Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Error: unable to load U velocity vector components #1161

Open
jabowery opened this issue Jun 29, 2021 · 18 comments
Open

Error: unable to load U velocity vector components #1161

jabowery opened this issue Jun 29, 2021 · 18 comments

Comments

@jabowery
Copy link

See this screencast.

The video shows, first, a silent failure to load a scalar slice and then a reported (see subject above) failure to load a vector slice.

I was able to load these slices and vectors yesterday.

Two things have changed that might have caused this:

  1. A lot more data being generated overnight.
  2. Changing my location from local -- on the machine's console -- (yesterday) to remote via x2go X-windows access today.
@rmcdermo
Copy link
Contributor

I would suspect that "more data" is the likely issue. 3D slice files get very large. Does it help if you "unload all" first?

Glenn is out this week, so not sure if he will respond very readily.

@gforney
Copy link
Contributor

gforney commented Jun 29, 2021

assuming x2go is not the issue, try opening the Files/Data/Coloring dialog box, then open the "Time" tab. reduce the data that is loaded by selecting the "frame skip" checkbox and setting some number, say 5. how big re these 3d slice files? how much memory on the system where you are running smokeview?

@jabowery
Copy link
Author

jabowery commented Jun 29, 2021 •
edited
Loading

I'm at the machine console (local) with a frame skip of 5 and it loads only about 20 seconds of the >200seconds of data. Skip of 10 loads slightly less (real) time. It has 64G RAM. Setting the start time to 200 seconds doesn't affect things, nor does setting start time to 0 with offset of 200 seconds. Using min and max I was able to get the smoke data to load for the time interval (200 to 210 secs with step 5) but loading the slice for a basis vector gave the "Error: unable to load U velocity vector components in...".

(base) jabowery@ML:~/dev/BalloonVortex$ smokeview square.smv

Smokeview   - Jun  1 2021

Revision         : SMV6.7.16-0-g485e0cd19-release
Revision Date    : Fri May 28 08:24:03 2021 -0400
Compilation Date : Jun  1 2021 12:05:12
Compiler         : Intel C/C++ 2021.1
Checksum(SHA1)   : efb84c04d5949328c0b9f16952a9487e3d45ab5c
Platform         : LINUX64

Smokeview        : /home/jabowery/FDS/FDS6/smvbin/smokeview
Smokezip         : /home/jabowery/FDS/FDS6/smvbin/smokezip
Texture directory: /home/jabowery/FDS/FDS6/smvbin/textures
Scratch directory: /home/jabowery/.smokeview/
    smokeview.ini: /home/jabowery/.smokeview//smokeview.ini

initializing Glut
complete
initializing OpenGL
  GPU shaders
    3D smoke(new) loaded
    3D smoke loaded
    volume smoke loaded
    3D slice loaded
    zone smoke loaded
complete

reading  square.smv
reading /home/jabowery/FDS/FDS6/smvbin/objects.svo - complete

  pass 1
  pass 2
  pass 3
  pass 4
  wrapping up
complete

clipping off

Startup time: 0.4 s

reading /home/jabowery/FDS/FDS6/smvbin/smokeview.ini - complete

research mode on, using global bounds

Loading square_1_3.sf(V-VEL) - 50.8 MB/0.5 s
Defining 3d slice textures for Mesh01 ...completed
global: V-VELOCITY(cell centered) m/s, min=    -25.0922, max=     25.1124

Loading square_1_2.sf(U-VEL) - 50.8 MB/0.4 s
Defining 3d slice textures for Mesh01 ...completed
Loading square_1_3.sf(V-VEL) - 50.8 MB/0.5 s
Defining 3d slice textures for Mesh01 ...completed
Loading square_1_4.sf(W-VEL) - 50.8 MB/0.5 s
Defining 3d slice textures for Mesh01 ...completed
Loading square_1_1.sf(vel) - 50.8 MB/0.5 s
Defining 3d slice textures for Mesh01 ...completed
global: VELOCITY(cell centered) m/s, min=         0.0, max=     24.5894

Loading square_1_1.sf(vel) - 58.1 MB/0.5 s
Defining 3d slice textures for Mesh01 ...completed
global: VELOCITY(cell centered) m/s, min=         0.0, max=     24.5894

Loading square_1_1.sf(vel) - 58.1 MB/0.5 s
Defining 3d slice textures for Mesh01 ...completed
global: VELOCITY(cell centered) m/s, min=         0.0, max=     24.5894

@rmcdermo
Copy link
Contributor

rmcdermo commented Jun 29, 2021 •
edited
Loading

All I can suggest is to look at the size of the files and make sure they are consistent with how much data you expect. For example, square_1_1.sf(vel) should contain NX x NY x NZ x NFRAMES x (4 Bytes) + a few more Bytes for headers info. There should be one file per mesh. Are we in the ball park with the file sizes?

@jabowery
Copy link
Author

jabowery commented Jun 29, 2021 •
edited
Loading

I just figured out that the number of frames is going to be the number of seconds in the simulation divided by "DT_SL3D" parameter the, so the following looks about correct:

96x96x19x4 = 7077890

10118428970 / 7077890 = 1429.58

So 1429 should be the approximate number of frames, right?

&DUMP DT_RESTART=60.0, DT_SL3D=0.25/

and

Time Step: 25300, Simulation Time: 353.43 s

so

353.43 / 0.25 = 1413.72 which is in the ballpark at least.

PS: There is only one mesh in this model and the files of length 2749034 are from a prior run which had multiple meshes the first of which was small.

(base) jabowery@ML:~/dev/BalloonVortex$ ls -alt square_1_*.sf
-rw-rw-r-- 1 jabowery jabowery 10118428970 Jun 29 14:23 square_1_8.sf
-rw-rw-r-- 1 jabowery jabowery 10118428970 Jun 29 14:23 square_1_7.sf
-rw-rw-r-- 1 jabowery jabowery 10118428970 Jun 29 14:23 square_1_6.sf
-rw-rw-r-- 1 jabowery jabowery 10118428970 Jun 29 14:23 square_1_5.sf
-rw-rw-r-- 1 jabowery jabowery 10118428970 Jun 29 14:23 square_1_4.sf
-rw-rw-r-- 1 jabowery jabowery 10118428970 Jun 29 14:23 square_1_3.sf
-rw-rw-r-- 1 jabowery jabowery 10118428970 Jun 29 14:23 square_1_2.sf
-rw-rw-r-- 1 jabowery jabowery 10118428970 Jun 29 14:23 square_1_1.sf
-rw-rw-r-- 1 jabowery jabowery     2749034 Jun 28 19:23 square_1_27.sf
-rw-rw-r-- 1 jabowery jabowery     2749034 Jun 28 19:23 square_1_28.sf
-rw-rw-r-- 1 jabowery jabowery     2749034 Jun 28 19:23 square_1_29.sf
-rw-rw-r-- 1 jabowery jabowery     2749034 Jun 28 19:23 square_1_19.sf
-rw-rw-r-- 1 jabowery jabowery     2749034 Jun 28 19:23 square_1_20.sf
-rw-rw-r-- 1 jabowery jabowery     2749034 Jun 28 19:23 square_1_21.sf
-rw-rw-r-- 1 jabowery jabowery     2749034 Jun 28 19:23 square_1_22.sf
-rw-rw-r-- 1 jabowery jabowery     2749034 Jun 28 19:23 square_1_23.sf
-rw-rw-r-- 1 jabowery jabowery     2749034 Jun 28 19:23 square_1_24.sf
-rw-rw-r-- 1 jabowery jabowery     2749034 Jun 28 19:23 square_1_25.sf
-rw-rw-r-- 1 jabowery jabowery     2749034 Jun 28 19:23 square_1_26.sf
-rw-rw-r-- 1 jabowery jabowery     2749034 Jun 28 19:23 square_1_11.sf
-rw-rw-r-- 1 jabowery jabowery     2749034 Jun 28 19:23 square_1_12.sf
-rw-rw-r-- 1 jabowery jabowery     2749034 Jun 28 19:23 square_1_13.sf
-rw-rw-r-- 1 jabowery jabowery     2749034 Jun 28 19:23 square_1_14.sf
-rw-rw-r-- 1 jabowery jabowery     2749034 Jun 28 19:23 square_1_15.sf
-rw-rw-r-- 1 jabowery jabowery     2749034 Jun 28 19:23 square_1_16.sf
-rw-rw-r-- 1 jabowery jabowery     2749034 Jun 28 19:23 square_1_17.sf
-rw-rw-r-- 1 jabowery jabowery     2749034 Jun 28 19:23 square_1_18.sf
-rw-rw-r-- 1 jabowery jabowery     2749034 Jun 28 19:23 square_1_10.sf
-rw-rw-r-- 1 jabowery jabowery     2749034 Jun 28 19:23 square_1_9.sf

@rmcdermo
Copy link
Contributor

If you don't change NFRAMES on DUMP in the input file, then the default is 1000 for the entire simulation. That means FDS will write one 3D slice every DT_SLCF = T_END/1000 seconds. For each STIME, the write looks like this:

WRITE(LU_SLCF(IQ,NM)) STIME
WRITE(LU_SLCF(IQ,NM)) (((QQ(I,J,K,1),I=I1,I2),J=J1,J2),K=K1,K2)

where QQ is an array of Four Byte Reals.

Please include your input file. This will help us guess what might be going on.

@jabowery
Copy link
Author

Here is square.fds:

&HEAD CHID='square'/
&TIME T_END=480.0/
&DUMP DT_RESTART=60.0, DT_SL3D=0.25/

&MESH ID='Mesh01', IJK=96,96,192, XB=-24.0,24.0, -24.0,24.0, 0,96/

&PART ID='TracerR',
      MASSLESS=.TRUE.,
      MONODISPERSE=.TRUE.,
      COLOR='RED',
      AGE=60.0/

&SURF ID='ParticleR',
      COLOR='RED',
      TMP_FRONT=40.0,
      EMISSIVITY=0.0,
      VEL=-1.0,
      PART_ID='TracerR'/

&VENT ID='VentR', SURF_ID='ParticleR', XB=-1,0, -5,-5, 0.0,0.0/

&PART ID='TracerB',
      MASSLESS=.TRUE.,
      MONODISPERSE=.TRUE.,
      COLOR='BLUE',
      AGE=60.0/

&SURF ID='ParticleB',
      COLOR='BLUE',
      TMP_FRONT=40.0,
      EMISSIVITY=0.0,
      VEL=-1.0,
      PART_ID='TracerB'/

&VENT ID='VentB', SURF_ID='ParticleB', XB=1,0, 5,5, 0.0,0.0/

&PART ID='Propane',
      SPEC_ID='PROPANE',
      DIAMETER=1000.0,
      MONODISPERSE=.TRUE.,
      RGB=240,5,6,
      AGE=60.0/

&REAC ID='Reaction1',
      FUEL='PROPANE',
      SOOT_YIELD=0.01/

&PROP ID='Fuel Spray',
      PART_ID='Propane',
      OFFSET=0.25,
      FLOW_RATE=1.21,
      FLOW_RAMP='FUEL RAMP',
      PARTICLE_VELOCITY=20.0,
      SPRAY_ANGLE=5.0,5.0/

&RAMP ID='FUEL RAMP', T=0.0, F=0.0/
&RAMP ID='FUEL RAMP', T=1.0, F=1.0/
&RAMP ID='FUEL RAMP', T=480.0, F=1.0/

&DEVC ID='NOZZLE', PROP_ID='Fuel Spray', XYZ=0,0,2.5, ORIENTATION=0.0,0.0,1.0, QUANTITY='TIME', SETPOINT=0.0/
&DEVC ID='FEDco', QUANTITY='FED', XYZ=0,0,1.5/
&DEVC ID='Temp', QUANTITY='TEMPERATURE', XYZ=0,0,1.5/
&DEVC ID='Rad', QUANTITY='RADIATIVE HEAT FLUX GAS', XYZ=0,0,1.5, ORIENTATION=0.0,0.0,1.0/

&VENT ID='Boundry Vent XMAX', SURF_ID='OPEN', XB=24.0,24.0, -24.0,24.0, 0.0,96.0, COLOR='INVISIBLE'/ 
&VENT ID='Boundry Vent XMIN', SURF_ID='OPEN', XB=-24.0,-24.0,   -24.0,24.0, 0.0,96.0, COLOR='INVISIBLE'/ 
&VENT ID='Boundry Vent YMAX', SURF_ID='OPEN', XB=-24.0,24.0, 24.0,24.0, 0.0,96.0, COLOR='INVISIBLE'/ 
&VENT ID='Boundry Vent YMIN', SURF_ID='OPEN', XB=-24.0,24.0,  -24.0,-24.0, 0.0,96.0, COLOR='INVISIBLE'/ 
&VENT ID='Boundry Vent ZMAX', SURF_ID='OPEN', XB=-24.0,24.0,  -24.0,24.0, 96.0,96.0, COLOR='INVISIBLE'/ 


-x
&OBST ID='Fan', XB=10.00,11.00,8.00,9.00,0.0,1.0/
&VENT ID='IN00', SURF_ID='HVAC', XB=11.00,11.00,8.00,9.00,0.0,1.0, COLOR='RED'/
&VENT ID='OUT00', SURF_ID='HVAC', XB=10.00,10.00,8.00,9.00,0.0,1.0, COLOR='BLUE', UVW=1,0,.18/
&OBST ID='Tarp00', XB=11.70,-10.70, 12.00,13.00, 0.0,20.0/

+x
&OBST ID='Fan', XB=-10.00,-11.00,-8.00,-9.00,0.0,1.0/
&VENT ID='IN01', SURF_ID='HVAC', XB=-11.00,-11.00,-8.00,-9.00,0.0,1.0, COLOR='RED'/
&VENT ID='OUT01', SURF_ID='HVAC',XB=-10.00,-10.00,-8.00,-9.00,0.0,1.0, COLOR='BLUE', UVW=-1,0,.18/
&OBST ID='Tarp01', XB=-11.70,10.70, -12.00,-13.00, 0.0,20.0/

-y
&OBST ID='Fan', XB=-8.00,-9.00,10.00,11.00,0.0,1.0/
&VENT ID='IN02', SURF_ID='HVAC', XB=-8.00,-9.00,11.00,11.00,0.0,1.0, COLOR='RED'/
&VENT ID='OUT02', SURF_ID='HVAC', XB=-8.00,-9.00,10.00,10.00,0.0,1.0, COLOR='BLUE', UVW=0,1,.18/
&OBST ID='Tarp00', XB=12.00,13.00, 11.70,-10.70, 0.0,20.0/

+y
&OBST ID='Fan', XB=8.00,9.00,-10.00,-11.00,0.0,1.0/
&VENT ID='IN03', SURF_ID='HVAC', XB=8.00,9.00,-11.00,-11.00,0.0,1.0, COLOR='RED'/
&VENT ID='OUT03', SURF_ID='HVAC',XB=8.00,9.00,-10.00,-10.00,0.0,1.0, COLOR='BLUE', UVW=0,-1,.18/
&OBST ID='Tarp01', XB=-12.00,-13.00, -11.70,10.70, 0.0,20.0/

&HVAC ID='IN00', TYPE_ID='NODE', DUCT_ID='JET FAN', VENT_ID='IN00'/
&HVAC ID='OUT00', TYPE_ID='NODE', DUCT_ID='JET FAN', VENT_ID='OUT00'/
&HVAC ID='JET FAN', TYPE_ID='DUCT', AREA=1.0, PERIMETER=4.0, VOLUME_FLOW=25.0, RAMP_ID='JET FAN RAMP', NODE_ID='IN00','OUT00'/
&HVAC ID='IN01', TYPE_ID='NODE', DUCT_ID='JET FAN01', VENT_ID='IN01'/
&HVAC ID='OUT01', TYPE_ID='NODE', DUCT_ID='JET FAN01', VENT_ID='OUT01'/
&HVAC ID='JET FAN01', TYPE_ID='DUCT', AREA=1.0, PERIMETER=4.0, VOLUME_FLOW=25.0, RAMP_ID='JET FAN RAMP', NODE_ID='IN01','OUT01'/
&HVAC ID='JET FAN02', TYPE_ID='DUCT', AREA=1.0, PERIMETER=4.0, VOLUME_FLOW=25.0, RAMP_ID='JET FAN RAMP', NODE_ID='IN02','OUT02'/
&HVAC ID='IN02', TYPE_ID='NODE', DUCT_ID='JET FAN02', VENT_ID='IN02'/
&HVAC ID='OUT02', TYPE_ID='NODE', DUCT_ID='JET FAN02', VENT_ID='OUT02'/
&HVAC ID='JET FAN03', TYPE_ID='DUCT', AREA=1.0, PERIMETER=4.0, VOLUME_FLOW=25.0, RAMP_ID='JET FAN RAMP', NODE_ID='IN03','OUT03'/
&HVAC ID='IN03', TYPE_ID='NODE', DUCT_ID='JET FAN03', VENT_ID='IN03'/
&HVAC ID='OUT03', TYPE_ID='NODE', DUCT_ID='JET FAN03', VENT_ID='OUT03'/

&RAMP ID='JET FAN RAMP', T=0.0, F=0.0/
&RAMP ID='JET FAN RAMP', T=1.0, F=1.0/
&RAMP ID='JET FAN RAMP', T=480.0, F=1.0/


&SLCF QUANTITY='VELOCITY', VECTOR=.TRUE., CELL_CENTERED=.TRUE., XB=-24.0,24.0, -24.0,24.0,0.0,96.0/
&SLCF QUANTITY='TEMPERATURE', CELL_CENTERED=.TRUE., XB=-24.0,24.0,-24.0,24.0,0.0,96.0/
&SLCF QUANTITY='VORTICITY Z', CELL_CENTERED=.TRUE., XB=-24.0,24.0,-24.0,24.0,0.0,96.0/
&SLCF QUANTITY='HRRPUV', CELL_CENTERED=.TRUE., XB=-24.0,24.0,-24.0,24.0,0.0,96.0/
&SLCF QUANTITY='INTEGRATED INTENSITY', CELL_CENTERED=.TRUE., XB=-24.0,24.0,-24.0,24.0,0.0,96.0/

&TAIL/

@gforney
Copy link
Contributor

gforney commented Jun 29, 2021 via email

@gforney
Copy link
Contributor

gforney commented Jun 29, 2021 via email

@jabowery
Copy link
Author

My system has 64G RAM.

@jabowery
Copy link
Author

There is only one mesh.

@rmcdermo
Copy link
Contributor

rmcdermo commented Jun 29, 2021 via email

@gforney
Copy link
Contributor

gforney commented Jun 29, 2021 via email

@jabowery
Copy link
Author

Looks like I have about 700G free on the device where the directory is located.

(base) jabowery@ML:~/dev/BalloonVortex$ df
Filesystem                                                                            1K-blocks       Used Available Use% Mounted on
udev                                                                                   32866312          0  32866312   0% /dev
tmpfs                                                                                   6579568       2980   6576588   1% /run
/dev/mapper/ubuntu--vg-root                                                          1920436504 1129135144 693678732  62% /
...

I'm getting another

@jabowery
Copy link
Author

Of some pertinence, I killed off the fds process and now I can get about 60 seconds loaded into smokeview.

@jabowery
Copy link
Author

jabowery commented Jul 16, 2021 •
edited
Loading

Ran into a similar problem on another simulation that renders only about 10seconds out of over 100seconds.

See this screencast.

As before, if I kill off fds I can render a few seconds more.

Particles render the whole timeline.

Here's the disk usage:

~/dev/BalloonVortex$ du -c coriolis1burn*
21084	coriolis1burn_1_1.s3d
8	coriolis1burn_1_1.s3d.sz
14293896	coriolis1burn_1_1.sf
28	coriolis1burn_1_1.sf.bnd
8928	coriolis1burn_1_2.s3d
8	coriolis1burn_1_2.s3d.sz
14293896	coriolis1burn_1_2.sf
28	coriolis1burn_1_2.sf.bnd
10784	coriolis1burn_1_3.s3d
8	coriolis1burn_1_3.s3d.sz
14293892	coriolis1burn_1_3.sf
28	coriolis1burn_1_3.sf.bnd
21468	coriolis1burn_1_4.s3d
8	coriolis1burn_1_4.s3d.sz
14293892	coriolis1burn_1_4.sf
28	coriolis1burn_1_4.sf.bnd
14293896	coriolis1burn_1_5.sf
28	coriolis1burn_1_5.sf.bnd
63348	coriolis1burn_1.prt5
16	coriolis1burn_1.prt5.bnd
4	coriolis1burn_1.prt5.sz
2596276	coriolis1burn_1.restart
1992	coriolis1burn_2_1.s3d
4	coriolis1burn_2_1.s3d.sz
3145220	coriolis1burn_2_1.sf
8	coriolis1burn_2_1.sf.bnd
1992	coriolis1burn_2_2.s3d
4	coriolis1burn_2_2.s3d.sz
3145220	coriolis1burn_2_2.sf
8	coriolis1burn_2_2.sf.bnd
1992	coriolis1burn_2_3.s3d
4	coriolis1burn_2_3.s3d.sz
3145220	coriolis1burn_2_3.sf
8	coriolis1burn_2_3.sf.bnd
1992	coriolis1burn_2_4.s3d
4	coriolis1burn_2_4.s3d.sz
3145220	coriolis1burn_2_4.sf
8	coriolis1burn_2_4.sf.bnd
3145220	coriolis1burn_2_5.sf
8	coriolis1burn_2_5.sf.bnd
4	coriolis1burn_2.prt5
4	coriolis1burn_2.prt5.bnd
4	coriolis1burn_2.prt5.sz
4	coriolis1burn.binfo
8	coriolis1burn_devc.csv
4	coriolis1burn.end
4	coriolis1burn_git.txt
28	coriolis1burn_hrr.csv
12	coriolis1burn.ini
52	coriolis1burn.out
4	coriolis1burn.prt5.gbnd
4	coriolis1burn.sinfo
24	coriolis1burn.smv
4	coriolis1burn_steps.csv
89925836	total

coriolis.zip

@gforney
Copy link
Contributor

gforney commented Jul 16, 2021

can you post your input file

@gforney
Copy link
Contributor

gforney commented Jul 19, 2021

your 3d slice files will get to be pretty big . approximately 1921921924(1200 s) *(4 frames/s) ~= 135GB

I suggest you replace them with 2D slice files at various locations of interest. ie replace XB=... with PBY=0.0 or some other position

do that and let me know if you still see a problem

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@jabowery @rmcdermo @gforney