Ioanna Vavelidou | Searcharxiv

arXiv2026

ITHICA: Intra-Thread Instruction Checking Approach for Defect-Induced Silent Data Corruptions

Hyperscalers are reporting silent data corruptions (SDCs), presumed to be caused by silicon manufacturing defects, as a threat to datacenter reliability. To support datacenter testing efforts to detect defective CPU servers, this paper presents ITHICA, an approach and tool for automatically generating functional tests for defect-induced errors from arbitrary programs by inserting intra-thread, instruction-level error checks, primarily leveraging instruction duplication and output comparison. Our key insight is that the most pernicious defects (those most likely to escape manufacturing testing) cause inconsistent errors: two executions of the same instruction given the same inputs within the same thread can produce different architectural outputs depending on the execution context in which they run. By exploiting this insight, ITHICA uniquely enables arbitrary programs to serve as tests and localizes affected instructions concurrently with error detection. We use ITHICA to transform industrial hyperscaler tests (our baseline), datacenter programs, and common libraries into functional tests, and evaluate them on over 3,000 CPU servers. ITHICA checks detect 39% more defective servers than baseline industrial checks and yield novel findings on defect behavior that challenge conclusions drawn by prior hyperscaler fleet studies.

cs.AR↗

arXiv2021

ML-EXray: Visibility into ML Deployment on the Edge

Benefiting from expanding cloud infrastructure, deep neural networks (DNNs) today have increasingly high performance when trained in the cloud. Researchers spend months of effort competing for an extra few percentage points of model accuracy. However, when these models are actually deployed on edge devices in practice, very often, the performance can abruptly drop over 10% without obvious reasons. The key challenge is that there is not much visibility into ML inference execution on edge devices, and very little awareness of potential issues during the edge deployment process. We present ML-EXray, an end-to-end framework, which provides visibility into layer-level details of the ML execution, and helps developers analyze and debug cloud-to-edge deployment issues. More often than not, the reason for sub-optimal edge performance does not only lie in the model itself, but every operation throughout the data flow and the deployment process. Evaluations show that ML-EXray can effectively catch deployment issues, such as pre-processing bugs, quantization issues, suboptimal kernels, etc. Using ML-EXray, users need to write less than 15 lines of code to fully examine the edge deployment pipeline. Eradicating these issues, ML-EXray can correct model performance by up to 30%, pinpoint error-prone layers, and guide users to optimize kernel execution latency by two orders of magnitude. Code and APIs will be released as an open-source multi-lingual instrumentation library and a Python deployment validation library.

cs.DC↗