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

Add response statistics #325

Merged
merged 10 commits into from
Feb 17, 2024
Merged

Add response statistics #325

merged 10 commits into from
Feb 17, 2024

Conversation

kthui
Copy link
Contributor

@kthui kthui commented Feb 7, 2024

Related PRs:

Add response statistics API.

Example statistics endpoint output to client:

{
  'model_stats': [{
    'name': 'square_int32',
    ...
    'inference_stats': {
      ...
      'response_stats': {
        '0': {
          'compute_infer': {'count': 3, 'ns': 1200514434},
          'compute_output': {'count': 3, 'ns': 600346215},
          'success': {'count': 2, 'ns': 1200431061},
          'fail': {'count': 1, 'ns': 600429588},
          'empty_response': {'count': 0, 'ns': 0}
        },
        '1': {
          'compute_infer': {'count': 3, 'ns': 1200250644},
          'compute_output': {'count': 3, 'ns': 600358040},
          'success': {'count': 1, 'ns': 600202056},
          'fail': {'count': 2, 'ns': 1200406628},
          'empty_response': {'count': 0, 'ns': 0}
        },
        '2': {
          'compute_infer': {'count': 3, 'ns': 1200252841},
          'compute_output': {'count': 2, 'ns': 400214433},
          'success': {'count': 1, 'ns': 600183751},
          'fail': {'count': 1, 'ns': 600188918},
          'empty_response': {'count': 1, 'ns': 400094605}
        },
        '3': {
          'compute_infer': {'count': 2, 'ns': 800160632},
          'compute_output': {'count': 1, 'ns': 200133237},
          'success': {'count': 0, 'ns': 0},
          'fail': {'count': 1, 'ns': 600214152},
          'empty_response': {'count': 1, 'ns': 400079717}
        },
        '4': {
          'compute_infer': {'count': 1, 'ns': 400086093},
          'compute_output': {'count': 1, 'ns': 200104114},
          'success': {'count': 0, 'ns': 0},
          'fail': {'count': 1, 'ns': 600190207},
          'empty_response': {'count': 0, 'ns': 0}
        },
        '5': {
          'compute_infer': {'count': 1, 'ns': 400089434},
          'compute_output': {'count': 0, 'ns': 0},
          'success': {'count': 0, 'ns': 0},
          'fail': {'count': 0, 'ns': 0},
          'empty_response': {'count': 1, 'ns': 400089434}}
        }
      }
    ...
    }
  }]
}

src/infer_response.h Outdated Show resolved Hide resolved
include/triton/core/tritonbackend.h Outdated Show resolved Hide resolved
include/triton/core/tritonbackend.h Outdated Show resolved Hide resolved
oandreeva-nv
oandreeva-nv previously approved these changes Feb 16, 2024
Copy link
Contributor

@oandreeva-nv oandreeva-nv left a comment

Choose a reason for hiding this comment

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

TMLG, but I would also let @rmccorm4 take a look

Copy link
Contributor

@rmccorm4 rmccorm4 left a comment

Choose a reason for hiding this comment

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

LGTM overall, main comment is on the use of "zero" compute time for "empty".

Copy link
Contributor

@rmccorm4 rmccorm4 left a comment

Choose a reason for hiding this comment

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

Nice work! 🚀

@kthui kthui merged commit 854bdcb into main Feb 17, 2024
1 check passed
@kthui kthui deleted the jacky-res-stats branch February 17, 2024 01:38
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Development

Successfully merging this pull request may close these issues.

4 participants