Algorithm-Based Fault Tolerance for Parallel Stencil Computations

Abstract

The increase in HPC systems size and complexity, together with increasing on-chip transistor density, power limitations, and number of components, render modern HPC systems subject to soft errors. Silent data corruptions (SDCs) are typically caused by such soft errors in the form of bit-flips in the memory subsystem and hinder the correctness of scientific applications. This work addresses the problem of protecting a class of iterative computational kernels, called stencils, against SDCs when executing on parallel HPC systems. Existing SDC detection and correction methods are in general either inaccurate, inefficient, or targeting specific application classes that do not include stencils. This work proposes a novel algorithm-based fault tolerance (ABFT) method to protect scientific applications that contain arbitrary stencil computations against SDCs. The ABFT method can be applied both online and offline to accurately detect and correct SDCs in 2D and 3D parallel stencil computations. We present a formal model for the proposed method including theorems and proofs for the computation of the associated checksums as well as error detection and correction. We experimentally evaluate the use of the proposed ABFT method on a real 3D stencil-based application (HotSpot3D) via a fault-injection, detection, and correction campaign. Results show that the proposed ABFT method achieves less than 8% overhead compared to the performance of the unprotected stencil application. Moreover, it accurately detects and corrects SDCs. While the offline ABFT version corrects errors more accurately, it may incur a small additional overhead than its online counterpart.

Resiliency in numerical algorithm design for extreme scale simulations

“…ABFT concepts have been extended to process failures for a wide range of matrix operations both for detection and mitigation purposes (Bosilca et al, 2009; Chen and Dongarra, 2008; Du et al, 2012; Jia et al, 2013; Kim et al, 1996) and general communication patterns (Kabir and Goswami, 2016). ABFT has also recently been proposed for parallel stencil-based operations to accurately detect and correct silent data corruptions (Cavelan and Ciorba, 2019). In these scenarios the general strategy is a combination of checkpointing and replication of checksums.…”