ParaView 3 Telecon 09-17-2009
| Item | People | Description |
|---|---|---|
| 1 | Ice | MARS/V&V Milestone issues |
| 2 | Moreland | Stronger Dashboard
|
| 3 | Thompson | Statistics Update |
| 4 | Ayachit | Partial Arrays with Plot Over Line (Bug #9252) |
| 5 | Moreland | Time pipeline changes owner |
| 6 | Moreland | Ultravis workshop |
| 7 | Bauer | In-situ paper |
| 8 | Moreland | PV scalar bar to VTK? |
MARS Milestone
Previously:
The scaling studies have stalled. Kitware does not have any CTH simulation that makes sense to scale past about 500 processors. Thus, the fraction of time spent in fragment analysis is not representative (the entire dataset is replicated in ghost cells.
As an action, someone at Sandia needs to work with someone in 1500 to get some reasonable scaling numbers.
Nathan has been able to run larger data. There are some scaling issues that still need to be worked out.
This week:
Nathan talked to Karen Devine for suggestions. She suggested using a "data dictionary" to resolve fragments. You do two all-to-all broadcasts to choose which id to assign to each fragment.
Berk reports that the biggest bottleneck to scaling is the exchange of ghost cell information. One approach would be to not do ghost cells and do fragment resolution later. The down side is that interfaces become non-watertight. We should at least try approaches that do no communication (such as CTH parts) to at least demonstrate the scalability.
Another possibility is to try to leverage the ghost levels provided by CTH.
Stronger Dashboard
Previously:
Dave P. is now monitoring the dashboard.
Utkarsh is still going to think about how to run more tests faster.
In Utkarsh's branding branch, allows you to run multiple tests in one invocation and have each have their own result images. Between tests, the GUI resets but the connection remains. You can also isolate tests that could effect others (like the color palette tests).
This Week:
Might be committing the branding stuff next week.
Statistics Update
Previously:
- Promises (empty).
This Week:
- Initial implementation complete.
- Should there be a "no second input" version of each filter?
- Approaches for handling shared points in distributed-memory mode?
- How to create tests for plugins?
- If/when to merge into 3.6 branch?
- Semi-offtopic: "Root" of multiblock of tables appears to be mashed version of first table?
- Semi-offtopic: Possible to include HTML/images in online docs?
