Global stateA global state of a distributed system consists of a collection of the local states of processes in that system along with the states of the communication channels that connect these processes. Global state may be used to enquire and determine system status such as a deadlock or whether all distributed computations have concluded. Distributed snapshot algorithms provide a snapshot of a global state of the system. Termination detection algorithms are commonly used to determine whether a distributed computation has terminated in all nodes of the system. Detection of a deadlock in a distributed system is needed to recover and resume computation. We review fundamental distributed snapshot, termination detection and deadlock detection algorithms and implement them in Python using mpi4py in this chapter.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Global State Analysis

  • K. Erciyes

摘要

Global stateA global state of a distributed system consists of a collection of the local states of processes in that system along with the states of the communication channels that connect these processes. Global state may be used to enquire and determine system status such as a deadlock or whether all distributed computations have concluded. Distributed snapshot algorithms provide a snapshot of a global state of the system. Termination detection algorithms are commonly used to determine whether a distributed computation has terminated in all nodes of the system. Detection of a deadlock in a distributed system is needed to recover and resume computation. We review fundamental distributed snapshot, termination detection and deadlock detection algorithms and implement them in Python using mpi4py in this chapter.