Authors

Dr. Samir Jafar

Abstract

 The study is researching the fault tolerance in the large distributed environments such as Grid-Computing, where characterized by the large number of their components and geographic breadth. However all that widening and the huge number of the nodes involved in the system has created great hurdles to achieve dependability in these large environments. So the orientation of this research is to a new mechanism of fault tolerance in the large scale and distributed environments where are based on the principle of rebuilding dataflow graph of the application in order to ensure the continuity of the application as well as to ensure the completion of execution in presence of faults, which resulting from the leave of the nodes and the interruption of the network at a given moment during the execution time

Keywords

grid computing, recovery, checkpointing, parallel programming, macro data flow, work stealing, dependability, fault tolerance