Data Driven Device Failure Prediction
Abstract
As society becomes more dependent upon computer systems to perform increasingly critical tasks, ensuring those systems do not fail also becomes more important. Many organizations depend heavily on desktop computers for day to day operations. Unfortunately, the software that runs on these computers is still written by humans and as such, is still subject to human error and consequent failure. A natural solution is to use statistical machine learning to predict failure. However, since failure is still a relatively rare event, obtaining labeled training data to train these models is not trivial. This work presents new simulated fault loads with an automated framework to predict failure in the Microsoft enterprise authentication service and Apache web server in an effort to increase up-time and improve mission effectiveness. These new fault loads were successful in creating realistic failure conditions that are accurately identified by statistical learning models.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 15, 2016
- Accession Number
- AD1017885
Entities
People
- Paul L. Jordan
Organizations
- Air Force Institute of Technology