Locking data structures in distributed systems

I am completely new to the concepts of distributed systems. Please let me know even if the question needs to be rephrased.

I am trying to create distributed systems with 10 clients and one server. On the server side, there is a queue that clients can access one at a time. So, what locking mechanism can you use to avoid false data? Are semaphores possible in this situation? If possible, kindly provide a link to have a much deeper knowledge of this.

+3


source to share


1 answer


Server-side semaphores are possible and in fact that is the way to go. On a GNU / Linux system such as Debian, see man 7 sem_overview

and man 1 lockfile

.

The simplest method probably allows the server to serve no more than one client at a time, discarding all requests from other clients. A rejected client waits for a random (undefined) period of time and then tries again.



Another technique might be to allow server queue requests, but this is more difficult (and may still involve discarding some requests).

+1


source







All Articles