onediff icon indicating copy to clipboard operation
onediff copied to clipboard
verified publisher icon siliconflow

Compilation optimization issue

Open chuck-ma opened this issue 1 year ago • 3 comments

I am currently using onediff, and the total time for compile_pipe and load_pipe is about 30 seconds (even after the first compilation and save_pipe). This duration is too exaggerated in a serverless scenario. Is there a way to utilize multi-core CPUs to reduce the time for compilation and loading?

chuck-ma avatar Apr 23 '24 11:04 chuck-ma

I am currently using onediff, and the total time for compile_pipe and load_pipe is about 30 seconds (even after the first compilation and save_pipe). This duration is too exaggerated in a serverless scenario. Is there a way to utilize multi-core CPUs to reduce the time for compilation and loading?

Currently, the compilation needs to be a single thread run.

We will do some optimization later for the compilation/loading time in later version.

strint avatar Apr 25 '24 07:04 strint

I am more than happy to provide some assistance, if there is anything I can do.

chuck-ma avatar Apr 26 '24 04:04 chuck-ma

I am more than happy to provide some assistance, if there is anything I can do.

Thanks! We are making a prototype for this, and it's expected to be released at the end of May. It should need help, stay tuned.

strint avatar Apr 26 '24 05:04 strint