Skip to Content.
Sympa Menu

h5part - Re: [H5part] bounding boxes

h5part AT lists.psi.ch

Subject: H5Part development and discussion

List archive

Re: [H5part] bounding boxes


Chronological Thread 
  • From: Andreas Adelmann <andreas.adelmann AT psi.ch>
  • To: Mr Prabhat <prabhat AT hpcrd.lbl.gov>
  • Cc: "Biddiscombe, John A." <biddisco AT cscs.ch>, Mark Howison <mhowison AT brown.edu>, Achim Gsell <achim.gsell AT psi.ch>, h5part AT lists.psi.ch, "K. John Wu" <kwu AT lbl.gov>, Suren Byna <sbyna AT lbl.gov>
  • Subject: Re: [H5part] bounding boxes
  • Date: Sat, 4 Feb 2012 00:12:29 +0100
  • List-archive: <https://lists.web.psi.ch/pipermail/h5part/>
  • List-id: H5Part development and discussion <h5part.lists.psi.ch>

Hello Prabhat please discuss this NOT offline this is very interesting for the H5Part/H5Block community.
Tnx Andreas 

On Feb 2, 2012, at 7:30 PM, Mr Prabhat wrote:

If your data is in H5Part/H5Block format, things should be pretty easy to get going. Might be better if we can chat offline on what exactly you might want to do down the line.

Prabhat

On Feb 2, 2012, at 10:11 AM, Biddiscombe, John A. wrote:

Prabhat

That sounds interesting to play with. Although I've read about fastbit and follow threads on the list to keep myself informed. I've still not used it on anything real. Would there be a step learning curve to go from zero to a test app?

Thanks

JB

-----Original Message-----
From: Mr Prabhat [mailto:prabhat AT hpcrd.lbl.gov]
Sent: 02 February 2012 19:07
To: Biddiscombe, John A.
Cc: Mark Howison; Achim Gsell; h5part AT lists.psi.ch
Subject: Re: [H5part] bounding boxes

Hi John,

Yes, if you are only going to do spatial subsets, bboxes should work just fine. FastBit enables you do queries on any variable, or combination of variables. John Wu and I have been working on FastQuery which enables one to apply FastBit to H5Part, generic HDF5 files and NetCDF files. Happy to share the code with you if you'd like.


Prabhat

On Feb 2, 2012, at 7:51 AM, Biddiscombe, John A. wrote:

Thank Mark, I'm interested in using fastbit as well, but in this case since I already know the BBoxes, I wondered if there was a standard way of using them. I've begun integrating this into my vtkH5PartReader for paraview and in one of the simulations, I will move the code into H5Part proper once it's finished if you would like it.

It's only useful if each process was using a regular hyperslab of particles (which we are).

JB


-----Original Message-----
From: Mark Howison [mailto:mhowison AT brown.edu]
Sent: 02 February 2012 16:25
To: Achim Gsell
Cc: Biddiscombe, John A.; h5part AT lists.psi.ch; Prabhat Prabhat
Subject: Re: [H5part] bounding boxes

Hi John, you can perform fast range queries on a H5Part data set by creating a corresponding FastBit index. I'm cc'ing Prabhat who can provide you with more information about a library developed at LBL to integrate FastBit indices into H5Part files, although I don't know if it is publicly available. Mark

On Feb 2, 2012, at 10:13 AM, Achim Gsell wrote:


On Feb 2, 2012, at 3:21 PM, Biddiscombe, John A. wrote:

Does H5Part have any built in support for storing/querying/extracting bounds of regions of particles.

No. But filed an enhancement request in our tracker. We may implement it in a future release.


Achim

_______________________________________________
H5Part mailing list
H5Part AT lists.psi.ch
https://lists.web.psi.ch/mailman/listinfo/h5part

_______________________________________________
H5Part mailing list
H5Part AT lists.psi.ch
https://lists.web.psi.ch/mailman/listinfo/h5part


_______________________________________________
H5Part mailing list
H5Part AT lists.psi.ch
https://lists.web.psi.ch/mailman/listinfo/h5part

------
Dr. sc. math. Andreas (Andy) Adelmann
Staff Scientist
Paul Scherrer Institut WLGB/132 CH-5232 Villigen PSI
Phone Office: xx41 56 310 42 33 Fax: xx41 56 310 31 91
Phone Home: xx41 62 891 91 44
-------------------------------------------------------
Thursdays: ETH CAB H 83.1  +41 44 632 36 72
============================================
The more exotic, the more abstract the knowledge, 
the more profound will be its consequences.
Leon Lederman 
============================================






Archive powered by MHonArc 2.6.19.

Top of Page