Towards Data-Driven Autonomics in Data Centers

TL;DR

This paper demonstrates how data-driven predictive models, built from large-scale data center logs, can forecast node failures with reasonable accuracy, paving the way for autonomous management of data centers.

Contribution

It introduces a methodology for building and evaluating a predictive failure model using ensemble classifiers on a large Google dataset, advancing autonomous data center management.

Findings

Achieved true positive rates between 27% and 88% at 5% false positive rate.

Predicted failures within a 24-hour window using ensemble classifiers.

Provided publicly available scripts for data processing and modeling.

Abstract

Continued reliance on human operators for managing data centers is a major impediment for them from ever reaching extreme dimensions. Large computer systems in general, and data centers in particular, will ultimately be managed using predictive computational and executable models obtained through data-science tools, and at that point, the intervention of humans will be limited to setting high-level goals and policies rather than performing low-level operations. Data-driven autonomics, where management and control are based on holistic predictive models that are built and updated using generated data, opens one possible path towards limiting the role of operators in data centers. In this paper, we present a data-science study of a public Google dataset collected in a 12K-node cluster with the goal of building and evaluating a predictive model for node failures. We use BigQuery, the big data SQL platform from the Google Cloud suite, to process massive amounts of data and generate a rich feature set characterizing machine state over time. We describe how an ensemble classifier can be built out of many Random Forest classifiers each trained on these features, to predict if machines will fail in a future 24-hour window. Our evaluation reveals that if we limit false positive rates to 5%, we can achieve true positive rates between 27% and 88% with precision varying between 50% and 72%. We discuss the practicality of including our predictive model as the central component of a data-driven autonomic manager and operating it on-line with live data streams (rather than off-line on data logs). All of the scripts used for BigQuery and classification analyses are publicly available from the authors' website.