Title :
EVEREST: an event recognition testbed
Author :
Spezialetti, Madalene ; Bernberg, Scott
Author_Institution :
Dept. of EECS, Lehigh Univ., Bethlehem, PA, USA
Abstract :
Due to their fundamental nature, analyzing the behavior of distributed computations is a complex task. One approach is to monitor the system activity for occurrences of user defined events. As the computation executes, the monitoring system automatically collects and evaluates information pertaining to the defined events, and recognizes event occurrences. The need exists to study various approaches to event recognition in a context in which the strategies can be created and analyzed. This paper presents an EVEnt REcognition teSTbed (EVEREST) specifically designed to address these needs. Commands are provided by which the user can define the behavior to be recognized the configuration of the monitoring system and the strategies to be used in recognizing the event occurrences. Utilizing these directives, EVEREST recognizes occurrences of the defined events during the distributed program´s execution. By providing flexible structuring and dynamic configuration of the monitoring system, multiple time view protocols and the separation of event and monitoring specification, EVEREST provides the capability of developing, testing and comparing various approaches to distributed event recognition
Keywords :
program testing; system monitoring; EVEREST; configuration; distributed computations; distributed event recognition; distributed program´s execution; event occurrences; event recognition testbed; monitoring system; user defined events; Clocks; Computerized monitoring; Concurrent computing; Data analysis; Distributed computing; Event detection; Information analysis; Protocols; System testing;
Conference_Titel :
Distributed Computing Systems, 1995., Proceedings of the 15th International Conference on
Conference_Location :
Vancouver, BC
Print_ISBN :
0-8186-7025-8
DOI :
10.1109/ICDCS.1995.500042