Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Measure the computing time of particle interpolation #1565

Closed
hlokavarapu opened this issue May 10, 2017 · 5 comments
Closed

Measure the computing time of particle interpolation #1565

hlokavarapu opened this issue May 10, 2017 · 5 comments

Comments

@hlokavarapu
Copy link
Contributor

At the moment, the time taken for the particle interpolation algorithm is not being measured. This piece of information will be important in better understanding the cost of different interpolation algorithms, the scaling properties, the time complexity as a function of number of particles, and etc.

@hlokavarapu
Copy link
Contributor Author

@gassmoeller

After adding call to computing_timer.enter_section(" Particles: Interpolate") in the function void Simulator<dim>::interpolate_particle_properties (const AdvectionField &advection_field) the timing summary is such that:

+---------------------------------------------+------------+------------+
| Total wallclock time elapsed since start    |      1.48s |            |
|                                             |            |            |
| Section                         | no. calls |  wall time | % of total |
+---------------------------------+-----------+------------+------------+
| Assemble Stokes system          |         2 |    0.0433s |       2.9% |
| Assemble composition system     |         2 |     0.209s |        14% |
| Assemble temperature system     |         2 |    0.0716s |       4.8% |
| Build Stokes preconditioner     |         1 |    0.0248s |       1.7% |
| Build composition preconditioner|         2 |    0.0134s |       0.9% |
| Particles: Interpolate          |         4 |    0.0268s |       1.8% |
| Solve Stokes system             |         2 |      0.07s |       4.7% |
| Solve composition system        |         2 |   0.00376s |      0.25% |
| Initialization                  |         1 |    0.0264s |       1.8% |
| Particles: Advect               |         4 |     0.287s |        19% |
| Particles: Exchange ghosts      |         3 |   0.00322s |      0.22% |
| Particles: Generate             |         1 |    0.0436s |       2.9% |
| Particles: Initialization       |         1 |  0.000129s |         0% |
| Particles: Initialize properties|         1 |    0.0458s |       3.1% |
| Particles: Sort                 |         4 |    0.0215s |       1.4% |
| Particles: Update properties    |         2 |      0.34s |        23% |
| Postprocessing                  |         2 |     0.695s |        47% |
| Setup dof systems               |         1 |     0.178s |        12% |
| Setup initial conditions        |         1 |       0.1s |       6.8% |
+---------------------------------+-----------+------------+------------+

Is there a method of controlling where Particles: Interpolate is listed in the summary?

@gassmoeller
Copy link
Member

I though it would be sorted in the order the std::map<std::string,Section> would sort the entries, which should be alphabetical. In your post above you have a space in front of the Particles do you add the same in the code? What happens if you remove the space?

@hlokavarapu
Copy link
Contributor Author

Super simple, thanks Rene!

+---------------------------------------------+------------+------------+
| Total wallclock time elapsed since start    |      1.42s |            |
|                                             |            |            |
| Section                         | no. calls |  wall time | % of total |
+---------------------------------+-----------+------------+------------+
| Assemble Stokes system          |         2 |    0.0414s |       2.9% |
| Assemble composition system     |         2 |     0.195s |        14% |
| Assemble temperature system     |         2 |    0.0686s |       4.8% |
| Build Stokes preconditioner     |         1 |    0.0245s |       1.7% |
| Build composition preconditioner|         2 |    0.0122s |      0.86% |
| Solve Stokes system             |         2 |    0.0406s |       2.9% |
| Solve composition system        |         2 |   0.00313s |      0.22% |
| Initialization                  |         1 |    0.0267s |       1.9% |
| Particles: Advect               |         4 |     0.286s |        20% |
| Particles: Exchange ghosts      |         3 |   0.00306s |      0.22% |
| Particles: Generate             |         1 |    0.0422s |         3% |
| Particles: Initialization       |         1 |  0.000125s |         0% |
| Particles: Initialize properties|         1 |    0.0461s |       3.3% |
| Particles: Interpolate          |         4 |    0.0251s |       1.8% |
| Particles: Sort                 |         4 |    0.0229s |       1.6% |
| Particles: Update properties    |         2 |      0.33s |        23% |
| Postprocessing                  |         2 |     0.685s |        48% |
| Setup dof systems               |         1 |     0.179s |        13% |
| Setup initial conditions        |         1 |    0.0978s |       6.9% |
+---------------------------------+-----------+------------+------------+

@bangerth
Copy link
Contributor

Will be fixed by #1573.

@gassmoeller
Copy link
Member

closed by #1573

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants