My publications trace a research line on AI for system reliability: predicting failures, explaining incidents, and classifying risky operational behavior in large-scale systems.

Featured

A Survey of AIOps Methods for Failure Management

ACM TIST · 2021 · first author

A structured map of AIOps research across monitoring, prediction, diagnosis, and recovery.

Read the paper
Featured

LogRule: Efficient Structured Log Mining for Root Cause Analysis

IEEE TNSM · 2023 · first author

An interpretable structured-log RCA method with strong diagnosis performance and substantially faster runtime.

Read the paper
AI-based Proactive Failure Management in Large-scale Cloud Environments
PhD thesis · 2024

Paolo Notaro

AIOps methods for proactive failure management across infrastructure, platform, and software layers.

Read thesis
Command-line Risk Classification using Transformer-based Neural Architectures
arXiv · 2024

Paolo Notaro, Soroush Haeri, Jorge Cardoso, Michael Gerndt

LLM-based classification of risky CLI commands for operational security workflows.

Read preprint
An Optical Transceiver Reliability Study based on SFP Monitoring and OS-level Metric Data
CCGRID · 2023

Paolo Notaro, Soroush Haeri, Qiao Yu, Jorge Cardoso, Michael Gerndt · Best Paper nominee

Failure-rate analysis and predictive signals for optical transceiver reliability in cloud datacenters.

Read paper
HiMFP: Hierarchical Intelligent Memory Failure Prediction for Cloud Service Reliability
DSN · 2023

Qiao Yu, Wengui Zhang, Soroush Haeri, Paolo Notaro, Jorge Cardoso, Odej Kao

Memory failure prediction using spatial and temporal correctable-error information.

Read paper
A Systematic Mapping Study in AIOps
ICSOC Workshop · 2020

Paolo Notaro, Jorge Cardoso, Michael Gerndt

An early taxonomy and mapping of AIOps contributions, with failure-related tasks as the dominant theme.

Read paper
Radar Emitter Classification with Attribute-specific Recurrent Neural Networks
arXiv · 2019

Paolo Notaro, Magdalini Paschali, Carsten Hopke, David Wittmann, Nassir Navab

Sequence modeling for radar pulse streams using attribute-specific recurrent processing.

Read preprint