Current Proceedings on Technology

Yazarlar: Farhad Soleimanian Gharehchopogh, Mutlu Avci, Zeinab Abbasi Khalifehlou

Konular:-

Anahtar Kelimeler:-

Özet: Particularly in operating systems, a quite a few of researches are focused on solving deadlock problems. Deadlock is a common problem bounded efficiency of the system and very often occurs where multi-processes compete for some sharable resources.. Deadlocked status means that in among of processes which each of them is waiting for an event while another process can cause it switches to ready state and then runs. The necessary conditions for deadlock are mutual exclusion, hold and wait, preemption, and circular wait. There are four strategies to handle deadlocks as if include ignorance, detection, avoidance and prevention. Deadlock Prevention (DP) is a true real-time solution; however some researchers see Deadlock Avoidance (DA) as less restrictive. In DP, the concern is to condition a system to remove any possibility of deadlocks occurring.This paper is a review of the literature of techniques for solving deadlock problems in operating system. The solution methods for DP were selected according to a set of criteria of work relevancy their internal architecture techniques were reviewed. Our work addresses the issue of DP via these methods. A classification considering the main goal of the methods has been made methods to handle deadlocks. A discussion is presented for the suitability of resolution methods of deadlock problems in operating systems.  Keyworks:  Deadlock Prevention, Deadlock Avoidance, Deadlock Detection, Deadlock Recovery, Operating Systems;

---