Feature request: continuous monitoring mode for long-running services (e.g. web servers)

[XaBerr]

Hi, first of all thanks for the great work on CodeGreen, it is a really interesting and promising tool.
I have been experimenting with it on a Python Flask application running behind Apache (mod_wsgi), and I wanted to share a potential use case that might be worth considering.
If I understood correctly, CodeGreen works very well for:

• measuring standalone scripts (codegreen measure)
• benchmarking commands (codegreen run)

However, in the case of long-running services (e.g. web servers), there is a limitation:

• the application is already running (e.g. Apache + Flask)
• incoming HTTP requests are handled internally by the server
• therefore, it is not possible to directly attach CodeGreen to monitor the process continuously

It could be interesting to support a continuous monitoring mode for this type of workload (e.g. attach to a running process or log energy over time).

[saurabh-singh-rajput]

Hi Federico, @XaBerr

Thank you so much for using the tool and giving us feedback. It's a great suggestion. I'll help us shape it as you need it and improve its usability. We recently added Python API support to it, in case you haven't checked (https://smart-dal.github.io/codegreen/api/python/), which will help users embed CodeGreen into their apps manually. However, on continually running apps as you said, I may need to test it on your specific example use case. I can replicate this on my end and share it with you.

Thanks,
Saurabh