| 36 | |
| 37 | |
| 38 | === Parallel !ParaView |
| 39 | * [http://www.paraview.org/Wiki/ParaView/ParaView_Readers_and_Parallel_Data_Distribution ParaView Readers] |
| 40 | |
| 41 | ==== Parallel Data Management |
| 42 | Data must be distributed across parallel processes to take advantage of resources |
| 43 | |
| 44 | * Some !ParaView readers import in parallel |
| 45 | * Explicit parallel formats use separate files for partitions (.pvti, global.silo) |
| 46 | * Implicit parallel formats – parallel processes figure out what they need – (.vti, brick-f-values) |
| 47 | * Some !ParaView readers may seem to import in parallel |
| 48 | * Actually, import serially and then distribute |
| 49 | * Bad bad bad – 1 process needs enough memory for entire dataset plus additional space for partitioning |
| 50 | * Some !ParaView readers do NOT read in parallel |
| 51 | * ... and leave it to you (D3 filter in Paraview) |
| 52 | * See Bad bad bad above |
| 53 | |
| 54 | ==== Test Parallel Data Mangement |
| 55 | * Click Sources->Sphere |
| 56 | * Max out Theta Resolution and Phi Resolution |
| 57 | * Click Filters->Alphabetical->Process Id Scalars |
| 58 | * => Segments are colored by which process handles them |