Title :
On the complexity of scheduling checkpoints for computational workflows
Author :
Robert, Yannick ; Vivien, F. ; Zaidouni, Dounia
Author_Institution :
INRIA, Ecole Normale Super. de Lyon, Lyon, France
Abstract :
This paper deals with the complexity of scheduling computational workflows in the presence of Exponentially distributed failures. When such a failure occurs, rollback and recovery is used so that the execution can resume from the last checkpointed state. The goal is to minimize the expected execution time, and we have to decide in which order to execute the tasks, and whether to checkpoint or not after the completion of each given task. We show that this scheduling problem is strongly NP-complete, and propose a (polynomial-time) dynamic programming algorithm for the case where the application graph is a linear chain. These results lay the theoretical foundations of the problem, and constitute a prerequisite before discussing scheduling strategies for arbitrary DAGS of moldable tasks subject to general failure distributions.
Keywords :
checkpointing; computational complexity; dynamic programming; failure analysis; graph theory; processor scheduling; NP-complete; application graph; arbitrary DAGS; computational workflows scheduling complexity; dynamic programming algorithm; execution time; exponentially distributed failures; failure distributions; linear chain; moldable tasks; rollback; scheduling checkpoints complexity; scheduling strategies; Checkpointing; Complexity theory; Dynamic programming; Equations; Mathematical model; Parallel processing; Program processors;
Conference_Titel :
Dependable Systems and Networks Workshops (DSN-W), 2012 IEEE/IFIP 42nd International Conference on
Conference_Location :
Boston, MA
Print_ISBN :
978-1-4673-2264-5
Electronic_ISBN :
978-1-4673-2265-2
DOI :
10.1109/DSNW.2012.6264675