Section: New Results

Checkpointing workflows for fail-stop errors

Participants : Louis-Claude Canon, Henri Casanova [University of Hawai`i at Manoa, USA] , Li Han, Yves Robert, Frédéric Vivien.

We consider the problem of orchestrating the execution of workflow applications structured as Directed Acyclic Graphs (DAGs) on parallel computing platforms that are subject to fail-stop failures. The objective is to minimize expected overall execution time, or makespan. A solution to this problem consists of a schedule of the workflow tasks on the available processors and of a decision of which application data to checkpoint to stable storage, so as to mitigate the impact of processor failures. For general DAGs this problem is hopelessly intractable. In fact, given a solution, computing its expected makespan is still a difficult problem. To address this challenge, we consider a restricted class of graphs, Minimal Series-Parallel Graphs (GSPGs). It turns out that many real-world workflow applications are naturally structured as GSPGs. For this class of graphs, we propose a recursive list-scheduling algorithm that exploits the GSPG structure to assign sub-graphs to individual processors, and uses dynamic programming to decide which tasks in these sub-gaphs should be checkpointed. Furthermore, it is possible to efficiently compute the expected makespan for the solution produced by this algorithm, using a first-order approximation of task weights and existing evaluation algorithms for 2-state probabilistic DAGs. We assess the performance of our algorithm for production workflow configurations, comparing it to (i) an approach in which all application data is checkpointed, which corresponds to the standard way in which most production workflows are executed today; and (ii) an approach in which no application data is checkpointed. Our results demonstrate that our algorithm strikes a good compromise between these two approaches, leading to lower checkpointing overhead than the former and to better resilience to failure than the latter. To the best of our knowledge, this is the first scheduling/checkpointing algorithm for workflow applications with fail-stop failures that considers workflow structures more general than mere linear chains of tasks.

This work is available as a research report and has been submitted to a journal. A short version appears in the proceedings of the IEEE Cluster'17 conference.