# Analog of pvtu file for vtkhdf
**URL:** https://discourse.paraview.org/t/analog-of-pvtu-file-for-vtkhdf/13869
**Category:** Development
**Tags:** vtk
**Created:** [February 5, 2024, 4:03pm UTC](https://discourse.paraview.org/t/analog-of-pvtu-file-for-vtkhdf/13869 "2024-02-05T16:03:48Z")
**Posts on this page:** 13
**Page:** 1
### Author:  [@OsKnoth](https://discourse.paraview.org/u/OsKnoth)
#### Post date: [February 5, 2024, 4:03pm UTC](https://discourse.paraview.org/t/analog-of-pvtu-file-for-vtkhdf/13869/1 "2024-02-05T16:03:48Z")
I have alraedy distributed vtkhdf files. How can I read in this via one pvtu like file.
Oswald
---
### Author:  [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [February 5, 2024, 4:11pm UTC](https://discourse.paraview.org/t/analog-of-pvtu-file-for-vtkhdf/13869/2 "2024-02-05T16:11:53Z")
Hi @OsKnoth
Sorry, I’m not sure to follow, are you not able to open a VTKHDF file ?
---
### Author:  [@OsKnoth](https://discourse.paraview.org/u/OsKnoth)
#### Post date: [February 5, 2024, 5:01pm UTC](https://discourse.paraview.org/t/analog-of-pvtu-file-for-vtkhdf/13869/3 "2024-02-05T17:01:15Z")
I am able to solve the files independently, one by one. The files are in binary format.
The header of the files with ncdump looks as
netcdf Orography\_1 {
group: VTKHDF {
dimensions:
phony\_dim\_0 = 2560 ;
phony\_dim\_1 = 1 ;
phony\_dim\_2 = 641 ;
phony\_dim\_3 = 3 ;
phony\_dim\_4 = 640 ;
variables:
int64 Connectivity(phony\_dim\_0) ;
int64 NumberOfCells(phony\_dim\_1) ;
int64 NumberOfConnectivityIds(phony\_dim\_1) ;
int64 NumberOfPoints(phony\_dim\_1) ;
int64 Offsets(phony\_dim\_2) ;
double Points(phony\_dim\_0, phony\_dim\_3) ;
ubyte Types(phony\_dim\_4) ;
// group attributes:
:Version = 2LL, 0LL ;
:Type = “UnstructuredGrid” ;
data:
---
### Author:  [@OsKnoth](https://discourse.paraview.org/u/OsKnoth)
#### Post date: [February 5, 2024, 5:01pm UTC](https://discourse.paraview.org/t/analog-of-pvtu-file-for-vtkhdf/13869/4 "2024-02-05T17:01:55Z")
I mean “I can open …”
---
### Author:  [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [February 9, 2024, 8:30am UTC](https://discourse.paraview.org/t/analog-of-pvtu-file-for-vtkhdf/13869/5 "2024-02-09T08:30:38Z")
please share your files
---
### Author:  [@OsKnoth](https://discourse.paraview.org/u/OsKnoth)
#### Post date: [February 9, 2024, 11:28am UTC](https://discourse.paraview.org/t/analog-of-pvtu-file-for-vtkhdf/13869/6 "2024-02-09T11:28:52Z")
Can I share the two files on GoogleDrive, what is your e-mail adress
---
### Author:  [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [February 9, 2024, 3:05pm UTC](https://discourse.paraview.org/t/analog-of-pvtu-file-for-vtkhdf/13869/7 "2024-02-09T15:05:36Z")
you can just share the link here
---
### Author:  [@OsKnoth](https://discourse.paraview.org/u/OsKnoth)
#### Post date: [February 10, 2024, 8:07am UTC](https://discourse.paraview.org/t/analog-of-pvtu-file-for-vtkhdf/13869/8 "2024-02-10T08:07:38Z")
[https://drive.google.com/drive/folders/1yV0nquosQ0EfUh\_iCLiHxgMaU7oD9N9r?usp=sharing](https://drive.google.com/drive/folders/1yV0nquosQ0EfUh_iCLiHxgMaU7oD9N9r?usp=sharing)
Hopefully this work.
---
### Author:  [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [February 12, 2024, 9:40am UTC](https://discourse.paraview.org/t/analog-of-pvtu-file-for-vtkhdf/13869/9 "2024-02-12T09:40:22Z")
So you have two different .vtkhdf files. You need a single distributed .vtkhdf file, not the same thing.
FYI @Lucas_Givord
---
### Author:  [@OsKnoth](https://discourse.paraview.org/u/OsKnoth)
#### Post date: [February 12, 2024, 10:24am UTC](https://discourse.paraview.org/t/analog-of-pvtu-file-for-vtkhdf/13869/10 "2024-02-12T10:24:55Z")
Yes, similar to a .pvtu file.
---
### Author:  [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [February 12, 2024, 10:39am UTC](https://discourse.paraview.org/t/analog-of-pvtu-file-for-vtkhdf/13869/11 "2024-02-12T10:39:43Z")
I’m afraid this is not how .vtkhdf describes distributed data.
---
### Author:  [@OsKnoth](https://discourse.paraview.org/u/OsKnoth)
#### Post date: [February 12, 2024, 10:48am UTC](https://discourse.paraview.org/t/analog-of-pvtu-file-for-vtkhdf/13869/12 "2024-02-12T10:48:27Z")
So in a parallel environment all processes have to write in one file like in MPI.WriteAt. Where can I read something about distributed ouput for the vtkhdf format.
I have found this link: [Composite Data Sets for the VTKHDF format - Development - VTK](https://discourse.vtk.org/t/composite-data-sets-for-the-vtkhdf-format/12190)
Is this work in “progress”.
---
### Author:  [@Lucas\_Givord](https://discourse.paraview.org/u/Lucas_Givord)
#### Post date: [February 12, 2024, 11:09am UTC](https://discourse.paraview.org/t/analog-of-pvtu-file-for-vtkhdf/13869/13 "2024-02-12T11:09:00Z")
Hi @OsKnoth,
Composite dataset in VTKHDF works in serial but not in distributed for now. I recommend to follow the roadmap to have a clear understanding about what is supported or not in VTKHDF : [VTKHDF roadmap - Development - VTK](https://discourse.vtk.org/t/vtkhdf-roadmap/13257)
It seems you use UnstructuredGrid, right? In this case, we can already read distributed UG in ParaView. The documentation is here : [VTK File Formats - VTK documentation](https://docs.vtk.org/en/latest/design_documents/VTKFileFormats.html#unstructured-grid)
For example, here is a dummy distributed UG for 2 rank :
[unstructured.vtkhdf](https://discourse.paraview.org/uploads/short-url/pIugAy0eWdKMoRfg9oTv9v0DYqD.hdf) (66.1 KB)