Teleconference 2019-03-27

Attendees: Ken Moreland (SNL), Allie Vacanti (Kitware), Rob Maynard (Kitware), Sujin Philip (Kitware), Matt Larsen (LLNL), Dave Pugmire (ORNL), Nick Davis (SNL), Berk Geveci (Kitware), Hank Childs (UO), Ollie Lo (LANL)

ECP Updates

• New round of P6 Activities
  • Feedback was to mash things up more and list more varied things done for each activity.
  • All PI's should review changes at: https://jira.exascaleproject.org/issues/?filter=14802
• Need new KPP-3 metrics
  • Only "Integration" metric is due
  • Due date moved back to April 8
  • Hank is working on this
• Looks like everyone has added to quarterly report
  • Don't forget to add to outreach and communication

ECP deliverable updates

FY19Q2

• [MS-19/04] Ghost Cells
  • Done.
• [MS-19/05] Merge Points
  • Done.
• [MS-19/06] Connected Components
  • Merge requests are in
    • Which is the most recent one?
    • Might be listed in closed instead of merged
  • Sent highlight slide
  • Missing short user's manual
• [MS-19/07] Advect Time Varying
  • MR is in
  • Should be an MR for the user's guide

FY19Q3

• [MS-19/08] Point Movement 2
  • There are already some filters in for point movement.
  • Still need some user's guide documentation

FY19Q4

• [MS-19/09] Path Geometry
  • Not started
  • Some of the functionality is already there
• [MS-20/01] Lightweight Cell Library
  • Been doing some changes on the interface to the VTK-m cell library.
  • Most of the underlying library is ready.
  • Almost ready.
  • Still need to implement VTK-m data around it.
• [MS-20/02] Specialized Data Models
  • Some work on how XGC used a rotated triangular grid for a bunch of wedges.
  • Need to generalize some code written by Matt.
• [MS-20/03] Feature-Sensitive Surface Norm
  • Still working on bitfield implementation
    • Needed for graph traversal
  • Everything else is complete
• [MS-20/04] VTK-m Release 3
  • On track
• [MS-20/05] Cell Metrics
  • Progress seems a bit slow.
  • Will make a sprint during the Spring semester
• [MS-20/06] Contouring
  • Need to support all cell types
• [MS-20/07] Advanced Flow Algorithms
  • Still deciding what we want to do.
    • Planning on talking during DOECGF
    • Hybrid parallel seems like a good way to go
  • Dave is working on generalizing cell lookups and evaluators to work on all grid types

Rountable notes

We should try to collect some ECP data so that as we do reporting we can quickly generate relevant images. It would be great if we could also share across ALPINE and other vis projects.

There have been some code checked in that does not follow VTK-m coding styles (such as member variables not being capitalized, always using this pointer). We should be more vigilant during code reviews.

RuntimeDeviceTracker now has a single instance per thread (enforced).

Looking at CUDA 10.1. Most of our suppression for CUDA are not needed there. While removing the suppression, some warnings of real problems are being shown. Going to work through that. Also looking at some warnings with the new versions of Thrust.

Can now force cast a virtual array among types.

Sujin working on VTK/VTK-m data set. Mostly a matter of merging the code it.

There appears to be a crash in Cuda 10 linking shared libraries with the move of CellLocators into a library.

Matt has been working on spacial searches and rays.