Any ideas on how to preload youtube-dl into RAM to save on disk i/o?

[TheBoroer]

Make sure you are using the latest version

• I've verified and I assure that I'm running youtube-dl 2018.03.10

Before submitting an issue make sure you have:

• At least skimmed through the README, most notably the FAQ and BUGS sections
• Searched the bugtracker for similar issues including closed ones
• Checked that provided video/audio/playlist URLs (if any) are alive and playable in a browser

What is the purpose of your issue?

• Bug report (encountered problems with youtube-dl)
• Site support request (request for adding support for a new site)
• Feature request (request for a new functionality)
• Question
• Other

---

Description of your issue, suggested solution and other information

When running multiple youtube-dl instances with a long queue of video, I'm noticing a lot of disk I/O and high CPU usage when using the -J argument to get JSON for a youtube video.

I'm pretty sure it's because of all the modules that have to be loaded every time the script runs. I'm thinking of ways to wrap youtube-dl in another python script that initializes it only once (and stores it all in RAM so when it needs to run again everything is already preloaded and ready to go). I feel like this could also speed up youtube-dl execution time greatly.

Does anyone have any suggestions on linux tools/commands or other scripts that can help me do this? I would hate to be re-inventing the wheel if I didn't have to 😋

[dstftw]

Just pass this long queue of videos as list of URLs to single youtube-dl instance so that modules will be loaded only once. Alternatively you can implement a simple daemon like https://github.com/jaimeMF/youtube-dl-api-server.

[TheBoroer]

@dstftw
Passing a long queue of videos as a list of urls isn't possible because the videos are getting added on the fly (not at the time I launch the first youtube-dl instance). Thanks for the daemon suggestion! I'll look into that 👍