[Feature request] create a standalone reference / timing framework for CRTM evaluation

[BenjaminTJohnson]

Objective:
Develop a testing application to thoroughly exercise the CRTM model under various configurations (e.g., Cloud LUTs, multiple profiles) and collect performance metrics.

Metrics to Track:

CPU Timing: Measure computational time across different setups.
Wall Clock Time: Record actual time elapsed during the runs.
Profiling: Provide detailed profiling data for performance analysis.
Requirements:

Maintain a history of all previous runs, including:
CRTM version
Architecture and compiler details
Configuration specifics (e.g., Cloud LUTs, profiles used, etc.)
The application should be capable of comparing results between versions to track performance changes over time.

[YingtaoMa]

To me this can be used as a benchmark testbed for evaluating the CRTM and other RT models, and any new features or updates implemented in CRTM.

…

On Thu, Sep 5, 2024 at 3:18 PM Benjamin Johnson ***@***.***> wrote: *Objective*: Develop a testing application to thoroughly exercise the CRTM model under various configurations (e.g., Cloud LUTs, multiple profiles) and collect performance metrics. *Metrics to Track*: CPU Timing: Measure computational time across different setups. Wall Clock Time: Record actual time elapsed during the runs. Profiling: Provide detailed profiling data for performance analysis. Requirements: *Maintain a history of all previous runs, including*: CRTM version Architecture and compiler details Configuration specifics (e.g., Cloud LUTs, profiles used, etc.) The application should be capable of comparing results between versions to track performance changes over time. — Reply to this email directly, view it on GitHub <#178>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AMP5X4ONYGXKP2XVU3BCEK3ZVCVAVAVCNFSM6AAAAABNXEQQZWVHI2DSMVQWIX3LMV43ASLTON2WKOZSGUYDQNJUGU3TCOA> . You are receiving this because you were assigned.Message ID: ***@***.***>