How to increase pypdf speed and formatting

[vedantv3]

I have tried many libaraies but pypdf has performed really well but it is cpu heavy
and takes a lot of time does any one have idea how to reduce the time and make it fast.
any ideas are appreciated

[stefan6419846]

Making it faster requires identifying current bottlenecks and fixing them. There have been some related changes recently.

Apart from this, your request is rather generic and answering it as-is thus rather impossible.

[vedantv3]

i have tried it with 2GB and 4GB ram the avg latency for both to run on pdf(2mb-3mb) is avg 5 sec for both with sending 10 requests concurrently
so is there any chance we can increase its speed

[stefan6419846]

As already said, there is nothing to help you with unless you provide more details.

What actions are you performing? How complex are the PDF files? ...

[vedantv3]

I am extracting resumes they may be two column ,they have various layout to for these things pypdf is taking much time

[stefan6419846]

This does not really answer my questions, but depending on the complexity, page.extract_text() could be slow, yes. There might be options to evaluate for speeding it up (like mypyc or seeking for other possible improvements), but this would be something for you to look into yourself and propose corresponding PRs. (For some very constrained use-cases like yours, extracting the text with pypdf might not be the best solution.)