Reliable Error Recovery In Distributed Computer Control Systems
James D. Schoeffler · 2005
The critical nature of computer control applications such as electric power generation and distribution has led to tightly coupled computer systems in which manually. multiple processors and devices may be electronically switched into various configurations for backup purposes. By distributing the tasks into separate but communicating modules, overall system reliability and availability may be achieved with less complex hardware Furthermore, distribution of the time critical tasks leads to less complex software. Error recovery in such a system is complicated not by the distribution of the application tasks but by the distribution of the data base shared among the tasks. Error recovery techniques are discussed, especially their effect an the resulting loading of the communication channels.