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.

Sign up for GitHub

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

Jump to bottom

Thread-Parallelize blended DG-FV #2138

Merged
merged 7 commits into from
Oct 31, 2024
Merged

Thread-Parallelize blended DG-FV #2138

merged 7 commits into from
Oct 31, 2024

Conversation

DanielDoehring
Copy link
Contributor

The current version of determining whether the DG-FV stabilization should be used for an element is not thread-parallelized. This is disadvantageous when considering scaling to large machines. The proposed implementation is thread-parallelized and reduces allocations.

Example for https://github.com/trixi-framework/Trixi.jl/blob/main/examples/tree_2d_dgsem/elixir_euler_sedov_blast_wave.jl without AMR on $2^8 \times 2^8$ mesh with 2 threads:

Old version:

────────────────────────────────────────────────────────────────────────────────────────────────────
 Simulation running 'CompressibleEulerEquations2D' with DGSEM(polydeg=3)
────────────────────────────────────────────────────────────────────────────────────────────────────
 #timesteps:                664                run time:       1.25009769e+02 s
 Δt:             1.38512407e-04                └── GC time:    0.00000000e+00 s (0.000%)
 sim. time:      5.00000000e-01 (100.000%)     time/DOF/rhs!:  2.98306094e-08 s
                                               PID:            3.60262119e-08 s
 #DOFs per field:       1048576                alloc'd memory:        721.856 MiB
 #elements:               65536

────────────────────────────────────────────────────────────────────────────────────────────────────

────────────────────────────────────────────────────────────────────────────────────────────────────
Trixi.jl simulation finished.  Final time: 0.5  Time steps: 664 (accepted), 664 (total)
────────────────────────────────────────────────────────────────────────────────────────────────────

──────────────────────────────────────────────────────────────────────────────────
            Trixi.jl                     Time                    Allocations      
                                ───────────────────────   ────────────────────────
       Tot / % measured:              125s /  84.6%           95.9MiB /  40.2%    

Section                 ncalls     time    %tot     avg     alloc    %tot      avg
──────────────────────────────────────────────────────────────────────────────────
rhs!                     3.32k     103s   97.4%  31.0ms   8.92MiB   23.2%  2.75KiB
  volume integral        3.32k    71.9s   67.9%  21.6ms   4.46MiB   11.6%  1.37KiB
    pure DG              3.32k    59.7s   56.4%  18.0ms   1.42MiB    3.7%     448B
    blending factors     3.32k    8.28s    7.8%  2.49ms   1.57MiB    4.1%     496B
    blended DG-FV        3.32k    3.06s    2.9%   921μs   1.42MiB    3.7%     448B
    ~volume integral~    3.32k    857ms    0.8%   258μs   48.6KiB    0.1%    15.0B
  interface flux         3.32k    8.79s    8.3%  2.65ms   1.27MiB    3.3%     400B
  prolong2interfaces     3.32k    7.56s    7.1%  2.28ms   1.17MiB    3.0%     368B
  surface integral       3.32k    7.09s    6.7%  2.13ms   1.06MiB    2.8%     336B
  Jacobian               3.32k    3.89s    3.7%  1.17ms   0.96MiB    2.5%     304B
  reset ∂u/∂t            3.32k    3.86s    3.6%  1.16ms     0.00B    0.0%    0.00B
  ~rhs!~                 3.32k   28.7ms    0.0%  8.65μs   9.33KiB    0.0%    2.88B
  prolong2boundaries     3.32k   1.54ms    0.0%   465ns     0.00B    0.0%    0.00B
  prolong2mortars        3.32k   1.43ms    0.0%   431ns     0.00B    0.0%    0.00B
  mortar flux            3.32k    837μs    0.0%   252ns     0.00B    0.0%    0.00B
  source terms           3.32k    127μs    0.0%  38.3ns     0.00B    0.0%    0.00B
  boundary flux          3.32k   79.2μs    0.0%  23.9ns     0.00B    0.0%    0.00B
calculate dt               665    2.48s    2.3%  3.73ms     0.00B    0.0%    0.00B
analyze solution             8    268ms    0.3%  33.5ms   29.6MiB   76.8%  3.70MiB
──────────────────────────────────────────────────────────────────────────────────

New version:

────────────────────────────────────────────────────────────────────────────────────────────────────
 Simulation running 'CompressibleEulerEquations2D' with DGSEM(polydeg=3)
────────────────────────────────────────────────────────────────────────────────────────────────────
 #timesteps:                664                run time:       1.20725893e+02 s
 Δt:             1.38512407e-04                └── GC time:    0.00000000e+00 s (0.000%)
 sim. time:      5.00000000e-01 (100.000%)     time/DOF/rhs!:  2.92807732e-08 s
                                               PID:            3.50903734e-08 s
 #DOFs per field:       1048576                alloc'd memory:        734.587 MiB
 #elements:               65536

────────────────────────────────────────────────────────────────────────────────────────────────────

────────────────────────────────────────────────────────────────────────────────────────────────────
Trixi.jl simulation finished.  Final time: 0.5  Time steps: 664 (accepted), 664 (total)
────────────────────────────────────────────────────────────────────────────────────────────────────

──────────────────────────────────────────────────────────────────────────────────
            Trixi.jl                     Time                    Allocations      
                                ───────────────────────   ────────────────────────
       Tot / % measured:              121s /  85.1%           92.4MiB /  37.9%    

Section                 ncalls     time    %tot     avg     alloc    %tot      avg
──────────────────────────────────────────────────────────────────────────────────
rhs!                     3.32k     100s   97.3%  30.2ms   7.31MiB   20.8%  2.25KiB
  volume integral        3.32k    70.8s   68.8%  21.3ms   2.84MiB    8.1%     896B
    ~volume integral~    3.32k    63.2s   61.5%  19.0ms   1.42MiB    4.1%     448B
    blending factors     3.32k    7.57s    7.4%  2.28ms   1.42MiB    4.0%     448B
  interface flux         3.32k    8.49s    8.3%  2.56ms   1.27MiB    3.6%     400B
  prolong2interfaces     3.32k    7.03s    6.8%  2.12ms   1.17MiB    3.3%     368B
  surface integral       3.32k    6.59s    6.4%  1.99ms   1.06MiB    3.0%     336B
  reset ∂u/∂t            3.32k    3.64s    3.5%  1.09ms     0.00B    0.0%    0.00B
  Jacobian               3.32k    3.55s    3.5%  1.07ms   0.96MiB    2.7%     304B
  ~rhs!~                 3.32k   27.8ms    0.0%  8.37μs   9.33KiB    0.0%    2.88B
  prolong2boundaries     3.32k   1.29ms    0.0%   388ns     0.00B    0.0%    0.00B
  prolong2mortars        3.32k   1.19ms    0.0%   359ns     0.00B    0.0%    0.00B
  mortar flux            3.32k    730μs    0.0%   220ns     0.00B    0.0%    0.00B
  boundary flux          3.32k    116μs    0.0%  34.9ns     0.00B    0.0%    0.00B
  source terms           3.32k   95.3μs    0.0%  28.7ns     0.00B    0.0%    0.00B
calculate dt               665    2.47s    2.4%  3.71ms     0.00B    0.0%    0.00B
analyze solution             8    266ms    0.3%  33.2ms   27.7MiB   79.2%  3.47MiB
──────────────────────────────────────────────────────────────────────────────────

@DanielDoehring DanielDoehring added performance We are greedy refactoring Refactoring code without functional changes labels Oct 30, 2024
@github-actions GitHub Actions
Copy link
Contributor

Review checklist

This checklist is meant to assist creators of PRs (to let them know what reviewers will typically look for) and reviewers (to guide them in a structured review process). Items do not need to be checked explicitly for a PR to be eligible for merging.

Purpose and scope

  • The PR has a single goal that is clear from the PR title and/or description.
  • All code changes represent a single set of modifications that logically belong together.
  • No more than 500 lines of code are changed or there is no obvious way to split the PR into multiple PRs.

Code quality

  • The code can be understood easily.
  • Newly introduced names for variables etc. are self-descriptive and consistent with existing naming conventions.
  • There are no redundancies that can be removed by simple modularization/refactoring.
  • There are no leftover debug statements or commented code sections.
  • The code adheres to our conventions and style guide, and to the Julia guidelines.

Documentation

  • New functions and types are documented with a docstring or top-level comment.
  • Relevant publications are referenced in docstrings (see example for formatting).
  • Inline comments are used to document longer or unusual code sections.
  • Comments describe intent ("why?") and not just functionality ("what?").
  • If the PR introduces a significant change or new feature, it is documented in NEWS.md with its PR number.

Testing

  • The PR passes all tests.
  • New or modified lines of code are covered by tests.
  • New or modified tests run in less then 10 seconds.

Performance

  • There are no type instabilities or memory allocations in performance-critical parts.
  • If the PR intent is to improve performance, before/after time measurements are posted in the PR.

Verification

  • The correctness of the code was verified using appropriate tests.
  • If new equations/methods are added, a convergence test has been run and the results
    are posted in the PR.

Created with ❤️ by the Trixi.jl community.

@DanielDoehring DanielDoehring changed the title Thread-Parallelize blended DG-FV Thread-Parallelize blended DG-FV for DGSEM Oct 30, 2024
@codecov Codecov
Copy link

codecov bot commented Oct 30, 2024 •
edited
Loading

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 96.37%. Comparing base (05d1556) to head (8881a9d).
Report is 1 commits behind head on main.

Additional details and impacted files 
@@            Coverage Diff             @@
##             main    #2138      +/-   ##
==========================================
+ Coverage   96.36%   96.37%   +0.01%     
==========================================
  Files         477      477              
  Lines       37760    37720      -40     
==========================================
- Hits        36385    36349      -36     
+ Misses       1375     1371       -4
Flag Coverage Δ
unittests 96.37% <100.00%> (+0.01%) ⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

@DanielDoehring DanielDoehring added the parallelization Related to MPI, threading, tasks etc. label Oct 30, 2024
ranocha
ranocha requested changes Oct 30, 2024
View reviewed changes
Copy link
Member

@ranocha ranocha left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Did you check how it scales with more threads?

src/solvers/dgsem_tree/dg_1d.jl Show resolved Hide resolved
src/solvers/dgsem_tree/dg_2d.jl Show resolved Hide resolved
src/solvers/dgsem_tree/dg_3d.jl Show resolved Hide resolved
@DanielDoehring
Copy link
Contributor Author

DanielDoehring commented Oct 31, 2024 •
edited
Loading

Here are some measured runtimes for other number of threads. First is always the old version, second comes the proposed change.
Note: These results are just snapshots, i.e., not averaged over multiple runs.

1

Old:

──────────────────────────────────────────────────────────────────────────────────
            Trixi.jl                     Time                    Allocations      
                                ───────────────────────   ────────────────────────
       Tot / % measured:              229s /  89.1%           59.8MiB /   4.8%    

Section                 ncalls     time    %tot     avg     alloc    %tot      avg
──────────────────────────────────────────────────────────────────────────────────
rhs!                     3.32k     202s   98.8%  60.7ms   2.89MiB  100.0%     913B
  volume integral        3.32k     138s   67.3%  41.4ms   2.88MiB   99.7%     910B
    pure DG              3.32k     118s   58.0%  35.7ms     0.00B    0.0%    0.00B
    blending factors     3.32k    12.4s    6.1%  3.74ms   1.00MiB   34.6%     316B
    blended DG-FV        3.32k    5.92s    2.9%  1.78ms     0.00B    0.0%    0.00B
    ~volume integral~    3.32k    756ms    0.4%   228μs   1.88MiB   65.1%     594B
  interface flux         3.32k    22.9s   11.2%  6.90ms     0.00B    0.0%    0.00B
  prolong2interfaces     3.32k    16.3s    8.0%  4.90ms     0.00B    0.0%    0.00B
  surface integral       3.32k    13.0s    6.4%  3.93ms     0.00B    0.0%    0.00B
  reset ∂u/∂t            3.32k    6.11s    3.0%  1.84ms     0.00B    0.0%    0.00B
  Jacobian               3.32k    5.85s    2.9%  1.76ms     0.00B    0.0%    0.00B
  ~rhs!~                 3.32k   33.2ms    0.0%  10.0μs   9.33KiB    0.3%    2.88B
  prolong2mortars        3.32k   1.58ms    0.0%   475ns     0.00B    0.0%    0.00B
  prolong2boundaries     3.32k   1.15ms    0.0%   346ns     0.00B    0.0%    0.00B
  mortar flux            3.32k    615μs    0.0%   185ns     0.00B    0.0%    0.00B
  source terms           3.32k    125μs    0.0%  37.8ns     0.00B    0.0%    0.00B
  boundary flux          3.32k    103μs    0.0%  31.1ns     0.00B    0.0%    0.00B
calculate dt               665    2.46s    1.2%  3.70ms     0.00B    0.0%    0.00B
──────────────────────────────────────────────────────────────────────────────────

New:

──────────────────────────────────────────────────────────────────────────────────
            Trixi.jl                     Time                    Allocations      
                                ───────────────────────   ────────────────────────
       Tot / % measured:              228s /  89.3%           57.9MiB /   1.7%    

Section                 ncalls     time    %tot     avg     alloc    %tot      avg
──────────────────────────────────────────────────────────────────────────────────
rhs!                     3.32k     201s   98.8%  60.6ms   1.01MiB  100.0%     319B
  volume integral        3.32k     138s   67.5%  41.4ms   1.00MiB   99.1%     316B
    ~volume integral~    3.32k     125s   61.3%  37.6ms      752B    0.1%    0.23B
    blending factors     3.32k    12.8s    6.3%  3.85ms   1.00MiB   99.0%     316B
  interface flux         3.32k    22.9s   11.3%  6.90ms     0.00B    0.0%    0.00B
  prolong2interfaces     3.32k    15.9s    7.8%  4.78ms     0.00B    0.0%    0.00B
  surface integral       3.32k    12.7s    6.2%  3.82ms     0.00B    0.0%    0.00B
  reset ∂u/∂t            3.32k    6.26s    3.1%  1.89ms     0.00B    0.0%    0.00B
  Jacobian               3.32k    5.87s    2.9%  1.77ms     0.00B    0.0%    0.00B
  ~rhs!~                 3.32k   35.0ms    0.0%  10.5μs   9.33KiB    0.9%    2.88B
  prolong2boundaries     3.32k   1.24ms    0.0%   374ns     0.00B    0.0%    0.00B
  prolong2mortars        3.32k   1.14ms    0.0%   343ns     0.00B    0.0%    0.00B
  mortar flux            3.32k    515μs    0.0%   155ns     0.00B    0.0%    0.00B
  boundary flux          3.32k    396μs    0.0%   119ns     0.00B    0.0%    0.00B
  source terms           3.32k   85.7μs    0.0%  25.8ns     0.00B    0.0%    0.00B
calculate dt               665    2.43s    1.2%  3.66ms     0.00B    0.0%    0.00B
──────────────────────────────────────────────────────────────────────────────────

4

Old:

──────────────────────────────────────────────────────────────────────────────────
            Trixi.jl                     Time                    Allocations      
                                ───────────────────────   ────────────────────────
       Tot / % measured:             95.7s /  71.8%           68.6MiB /  17.1%    

Section                 ncalls     time    %tot     avg     alloc    %tot      avg
──────────────────────────────────────────────────────────────────────────────────
rhs!                     3.32k    65.9s   95.9%  19.8ms   11.8MiB  100.0%  3.63KiB
  volume integral        3.32k    41.4s   60.2%  12.5ms   7.29MiB   62.0%  2.25KiB
    pure DG              3.32k    33.4s   48.7%  10.1ms   1.42MiB   12.1%     448B
    blending factors     3.32k    5.34s    7.8%  1.61ms   2.57MiB   21.9%     812B
    blended DG-FV        3.32k    1.71s    2.5%   514μs   1.42MiB   12.1%     448B
    ~volume integral~    3.32k    869ms    1.3%   262μs   1.88MiB   16.0%     594B
  interface flux         3.32k    6.61s    9.6%  1.99ms   1.27MiB   10.8%     400B
  surface integral       3.32k    6.12s    8.9%  1.84ms   1.06MiB    9.0%     336B
  prolong2interfaces     3.32k    5.56s    8.1%  1.67ms   1.17MiB    9.9%     368B
  Jacobian               3.32k    3.68s    5.4%  1.11ms   0.96MiB    8.2%     304B
  reset ∂u/∂t            3.32k    2.52s    3.7%   759μs     0.00B    0.0%    0.00B
  ~rhs!~                 3.32k   32.6ms    0.0%  9.82μs   9.33KiB    0.1%    2.88B
  prolong2boundaries     3.32k   1.60ms    0.0%   482ns     0.00B    0.0%    0.00B
  prolong2mortars        3.32k   1.12ms    0.0%   337ns     0.00B    0.0%    0.00B
  mortar flux            3.32k    637μs    0.0%   192ns     0.00B    0.0%    0.00B
  boundary flux          3.32k    234μs    0.0%  70.4ns     0.00B    0.0%    0.00B
  source terms           3.32k    166μs    0.0%  49.9ns     0.00B    0.0%    0.00B
calculate dt               665    2.79s    4.1%  4.19ms     0.00B    0.0%    0.00B
──────────────────────────────────────────────────────────────────────────────────

New:

──────────────────────────────────────────────────────────────────────────────────
            Trixi.jl                     Time                    Allocations      
                                ───────────────────────   ────────────────────────
       Tot / % measured:             94.9s /  72.1%           65.2MiB /  12.7%    

Section                 ncalls     time    %tot     avg     alloc    %tot      avg
──────────────────────────────────────────────────────────────────────────────────
rhs!                     3.32k    65.6s   96.0%  19.8ms   8.31MiB  100.0%  2.56KiB
  volume integral        3.32k    41.0s   60.0%  12.3ms   3.84MiB   46.2%  1.18KiB
    ~volume integral~    3.32k    35.8s   52.3%  10.8ms   1.42MiB   17.1%     448B
    blending factors     3.32k    5.24s    7.7%  1.58ms   2.42MiB   29.1%     764B
  interface flux         3.32k    6.63s    9.7%  2.00ms   1.27MiB   15.3%     400B
  surface integral       3.32k    6.15s    9.0%  1.85ms   1.06MiB   12.8%     336B
  prolong2interfaces     3.32k    5.57s    8.1%  1.68ms   1.17MiB   14.0%     368B
  Jacobian               3.32k    3.70s    5.4%  1.11ms   0.96MiB   11.6%     304B
  reset ∂u/∂t            3.32k    2.52s    3.7%   760μs     0.00B    0.0%    0.00B
  ~rhs!~                 3.32k   38.0ms    0.1%  11.4μs   9.33KiB    0.1%    2.88B
  prolong2boundaries     3.32k   1.73ms    0.0%   519ns     0.00B    0.0%    0.00B
  prolong2mortars        3.32k   1.51ms    0.0%   455ns     0.00B    0.0%    0.00B
  mortar flux            3.32k    502μs    0.0%   151ns     0.00B    0.0%    0.00B
  boundary flux          3.32k    259μs    0.0%  78.0ns     0.00B    0.0%    0.00B
  source terms           3.32k    170μs    0.0%  51.2ns     0.00B    0.0%    0.00B
calculate dt               665    2.75s    4.0%  4.14ms     0.00B    0.0%    0.00B
──────────────────────────────────────────────────────────────────────────────────

8

Old:

──────────────────────────────────────────────────────────────────────────────────
            Trixi.jl                     Time                    Allocations      
                                ───────────────────────   ────────────────────────
       Tot / % measured:              116s /  72.1%           68.6MiB /  17.1%    

Section                 ncalls     time    %tot     avg     alloc    %tot      avg
──────────────────────────────────────────────────────────────────────────────────
rhs!                     3.32k    77.6s   92.6%  23.4ms   11.8MiB  100.0%  3.63KiB
  volume integral        3.32k    54.2s   64.6%  16.3ms   7.29MiB   62.0%  2.25KiB
    pure DG              3.32k    45.4s   54.1%  13.7ms   1.42MiB   12.1%     448B
    blending factors     3.32k    5.45s    6.5%  1.64ms   2.57MiB   21.9%     812B
    blended DG-FV        3.32k    1.98s    2.4%   596μs   1.42MiB   12.1%     448B
    ~volume integral~    3.32k    1.37s    1.6%   412μs   1.88MiB   16.0%     594B
  interface flux         3.32k    7.75s    9.2%  2.33ms   1.27MiB   10.8%     400B
  surface integral       3.32k    5.31s    6.3%  1.60ms   1.06MiB    9.0%     336B
  prolong2interfaces     3.32k    5.17s    6.2%  1.56ms   1.17MiB    9.9%     368B
  Jacobian               3.32k    2.75s    3.3%   827μs   0.96MiB    8.2%     304B
  reset ∂u/∂t            3.32k    2.40s    2.9%   722μs     0.00B    0.0%    0.00B
  ~rhs!~                 3.32k   60.0ms    0.1%  18.1μs   9.33KiB    0.1%    2.88B
  prolong2boundaries     3.32k   1.99ms    0.0%   600ns     0.00B    0.0%    0.00B
  prolong2mortars        3.32k   1.88ms    0.0%   567ns     0.00B    0.0%    0.00B
  mortar flux            3.32k   1.47ms    0.0%   443ns     0.00B    0.0%    0.00B
  boundary flux          3.32k    193μs    0.0%  58.2ns     0.00B    0.0%    0.00B
  source terms           3.32k    145μs    0.0%  43.6ns     0.00B    0.0%    0.00B
calculate dt               665    6.20s    7.4%  9.32ms     0.00B    0.0%    0.00B
──────────────────────────────────────────────────────────────────────────────────

New:

──────────────────────────────────────────────────────────────────────────────────
            Trixi.jl                     Time                    Allocations      
                                ───────────────────────   ────────────────────────
       Tot / % measured:              109s /  73.4%           65.2MiB /  12.7%    

Section                 ncalls     time    %tot     avg     alloc    %tot      avg
──────────────────────────────────────────────────────────────────────────────────
rhs!                     3.32k    76.0s   94.8%  22.9ms   8.31MiB  100.0%  2.56KiB
  volume integral        3.32k    52.7s   65.7%  15.9ms   3.84MiB   46.2%  1.18KiB
    ~volume integral~    3.32k    47.4s   59.1%  14.3ms   1.42MiB   17.1%     448B
    blending factors     3.32k    5.33s    6.7%  1.60ms   2.42MiB   29.1%     764B
  interface flux         3.32k    7.78s    9.7%  2.34ms   1.27MiB   15.3%     400B
  surface integral       3.32k    5.32s    6.6%  1.60ms   1.06MiB   12.8%     336B
  prolong2interfaces     3.32k    5.19s    6.5%  1.56ms   1.17MiB   14.0%     368B
  Jacobian               3.32k    2.69s    3.4%   811μs   0.96MiB   11.6%     304B
  reset ∂u/∂t            3.32k    2.27s    2.8%   683μs     0.00B    0.0%    0.00B
  ~rhs!~                 3.32k   52.2ms    0.1%  15.7μs   9.33KiB    0.1%    2.88B
  prolong2mortars        3.32k   1.89ms    0.0%   570ns     0.00B    0.0%    0.00B
  prolong2boundaries     3.32k   1.63ms    0.0%   489ns     0.00B    0.0%    0.00B
  mortar flux            3.32k   1.04ms    0.0%   312ns     0.00B    0.0%    0.00B
  boundary flux          3.32k    331μs    0.0%   100ns     0.00B    0.0%    0.00B
  source terms           3.32k    161μs    0.0%  48.6ns     0.00B    0.0%    0.00B
calculate dt               665    4.14s    5.2%  6.23ms     0.00B    0.0%    0.00B
──────────────────────────────────────────────────────────────────────────────────

@DanielDoehring DanielDoehring changed the title Thread-Parallelize blended DG-FV for DGSEM Thread-Parallelize blended DG-FV Oct 31, 2024
ranocha
ranocha approved these changes Oct 31, 2024
View reviewed changes
Copy link
Member

@ranocha ranocha left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks!

@ranocha ranocha enabled auto-merge (squash) October 31, 2024 15:44
@ranocha ranocha merged commit 8743712 into trixi-framework:main Oct 31, 2024
34 of 37 checks passed
@DanielDoehring DanielDoehring deleted the ThreadParaBlendedDG_FV branch November 1, 2024 10:32
@huiyuxie
Copy link
Member

huiyuxie commented Nov 5, 2024

Code change cc @huiyuxie

@huiyuxie huiyuxie mentioned this pull request Nov 5, 2024
Open
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@ranocha ranocha ranocha approved these changes

Assignees
No one assigned
Labels
parallelization Related to MPI, threading, tasks etc. performance We are greedy refactoring Refactoring code without functional changes
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@DanielDoehring @huiyuxie @ranocha