Measurement-based study of failures in a UNIX server /

analysis is also performed to identify server modules

Bibliographic Details
Main Author: Lal, Ronjeet Singh, 1970-
Format: Thesis eBook
Language:English
Published: [Place of publication not identified] : [publisher not identified] ; 1998.
Subjects:
Online Access:Link to OAKTrust copy
Description
Summary:analysis is also performed to identify server modules
analyzed to identify error patterns which may be used
basis. Message logs spanning approximately eleven
characterize the dependability of the system.
Component
data are classified and categorized to calculate
measurement data from the event logs.
months were collected for this study. The event log
parameters such as MTBF and availability to
reported on computer systems are compared to the
results of fault-injection experiments previously
server are collected to form a dependability data
study of a UNIX server. The event logs of a UNIX
system error activity preceding each system failure is
that are prone to errors in the system. Next, the
This thesis presents a measurement-based dependability ics.
to predict future failures. Lastly, the error/failure
Item Description:"Major subject: Electrical Engineering".
Vita.
Physical Description:viii, 43 leaves : illustrations ; 28 cm.
Also available online.
Bibliography:Includes bibliographical references: pages 39-42.