BEGIN:VCALENDAR
PRODID:-//Microsoft Corporation//Outlook MIMEDIR//EN
VERSION:1.0
BEGIN:VEVENT
DTSTART:20131121T213000Z
DTEND:20131121T214500Z
LOCATION:601/603
DESCRIPTION;ENCODING=QUOTED-PRINTABLE:ABSTRACT: In modern cloud computing systems, hundreds and even thousands of cloud servers are interconnected by multi-layer networks. In such large-scale and complex systems, failures are common. Autonomic failure detection and diagnosis are crucial technologies for understanding emergent, cloud-wide phenomena and self-managing resource burdens for cloud availability and productivity enhancement. They are determining the quality of the services that cloud could provide to the customers. In my PhD dissertation, I am focusing on developing an autonomic failure management mechanism with the assurance of dependability to the cloud services. My work first sets out by studying the health-related performance metric. With metric selection and metric extraction technologies, failures related metrics are selected. Then learning from the characteristics of metrics from time domain and frequency domain separately, failure detection and diagnosis are achieved dynamically.
SUMMARY:Autonomic Failure Identification and Diagnosis for Building Dependable Computing Systems
PRIORITY:3
END:VEVENT
END:VCALENDAR