Never free and periodically crashing by OOM killer is rarely used but can be useful in limited circumstances. There are some shops that arbitrarily kill any worker over X hours old under the hypothesis that a memory leak is present or probably present.
AFAIK it's how you get memory safety in missile guidance software: you just have to put enough RAM on the board so it never run out of memory before the end of its flight.
Edit: TFA just talks about that in the “never free”
I'm aware of several Java service backends that just disable the GC entirely and take a once-a-day crash as a cost of doing business. As long as you have enough nodes that the crashes aren't correlated, your overall service can maintain decent uptime
