feat: put the frontend outside of docker container

[xnought]

Draft work in progress, don't merge yet

#116

Please let me know if this new method is faster (moving the frontend out of the docker container)

7432c8ed-e333-469f-aea0-791049e2ecf5.mp4

Watch the video where I go through the following steps:

All I do is

Go to this branch

git checkout faster

hard restart

./run.sh hard_restart

run the frontend not in the docker

cd frontend
yarn install
yarn dev --open

upload all the proteins to test stuff out

cd .. # go back to project root
./run.sh upload_all 

Go to http://localhost:5173

[ansengarvin]

I just took a look at this! I tested the website, once with the changes here and once without them. TL;DR: There didn't seem to be a significant increase in speed between either version, but both versions were loading quickly.

Both versions continue to have a problem on my machine where Vmmem (which Docker instantiates) keeps consuming memory once the docker containers are stopped. I suspect this has nothing to do with our code, however, and that it's just a problem with how Windows interacts with WSL (maybe even only on my machine).

Some statistics! This is what my task manager showed after closing out docker:

Version without frontend in docker (this version):

Version with frontend in docker (in main):

You can see that there was a small reduction in memory usage, but it didn't seem to greatly increase load times for me. I ran both of these tests immediately after restarting my computer and with no additional programs running, which I believe may have been a large factor in it being slower on my computer.

While it didn't seem to greatly increase speeds for me, I think we should have the others try it on their machines and see if they notice a significant difference. I also think it's going to be important to try to test this on the deployment server, just so we can bypass any weirdness that our individual dev environments might bring.

[xnought]

So weird. Thank you so much for the effort in trying, I appreciate it greatly! Atleast we know now it wasn't the frontend that was super slow (probably).

[xnought]

Conclusion, no big difference, so look into other ways to speedup.