Difference between revisions of "Teleconference 2016-03-16"

From VTKM
Jump to navigation Jump to search
(Created page with "Attendees: Ken Moreland (SNL), The VTK-m paper submitted to IEEE CG&A has been accepted. However, there were not enough accepted entries to justify the special edition that w...")
 
Line 1: Line 1:
Attendees: Ken Moreland (SNL),
+
Attendees: Ken Moreland (SNL), David Pugmire (ORNL), Chris Sewell (LANL), Rob Maynard (Kitware), Berk Geveci (Kitware), Tom Fogal (NVIDIA)
  
 
The VTK-m paper submitted to IEEE CG&A has been accepted. However, there were not enough accepted entries to justify the special edition that was planned, so it looks like the data of publication will be bumped back. Ken is preparing the final materials for the publication. He is currently collecting short bios from everyone.
 
The VTK-m paper submitted to IEEE CG&A has been accepted. However, there were not enough accepted entries to justify the special edition that was planned, so it looks like the data of publication will be bumped back. Ken is preparing the final materials for the publication. He is currently collecting short bios from everyone.
Line 8: Line 8:
  
 
There are inconsistencies between the interfaces of VTKDataSetReader and VTKDataSetWriter. There does not appear to be any logical reason for the difference. It would be better if they were made to match each other.
 
There are inconsistencies between the interfaces of VTKDataSetReader and VTKDataSetWriter. There does not appear to be any logical reason for the difference. It would be better if they were made to match each other.
 +
 +
Jeremy made a little progress on annotation.
 +
 +
Example for GTC tutorial: Chris is waiting for Matt's example. Chris wants to put this together next week.
 +
 +
Chris is working with Hamish Carr on a publication on contour trees.
 +
 +
Rob spending time getting VTK-m dashboard green. In some cases a bit more red.
 +
 +
Merged in filters branch.
 +
 +
We have been getting weird test failures with ICC. Were tickling a compiler bug. We updated from ICC 14 to ICC 15 and this seems to have fixed the problem. However, this also caused more things to break. We have also added a check in the code to disable vectorization on ICC 14.
 +
 +
Tom brought up the desire to have another code sprint/hackathon. We are thinking on the June/July time frame.

Revision as of 13:40, 16 March 2016

Attendees: Ken Moreland (SNL), David Pugmire (ORNL), Chris Sewell (LANL), Rob Maynard (Kitware), Berk Geveci (Kitware), Tom Fogal (NVIDIA)

The VTK-m paper submitted to IEEE CG&A has been accepted. However, there were not enough accepted entries to justify the special edition that was planned, so it looks like the data of publication will be bumped back. Ken is preparing the final materials for the publication. He is currently collecting short bios from everyone.

We submitted a VTK-m preproposal in response to the ECP RFI. It went in Monday. We have not heard back anything about it yet.

We have gotten permission to assert copyright of VTK-m version 1.0, so we can release it when we are ready.

There are inconsistencies between the interfaces of VTKDataSetReader and VTKDataSetWriter. There does not appear to be any logical reason for the difference. It would be better if they were made to match each other.

Jeremy made a little progress on annotation.

Example for GTC tutorial: Chris is waiting for Matt's example. Chris wants to put this together next week.

Chris is working with Hamish Carr on a publication on contour trees.

Rob spending time getting VTK-m dashboard green. In some cases a bit more red.

Merged in filters branch.

We have been getting weird test failures with ICC. Were tickling a compiler bug. We updated from ICC 14 to ICC 15 and this seems to have fixed the problem. However, this also caused more things to break. We have also added a check in the code to disable vectorization on ICC 14.

Tom brought up the desire to have another code sprint/hackathon. We are thinking on the June/July time frame.