Near Real Time Log Aggregation and Analysis in Distributed Systems

U ovom diplomskom radu opisane su funkcionalnosti i arhitekture postojećih sustava za prikupljanje, obradu, spremanje i analizu dnevničkih zapisa u skoro realnom vremenu. Također, osmišljen je prototip sustava koji bi uz pomoć Apache Flinka imao iste funkcionalnosti. Međutim, pokazalo se kako takav sustav ne bi bio održiv u slučaju Edgara. Na koncu, informacijski sustav Edgar je prilagođen kako bi se uz njega instalirao Elastic Stack za bolje praćenje dnevničkih zapisa iz cijelog Edgarovog sustava.This graduation work describes the functionalities and architectures of existing systems for collecting, processing, storing and analyzing logs in near real time. Also, a prototype system was designed that would have the same functionalities with the help of Apache Flink. However, it turned out that such a system would not be sustainable in the case of Edgar. Finally, the Edgar information system has been customized to install the Elastic Stack to better track logs from across the entire Edgar system