Full Record

New Search | Similar Records

Author
Title Modeling operating system crash behavior through multifractal analysis, long range dependence and mining of memory usage patterns
URL
Publication Date
University/Publisher West Virginia University
Abstract Software Aging is a phenomenon where the state of the operating systems degrades over aperiod of time due to transient errors. These transient errors can result in resourceexhaustion and operating system hangups or crashes.Three different techniques from fractal geometry are studied using the same datasets foroperating system crash modeling and prediction. Holder Exponent is an indicator of howchaotic a signal is. M5 Prime is a nominal classification algorithm that allows predictionof a numerical quantity such as time to crash based on current and previous data. Hurstexponent measures the self similarity and long range dependence or memory of a processor data set and has been used to predict river flows and network usage.For each of these techniques, a thorough investigation was conducted using crash, hangupand nominal operating system monitoring data. All three approaches demonstrated apromising ability to identify software aging and predict upcoming operating systemcrashes. This thesis describes the experiments, reports the best candidate techniques andidentifies the topics for further investigation.
Subjects/Keywords Computer system failures – Mathematical models.; Multifractals.
Country of Publication us
Record ID oai:wvuscholar.wvu.edu:21029
Repository wvu
Date Retrieved
Date Indexed 2015-05-26
.