Process synchronization

from Wikipedia, the free encyclopedia

In programming , process synchronization (or simply synchronization for short ) means the coordination of the timing of several concurrent processes or threads . It does not matter whether it is threads in a program, programs on a computer or processes in a distributed system that communicate over a network .

The purpose of coordination is usually one of the following:

The synchronization of the processes now consists in the fact that, in general, one process has to wait for an event that is triggered by another. It is the job of the scheduler to implement the resulting causal dependencies in such a way that a process only gets computing time if all the necessary conditions are met. Specifically, this is usually solved by the classic control mechanisms such as locks , semaphores and monitors .

The dependencies between the processes can be modeled formally using a Petri net , for example . A typical problem in this context is that processes can get jammed if circular dependencies are formed - this becomes clear , for example, with the problem of the dining philosophers . Another prototypical problem is the control of a producer / consumer system .

See also

literature

  • James H. Anderson, Yong-Jik Kim, Ted Herman: Shared-memory mutual exclusion: major research trends since 1986 . In: Distrib. Comput. tape 16 , no. 2-3 . Springer-Verlag, September 2003, ISSN  0178-2770 , p. 75-110 , doi : 10.1007 / s00446-003-0088-6 .
  • M. Raynal, D. Beeson: Algorithms for mutual exclusion . MIT Press, Cambridge MA 1986, ISBN 0-262-18119-3 .