Fluids

Fluids

Compute processes interrupted. Processing can be resumed –> Error message after y+ adaptation

    • mmoataz
      Subscriber
      I am running a simulation on 120 nodes. When I execute the tui command adapt-to-y+ and then try to continue iterations, sometimes I get the error message Compute processes interrupted. Processing can be resumed and the process continues without iteration till the job time ends!nWhat causes this error? and how to fix it?nIt doesn't happen each time but when it happens it always happens after adaptationnBelow is a copy of the output filenn=========================================================================================n> adapt nThe adapt/ menu will be removed in a future release, as its functionality has been integrated into other text command menus:n - The functionality related to adaption registers is available in the mesh/adapt/cell-registers menu.n - The functionality related to adaption is available in the mesh/adapt/ menu.nadapt-to-y+ 0 4nn29127 cells marked for refinement, 0 cells marked for coarseningnAdditional cells might have been marked because of the requirements of the nadaption algorithms.nMinimum allowable cell volume (m3) [0] 0nMaximum allowable number of cells [1000000] 0nOk to change the mesh? [no] yesnntdeleting conduction zones for wall 4ntdeleting conduction zones for wall 5ntdeleting conduction zones for wall 6ntdeleting conduction zones for wall 11ntdeleting conduction zones for wall 23ntdeleting conduction zones for wall 24ntdeleting conduction zones for wall 56, layer 2ntdeleting conduction zones for wall 56, layer 1ntdeleting conduction zones for wall 57, layer 2ntdeleting conduction zones for wall 57, layer 1ntdeleting conduction zones for wall 58, layer 2ntdeleting conduction zones for wall 58, layer 1ntdeleting conduction zones for wall 63, layer 2ntdeleting conduction zones for wall 63, layer 1ntdeleting conduction zones for wall 64ntdeleting conduction zones for wall 65ntdeleting conduction zones for wall 66n>> Dividing domain into 120 partitions using Metis.n  Time = 1.60821 seconds.n  Done.n>> 120 Stored Partitions:n  ----------------------------------------------------------------------n  Collective Partition Statistics:    Minimum  Maximum  Total   n  ----------------------------------------------------------------------n  Cell count               154573  166110  18986150  n  Mean cell count deviation        -2.3%   5.0%    n  Partition boundary cell count      5646   11420   1024358  n  Partition boundary cell count ratio   3.6%   7.3%   5.4%    nn  Face count               350359  390432  42785256  n  Mean face count deviation        -3.1%   8.0%    n  Partition boundary face count      6405   13213   588009   n  Partition boundary face count ratio   1.8%   3.7%   1.4%    nn  Cell weights              7e+05   1e+06   1e+08   n  Mean cell weight deviation       -43.7%  15.1%   nn  Fluid cells               154573  166110  18986150  n  Mean fluid cell deviation        -2.3%   5.0%    nn  Solid cells               0     0     0     n  Mean solid cell deviation        0.0%   0.0%    nn  Partition neighbor count        4     23     n  ----------------------------------------------------------------------n  Partition Method            Metis                 n  Stored Partition Count         120                  nDone.nn  Smoothing partition boundaries...n Dump usage:18539636 cells, 41602074 faces, 6354824 nodesnn Dump usage:18539636 cells, 41602074 faces, 6354824 nodesnn Dump usage:18986150 cells, 42785256 faces, 6609394 nodesnn Grid size ( original /  adapted /  change)n   cells ( 18539636 / 18986150 /  446514)n   faces ( 41602074 / 42785256 /  1183182)n   nodes ( 6354824 /  6046186 /  -308638)nChecking contact points...nSkipping shell creation for wall 4 because its an adapted wall.nSkipping shell creation for wall 56 because its an adapted wall.nn   51 partition interfaces of the 61825 cells aspect ratio > 50 n   18511492 groups for partitioning (including overlaps)n   Merging groups across partitions:n   1 139n   2 0nn   Using model-weighted partitioning with 2 constraints (base and solid weights).n   Partition 18511456 groups out of 18986150 cellsn   24 partition interfaces of the 572093 cells aspect ratio > 50 nMigrating 10268238 cell(s) across compute nodes...Done.ntcreating conduction zones for wall 5, thickness = 0.008, material = steel, heat generation rate = constant (0).ntcreating conduction zones for wall 6, thickness = 0.003, material = steel, heat generation rate = constant (0).ntcreating conduction zones for wall 11, thickness = 0.003, material = steel, heat generation rate = constant (0).ntcreating conduction zones for wall 23, thickness = 0.003, material = steel, heat generation rate = constant (0).ntcreating conduction zones for wall 24, thickness = 0.003, material = steel, heat generation rate = constant (0).ntcreating conduction zones for wall 57, layer 1, thickness = 0.003, material = steel, heat generation rate = constant (0).ntcreating conduction zones for wall 57, layer 2, thickness = 0.075, material = wool, heat generation rate = constant (0).ntcreating conduction zones for wall 58, layer 1, thickness = 0.008, material = steel, heat generation rate = constant (0).ntcreating conduction zones for wall 58, layer 2, thickness = 0.12, material = wool, heat generation rate = constant (0).ntcreating conduction zones for wall 63, layer 1, thickness = 0.008, material = steel, heat generation rate = constant (0).ntcreating conduction zones for wall 63, layer 2, thickness = 0.12, material = wool, heat generation rate = constant (0).ntcreating conduction zones for wall 64, thickness = 0.008, material = steel, heat generation rate = constant (0).ntcreating conduction zones for wall 65, thickness = 0.008, material = steel, heat generation rate = constant (0).ntcreating conduction zones for wall 66, thickness = 0.008, material = steel, heat generation rate = constant (0).nSlitting two sided shell conduction walls...nnSlitting wall zone 128 into a coupled wall.nSlitting wall zone 118 into a coupled wall.nSlitting wall zone 108 into a coupled wall.nSlitting wall zone 5 into a coupled wall.nSlitting wall zone 67 into a coupled wall.nSlitting wall zone 6 into a coupled wall.nSlitting wall zone 78 into a coupled wall.nSlitting wall zone 11 into a coupled wall.nSlitting wall zone 81 into a coupled wall.nSlitting wall zone 23 into a coupled wall.nSlitting wall zone 76 into a coupled wall.nSlitting wall zone 24 into a coupled wall.nSlitting wall zone 86 into a coupled wall.nSlitting wall zone 57 into a coupled wall.nSlitting wall zone 58 into a coupled wall.nSlitting wall zone 63 into a coupled wall.nSlitting wall zone 64 into a coupled wall.nSlitting wall zone 65 into a coupled wall.nSlitting wall zone 66 into a coupled wall.ndone.nnUpdating shell metrics...done.nnShell junction connectivity...donennShell creation done.nn> nadapt/         file/          server/nadjoint/        mesh/          solve/ndefine/         parallel/        surface/ndisplay/        plot/          switch-to-meshing-modenexit          report/         views/nn> parallel timer resetnn> nadapt/         file/          server/nadjoint/        mesh/          solve/ndefine/         parallel/        surface/ndisplay/        plot/          switch-to-meshing-modenexit          report/         views/nn> solve iterate 25nn iter continuity x-velocity y-velocity z-velocity   energy      k    omega do-intensi    heat    otemp    emass    ovel    yplus   time/itern  125 2.8963e-02 1.9795e-04 1.9622e-04 7.8200e-05 9.2095e-07 6.3045e-03 1.1002e-02 1.2082e-05 2.4309e+04 5.2268e+02 -1.4222e-02 1.1828e+00 9.8612e+00 0:03:29  25nn===============Message from the Cortex Process================================nnCompute processes interrupted. Processing can be resumed.nn==============================================================================n
    • RK
      Ansys Employee
      Hello, nThe issue might be because of the global courant number limit. Can you please check that? Thanks! n
    • mmoataz
      Subscriber
      But this is a steady simulation. Does courant number have any effect here? n
Viewing 2 reply threads
  • You must be logged in to reply to this topic.