Is it common for people in academia to experience prolonged lack of sleep and how to deal with it?
There are three basic approaches to handling this:
- Paul Erdos' approach. Pros: it worked for Erdos, after a fashion. Cons: you are not Erdos.
- Crack, maybe have a psychotic episode, and leave academia. Pros: it's the easiest course, and the one you're on now. Cons: everything else about it.
- Recognize that your career is not a sprint, but a marathon, and start triaging less important work in favor of sleep. Working sleep-deprived is exactly as smart as working drunk, so you are already cutting your productivity, just in an uncontrolled manner. Pros: Survival, possibly having a career. Cons: you will have to say "no" to people.
The truth is, in academia there is an endless amount of work that you could be doing: you will never be done. The only choice you get to make is whether you will choose the things that you do not do, or whether you will make self-destructive choices that mean those choices get made for you arbitrarily by the fracture points of your body and mind.
You are following the "path of the burn out". You may be interested in reading this question and its answers :
What can I do to recover from a short term burnout?
By main advice is always the same : seek advice from a professional, there is nothing that can be said here except "don't do that", "enjoy life outside the lab" and "you have to realize that you won't miss anything by not being 24/7 at your desk". For the rest, a professional is needed.
There are short-term and long-term approaches. Sometimes, a lack of sleep is necessary because of, eg, unexpected and valuable time on expensive valuable equipment X. Or, because the grant really is due today and you didn't budget time correctly. Or because process X has to be babysat and really does take 72 hours. For those short-term upsets, coffee and lots of water works, followed by about twice as much rest as the time you missed sleeping. Never pretend you can make time up on your long-term schedule by missing sleep. For processes, that part just sucks, either try to work in shifts with another student or set up a nearby sleeping area. If you're gadget-friendly, simple setups can often improve efficiency by allowing uninterrupted sleep.*
Long-term, missing sleep does not work for most people. Planning can help a lot. So, start by figuring out everything you want to do, estimate how long it should take, and then prioritize. Start by filling in an 8 hour workday and try to get everything on your list done. Listing tasks such as 'try X' works better than 'solve Y'. For most people, it won't come close to happening in 8 hours. Next, measure the ratio of estimated to actual time and use that to scale your future estimates. Keep reviewing.
Next, find a weekly hour limit that keeps you healthy. Then, optimize for efficiency. (commuting 2 hours?? move. Reading StackOverFlow instead of working? Um... Stressed, not getting much done? - Try the gym.)
Finally, after a few iterations, compare your productivity to your peer group while adjusting for your career goals and choose realistically. If you aren't close to living healthily while being as or more productive as people likely to achieve your career goals, it would be reasonable to rethink them.
*One of my friends moved into his office. There was a lab shower, so it wasn't bad - and saved a ton on housing. Add a lockable door == no problem.