Detecting silent data corruptions in the wild

Silent Errors within hardware devices occur when an internal defect manifests in a part of the circuit which does not have check logic to detect the incorrect circuit operation. The results of such a defect can range from flipping a single bit in a single data value, up to causing the software to execute the wrong instructions. Silent data corruptions (SDC) in hardware impact computational integrity for large-scale applications. Manifestations of silent errors are accelerated by datapath variations, temperature variance, and age, among other silicon factors. These errors do not leave any record or trace in system logs. As a result, silent errors stay undetected within workloads, and their effects can propagate across several services, causing problems to appear in systems far removed from the original defect. In this paper, we describe testing strategies to detect silent data corruptions within a large scale infrastructure. Given the challenging nature of the problem, we experimented with different methods for detection and mitigation. We compare and contrast two such approaches - 1. Fleetscanner (out-of-production testing) and 2. Ripple (in-production testing).We evaluate the infrastructure tradeoffs associated with the silicon testing funnel across 3+ years of production experience.