Teleconference 2019-02-20

Attendees: Ken Moreland (SNL), Nick Davis (SNL), Dave Pugmire (ORNL), Hank Childs (UO), Matt Larsen (LLNL)

ECP Updates

• FY20-23 planning document submitted

ECP deliverable updates

FY19Q2

• [MS-19/04] Ghost Cells
  • Ghost cell filter for structured cells has been merged in.
  • Matt Larsen has checked in some similar code for extracting regions from structured cells.
  • Going to absorb that back into Dave's work
  • After that we can declare success.
  • Still need to add user's guide documentation.
  • Matt brought up that the ghost cell field is stored in UInt8, which is not recognized by filters by default.
• [MS-19/05] Merge Points
  • Ken has an implementation in a merge request.
  • Working through CUDA compile error.
• [MS-19/06] Connected Components
  • Ollie working on bug fixes.
  • Adding a filter in addition to a worklet.
  • Almost at the point to submit an MR
• [MS-19/07] Advect Time Varying
  • Some new grid evaluator MRs are in, but there are some issues with CUDA. Still struggling with this. Cannot replicate on local machine.

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
• [MS-20/01] Lightweight Cell Library
  • Sujin is wrapping up tests for hex and quad and will then be able to show what is available.
• [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
  • Working on fixing normals to be consistent across surface.
• [MS-20/04] VTK-m Release 3
  • On track
• [MS-20/05] Cell Metrics
  • Brent is working on this.
  • Steven Walten will take over soon as Brent is about to graduate.
• [MS-20/06] Contouring
  • Need to support all cell types
• [MS-20/07] Advanced Flow Algorithms
  • Still deciding what we want to do.
  • Dave is working on generalizing cell lookups and evaluators to work on all grid types

Rountable notes

Matt ran into an issue where DIY is faking MPI symbols. No way to prevent these from leaking elsewhere. Ascent and VisIt make compilations for both with MPI and not MPI. Maybe VTK-m should do the same.