Skip to Main content Skip to Navigation
Reports

Two-level checkpointing and partial verifications for linear task graphs

Abstract : Fail-stop and silent errors are unavoidable on large-scale platforms. Efficient resilience techniques must accommodate both error sources. A traditional checkpointing and rollback recovery approach can be used, with added verifications to detect silent errors. A fail-stop error leads to the loss of the whole memory content, hence the obligation to checkpoint on a stable storage (e.g., an external disk). On the contrary, it is possible to use in-memory checkpoints for silent errors, which provide a much smaller checkpoint and recovery overhead. Furthermore, recent detectors offer partial verification mechanisms, which are less costly than guaranteed verifications but do not detect all silent errors. In this paper, we show how to combine all these techniques for HPC applications whose dependence graph is a chain of tasks, and provide a sophisticated dynamic programming algorithm returning the optimal solution in polynomial time. Simulations demonstrate that the combined use of multi-level checkpointing and partial verifications further improves performance.
Complete list of metadatas

https://hal.inria.fr/hal-01216850
Contributor : Equipe Roma <>
Submitted on : Saturday, October 17, 2015 - 6:10:52 PM
Last modification on : Wednesday, February 26, 2020 - 11:14:31 AM
Long-term archiving on: : Thursday, April 27, 2017 - 6:16:17 AM

File

RR-8794.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01216850, version 1

Citation

Anne Benoit, Aurélien Cavelan, Yves Robert, Hongyang Sun. Two-level checkpointing and partial verifications for linear task graphs. [Research Report] RR-8794, ENS Lyon; Inria Grenoble Rhône-Alpes, Université de Grenoble. 2015. ⟨hal-01216850v1⟩

Share

Metrics

Record views

64

Files downloads

37