Mainz 2014 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

T: Fachverband Teilchenphysik

T 35: GRID-Computing

T 35.1: Talk

Monday, March 24, 2014, 16:45–17:00, P15

The Smart Grid Monitoring System - ANFIS — Gen Kawamura, •Erekle Magradze, Haykuhi Musheghyan, Jordi Nadal, and Arnulf Quadt — II. Physikalisches Institut, Georg-August Universität Göttingen, Friedrich-Hund-Platz 1 37077 Göttingen Deutschland

The demand on the various digital services is increasing and therefore the stable performance of the computing infrastructure is playing a critical role in the robustness of the provided services. Stability and high availability of the high performance computing infrastructure is achievable by an efficient administration, in time failure detection, the proper root cause analysis of the failures and finally by the proper action taking. All these tasks are mainly performed by systems administrators, which are widely using different monitoring and automated action taking systems. These systems were developed and continue evolving to help in the process of administration. The disadvantage of the monitoring or automated action taking tools is an absence of analytic properties, such as autonomous root cause analysis of the service degradation or short term predictions of a failure of the service or group of services. This fact served as a motivation to develop an Artificial Intelligence methods for development of the smart monitoring system. Latest results of the smart monitoring system are presented.