This study addresses the fundamental approaches and concepts used in the management of time series data generated by sensor-based systems.In such systems, measurements are sometimes produced at regular intervals and sometimes in an intermittent and high-rate manner. Therefore, the processes of data collection, secure transmission, storage in time-oriented structures, and subsequent processing take place through distinct yet interconnected steps. This research is, thus, concerned with demonstrating how these phases work with each other and flows through a single integrated system.
The main reason why time series–oriented databases have gained significant importance in recent years is that traditional database models have certain limitations. In particular, insufficient write speed, low compression capacity, and performance degradation as the scale increases have necessitated the search for more suitable solutions. For this reason, architectural models developed in recent years enable data to be processed at the edge, in intermediate layers, and at the central level. The primary objective of this approach is to reduce latency and alleviate the load on the communication channel. In this way, the operational stability of the system is maintained.
The evaluations conducted clearly demonstrate that time series data management is not limited solely to the storage process. Data processing, detection of anomalous situations, generation of alerts at specific threshold values, and the automatic execution of certain operations have now become a natural part of these systems.An integrated approach in which these features operate together is among the fundamental requirements of modern data flow systems.