Near-zero Downtime Recovery from Transient-error-induced Crashes

TL;DR

IterPro is a lightweight, compiler-assisted resilience technique that enables near-zero downtime recovery from transient errors in HPC systems by repairing process states on-the-fly with minimal overhead.

Contribution

It introduces IterPro, a novel compiler-based method with new code transformations for fast, accurate recovery from transient errors in high-performance computing.

Findings

Recovers 83.55% of crash errors within dozens of milliseconds

Incurs almost zero runtime overhead during normal execution

Uses small, fixed 27MB memory overhead

Abstract

Due to the system scaling, transient errors caused by external noises, e.g., heat fluxes and particle strikes, have become a growing concern for the current and upcoming extreme-scale high-performance-computing (HPC) systems. However, since such errors are still quite rare as compared to no-fault cases, desirable solutions call for low/no-overhead systems that do not compromise the performance under no-fault conditions and also allow very fast fault recovery to minimize downtime. In this paper, we present IterPro, a light-weight compiler-assisted resilience technique to quickly and accurately recover processes from transient-error-induced crashes. IterPro repairs the corrupted process states on-the-fly upon occurrences of errors, enabling applications to continue their executions instead of being terminated. IterPro also exploits side effects introduced by induction variable based code optimization techniques to improve its recovery capability. To this end, two new code transformation passes are introduced to expose the side effects for resilience purposes. We evaluated IterPro with 4 scientific workloads as well as the NPB benchmarks suite. During their normal execution, IterPro incurs almost zero runtime overhead and a small, fixed 27MB memory overhead. Meanwhile, IterPro can recover on an average 83.55% of crash-causing errors within dozens of milliseconds with negligible downtime. With such an effective recovery mechanism, IterPro could tremendously mitigate the overheads and resource requirements of the resilience subsystem in future extreme-scale systems.