Applying Machine Learning to Study Infrastructure Anomalies in a Mid-size Data Center -- Preliminary Considerations

Date

2021-01-05

Contributor

Advisor

Department

Instructor

Depositor

Speaker

Researcher

Consultant

Interviewer

Narrator

Transcriber

Annotator

Journal Title

Journal ISSN

Volume Title

Publisher

Volume

Number/Issue

Starting Page

218

Ending Page

Alternative Title

Abstract

Today, data centers deal with fast growing data volumes. To deliver services, they deploy growing amount of heterogeneous hardware. As a result, it becomes practically impossible to apply human-based data center management. For instance, in a real-world data center, with 500+ computers, delivering data, computational, and network services, it becomes impossible to visualize, and understand, causal relationships among variables describing performance of monitored resources. However, it is possible to collect data describing behavior of individual nodes. Hence, such data may be used to analyze/model system performance. In particular, it may be applied to recognize and predict anomalies in system behavior. Furthermore, collected data should allow finding the cause(s) of anomalies. Therefore, “data-driven approaches” have been applied to the real-world data, to find, so called, Root Cause of anomalies.

Description

Keywords

Business Intelligence and Big Data for Innovative and Sustainable Development of Organizations, anomaly detection, causality, data center, root cause analysis, time series

Citation

Extent

10 pages

Format

Geographic Location

Time Period

Related To

Proceedings of the 54th Hawaii International Conference on System Sciences

Related To (URI)

Table of Contents

Rights

Attribution-NonCommercial-NoDerivatives 4.0 International

Rights Holder

Local Contexts

Email libraryada-l@lists.hawaii.edu if you need this content in ADA-compliant format.