A temporary fix for overloaded lemmy.ml servers, via THAIO (Throw-Hardware-At-It Optimisation)…
Nothing more permanent than a temporary solution
Bro that’s the Reddit server.
I actually ran a moderately active (like 20,000 hits a day) small business site from a laptop for a couple years. Of course one of the first thing I did was put a “SERVER DO NOT SHUT DOWN” sticker on it, and set the power settings so closing the lid did not shut down or sleep the computer. It was a Dell 7000 series with 16GB IIRC, it did great.
Not advertising here, but with this low traffic you could be in a permanent free tier with AWS with all the availability guarantees. It doesn’t work with EC2, but for serverless solutions (ApiGateway, Lambda, DynamoDB) they have something like “we start charging after 1M calls per month” (don’t quote me on this exact number). I have a couple of pet projects working this way
This looks like my old laptop - Lenovo Y510P. Even down to the slight abrasion below the mousepad from the user wearing a watch with a metal band.
It had SLI GPUs in a laptop through the ultra bay. It was a beast for about 20 minutes until the heat built up.
Unplug it real quick so we can find out
