SLA guarantee real-time monitoring system with soft deadline constraint

Abstract

Real time monitoring of data streams can be used for monitoring performance, availability or system anomaly detection which can analyze and correlate events from logs file. However, due to the increasing complexity of modern distributed systems, the size of logs can become very large. Monitoring with timing constraint becomes more important in facilitating the enforcement of conditional guarantees. In general, real-time systems are usually classified into soft and hard which need to comply the term of Service Level Agreements (SLAs). In this paper, we focused on the probabilistic deadline in order to guarantee the achievement of SLA deadline for verification of soft deadlines in real-time system by adopt the Central limit theorem in order to approximate the distributions and calculate the probability of a violation of SLA guarantee which can be used to determine the deadline constraint for the system.