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

Improve performance of ImageDecoder (required nvImageCodec >= 0.4.0) #5576

Draft
wants to merge 14 commits into
base: main
Choose a base branch
from
Draft

Improve performance of ImageDecoder (required nvImageCodec >= 0.4.0) #5576

wants to merge 14 commits into from
+267 −247

Conversation

jantonguirao
Copy link
Contributor

@jantonguirao jantonguirao commented Jul 22, 2024 •
edited
Loading

Category:

Other Performance

Description:

  • Populate the CUDA event pool with a few events per thread, to avoid creation of threads during the execution of the pipeline
  • Use task priority in the thread pool used by nvimagecodec, so that tasks are executed on FIFO order
  • Fix the preallocated_batch_size to take into account the hw_load. Not doing so, causes cuMemFree calls during the pipeline run.

Additional information:

Affected modules and functionalities:

ImageDecoder mostly (any operator that has a mixed backend thread pool)

  • ImageDecoder

Key points relevant for the review:

Tests:

  • Existing tests apply
  • New tests added
    • Python tests
    • GTests
    • Benchmark
    • Other
  • N/A

Checklist

Documentation

  • Existing documentation applies
  • Documentation updated
    • Docstring
    • Doxygen
    • RST
    • Jupyter
    • Other
  • N/A

DALI team only

Requirements

  • Implements new requirements
  • Affects existing requirements
  • N/A

REQ IDs: N/A

JIRA TASK: N/A

@jantonguirao
Copy link
Contributor Author

!build

@dali-automaton
Copy link
Collaborator

CI MESSAGE: [16808221]: BUILD STARTED

@dali-automaton
Copy link
Collaborator

CI MESSAGE: [16808221]: BUILD FAILED

@dali-automaton
Copy link
Collaborator

CI MESSAGE: [16811949]: BUILD STARTED

@dali-automaton
Copy link
Collaborator

CI MESSAGE: [16813241]: BUILD STARTED

@dali-automaton
Copy link
Collaborator

CI MESSAGE: [16813356]: BUILD STARTED

@dali-automaton
Copy link
Collaborator

CI MESSAGE: [16813388]: BUILD STARTED

@dali-automaton
Copy link
Collaborator

CI MESSAGE: [16814374]: BUILD STARTED

@dali-automaton
Copy link
Collaborator

CI MESSAGE: [16811949]: BUILD PASSED

@dali-automaton
Copy link
Collaborator

CI MESSAGE: [16814374]: BUILD PASSED

@dali-automaton
Copy link
Collaborator

CI MESSAGE: [16825232]: BUILD STARTED

@dali-automaton
Copy link
Collaborator

CI MESSAGE: [16825232]: BUILD PASSED

@jantonguirao jantonguirao marked this pull request as draft July 23, 2024 08:31
@dali-automaton
Copy link
Collaborator

CI MESSAGE: [16836552]: BUILD STARTED

@dali-automaton
Copy link
Collaborator

CI MESSAGE: [16836608]: BUILD STARTED

@dali-automaton
Copy link
Collaborator

CI MESSAGE: [16836608]: BUILD PASSED

@dali-automaton
Copy link
Collaborator

CI MESSAGE: [16842081]: BUILD STARTED

@dali-automaton
Copy link
Collaborator

CI MESSAGE: [16842081]: BUILD FAILED

@review-notebook-app Review Notebook App
Copy link

Check out this pull request on  ReviewNB

See visual diffs & provide feedback on Jupyter Notebooks.

Powered by ReviewNB

@jantonguirao jantonguirao changed the title Create a few CUDA events per thread pool thread Optimize thread pool Jul 23, 2024
@dali-automaton
Copy link
Collaborator

CI MESSAGE: [16850269]: BUILD STARTED

@dali-automaton
Copy link
Collaborator

CI MESSAGE: [16850269]: BUILD PASSED

@dali-automaton
Copy link
Collaborator

CI MESSAGE: [19606907]: BUILD STARTED

@dali-automaton
Copy link
Collaborator

CI MESSAGE: [19606907]: BUILD FAILED

@dali-automaton
Copy link
Collaborator

CI MESSAGE: [19651072]: BUILD STARTED

@dali-automaton
Copy link
Collaborator

CI MESSAGE: [19651072]: BUILD FAILED

@dali-automaton
Copy link
Collaborator

CI MESSAGE: [19682967]: BUILD STARTED

@jantonguirao jantonguirao changed the title Improve performance of ImageDecoder Improve performance of ImageDecoder (required nvImageCodec >= 0.4.0) Nov 13, 2024
@dali-automaton
Copy link
Collaborator

CI MESSAGE: [20363924]: BUILD STARTED

@dali-automaton
Copy link
Collaborator

CI MESSAGE: [20363924]: BUILD FAILED

@jantonguirao
Attempt fix
6a6fe76 
Signed-off-by: Joaquin Anton Guirao <janton@nvidia.com>
@jantonguirao
Improve performance of ImageDecoder
fcd2b4f 
Signed-off-by: Joaquin Anton Guirao <janton@nvidia.com>
@jantonguirao
Document DALI_EVENT_POOL_INITIAL_SIZE
87de4c9 
Signed-off-by: Joaquin Anton Guirao <janton@nvidia.com>
@jantonguirao
Fixes
5bfd919 
Signed-off-by: Joaquin Anton Guirao <janton@nvidia.com>
@jantonguirao
Refactoring
b696ad6 
Signed-off-by: Joaquin Anton Guirao <janton@nvidia.com>
@jantonguirao
Update
e9bec0a 
Signed-off-by: Joaquin Anton Guirao <janton@nvidia.com>
@jantonguirao
Changes in header
f2302aa 
Signed-off-by: Joaquin Anton Guirao <janton@nvidia.com>
@jantonguirao
Update nvimgcodec header
0c838e3 
Signed-off-by: Joaquin Anton Guirao <janton@nvidia.com>
@jantonguirao
Use skip_pre_sync
acf2e04 
Signed-off-by: Joaquin Anton Guirao <janton@nvidia.com>
@jantonguirao
Install nvtiff in tests
20d5af1 
Signed-off-by: Joaquin Anton Guirao <janton@nvidia.com>
@jantonguirao
Bump nvImageCodec tag
811a760 
Signed-off-by: Joaquin Anton Guirao <janton@nvidia.com>
@jantonguirao
Use release of nvimagecodec
fcb816e 
Signed-off-by: Joaquin Anton Guirao <janton@nvidia.com>
@jantonguirao
Bump to nvimmagecodec 0.4.0
5619c21 
Signed-off-by: Joaquin Anton Guirao <janton@nvidia.com>
@jantonguirao
Bump nvimagecodec version in conda
0130829 
Signed-off-by: Joaquin Anton Guirao <janton@nvidia.com>
@dali-automaton
Copy link
Collaborator

CI MESSAGE: [20528431]: BUILD STARTED

@dali-automaton
Copy link
Collaborator

CI MESSAGE: [20528431]: BUILD FAILED

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@awolant awolant awolant approved these changes

@klecki klecki klecki approved these changes

Assignees

@klecki klecki

@awolant awolant

Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
4 participants
@jantonguirao @dali-automaton @klecki @awolant