0.393
095 1710176987.202 0.629 0.017 0.003 0.002 566.586 0.194 0.652 0.672 0.117 0.083 0.131 0.057 0.001 10.958 31.435 36.180 0.401
096 1710176987.854 0.644 0.017 0.003 0.002 566.027 0.194 0.669 0.596 0.117 0.083 0.126 0.056 0.001 10.909 31.394 36.098 0.399
097 1710176988.506 0.636 0.017 0.003 0.001 566.741 0.192 0.636 0.651 0.118 0.084 0.125 0.056 0.002 11.258 31.340 35.949 0.392
098 1710176989.158 0.640 0.017 0.003 0.002 569.880 0.201 0.642 0.664 0.117 0.083 0.123 0.055 0.001 10.825 31.324 35.999 0.404
099 1710176989.813 0.628 0.016 0.003 0.001 566.983 0.202 0.645 0.596 0.118 0.083 0.124 0.056 0.001 11.322 31.347 36.264 0.399
evt t00(sec) ms: t01 t02 t03 t04 t05 t06 t07 t08 t09 t10 t11 t12 t13 t14 t15 t16 t17
job #1 median: 0.637 0.018 0.003 0.002 567.135 0.195 0.646 0.611 0.118 0.083 0.126 0.056 0.001 11.075 31.371 36.003 0.396 other
other jobs
job #2 median: 0.628 0.017 0.003 0.001 566.815 0.194 0.638 0.639 0.117 0.084 0.128 0.055 0.001 11.071 31.411 36.020 0.391
job #3 median: 0.628 0.016 0.003 0.001 567.021 0.194 0.689 0.601 0.118 0.084 0.125 0.055 0.001 11.060 31.398 35.948 0.395
job #4 median: 0.638 0.016 0.003 0.002 567.018 0.195 0.627 0.629 0.117 0.083 0.127 0.057 0.001 11.080 31.404 35.974 0.396
job #5 median: 0.805 0.014 0.003 0.001 567.475 0.212 0.654 0.624 0.120 0.085 0.125 0.057 0.001 11.093 31.389 35.989 0.400 |
Summary
w/o loops over panels | with loops over panels | |||||
---|---|---|---|---|---|---|
dt | meaning | array size in (512,1024) | msec | array size in (512,1024) | msec | comment |
01 | get det.raw(evt) | 8 | 0.664 | 8 | 0.638 | |
02 | get detector name from source | 0.015 | 0.016 | |||
03 | access cache object for detname | 0.003 | 0.003 | |||
04 | get peds, gfac, mask, out, cmps | 3x8 | 0.002 | 3x8 | 0.002 | |
05 | single panel begin processing | begin processing entire array | 0.001 | begin processing last panel | 567 | meaningless |
06 | make gain range indices, gr0, 1, 2 | 8 | 1.408 | 1 | 0.195 | |
07 | np.select for gain factor | 8 | 5.34 | 1 | 0.627 | |
08 | np.select for gain peds+offset | 8 | 5.00 | 1 | 0.629 | |
09 | apply mask for data bits, arr & MSK | 8 | 1.73 | 1 | 0.117 | |
10 | subtract pedestals | 8 | 1.19 | 1 | 0.083 | |
11 | massaging common mode parameters | 0.272 | 0.127 | |||
12 | apply pixel mask for gr0 | 8 | 0.674 | 1 | 0.057 | |
13 | begin loop over panels for CMC | time to work with last panel | 549 | time to work with current panel | 0.001 | meaningless |
14 | CMC in banks: (512/2,1024/16) = (256,64) | 1 | 11.2 | 1 | 11.1 | CMC always loop over panels |
15 | CMC in rows per bank: 1024/16 = 64 pixels | 1 | 31.3 | 1 | 31.4 | |
16 | CMC in cols per bank: 512/2 = 256 pixels | 1 | 35.9 | 1 | 36.0 | |
17 | Apply gain correction and mask | 8 | 2.87 | 1 | 0.40 | |
Total time per event | 8 | 576 | 8 | 582 |
- where CMC stands for common mode correction
References
- Scaling behavior of psana1 - Part 1 - det.calib method in multicore processing with mpi
- Scaling behavior of psana1 - Part 2 - test with command perf stat
...