Deadlock Resolution For Process Graphs
Original Publication Date: 1979-Apr-01
Included in the Prior Art Database: 2005-Feb-20
Given n processes, F1,2,....,nJ, a "process graph" is formed by representing each process as a node and then connecting the nodes according to the input/output relationships among the processes. For simplicity, assume each process produces one output and accepts one or two inputs. Such a process graph is shown in Fig. 1. The node number is used to identify the process (P(i)) and the file (F(i)) it produces. For example, process p(7) produces file F(7), which in turn is an input to P(10) and P(11). The triangle nodes are source data (files). For this discussion the inputs and outputs are considered to be files.