
Ariel님의 댓글
Ariel 작성일Memory overcommitment is an idea in computing that covers the task of extra memory to virtual computing devices (or processes) than the physical machine they're hosted, or operating on, truly has. This is feasible because virtual machines (or processes) don't essentially use as much memory at anyone point as they are assigned, creating a buffer. If 4 virtual machines each have 1 GB of memory on a physical machine with four GB of memory, but these digital machines are solely using 500 MB, it is possible to create additional virtual machines that benefit from the 500 MB each current machine is leaving free. Memory swapping is then used to handle spikes in memory usage. The disadvantage of this approach is that memory swap information are slower to learn from than 'actual' memory, which might lead to efficiency drops. Another drawback is that, when operating out of real memory, the system is counting on the purposes to not use the extra memory despite it being allocated to them. Ought to a program achieve this anyway, it or another has to be killed as a way to free up memory to stop the system from freezing. The OOM Killer is what performs this process. Portnoy, Matthew (2012). Virtualization Necessities. John Wiley & Sons. Siebert, Eric (2009). VMware VI3 Implementation and Administration. Prentice Corridor Professional. pp. Santosa, Mulyadi. "When Linux Runs Out of Memory". This computer science article is a stub. You will help Wikipedia by expanding it. http://inprokorea.com/bbs/board.php?bo_table=free&wr_id=2121261


