IOT SİSTEMLERİ İÇİN VERİTABANI SİSTEM ÖNERİLERİ

Abstract

Internet of Things (IoT) is an interconnection of different types of information assets in which data is continuously generated and transmitted over the Internet. Technologies of the sensor, RFID, GPS, mobile devices, and Internet-enabled actuators play a significant role in IoT systems. IoT brings out new challenges in terms of data and information management because it is not easy to collect and manage a large amount of heterogeneous data that is aggregated at very high velocity as well as to retrieve and manage the information that is hidden within this large volume of data. In this paper, I discuss the main factors affecting the efficiency of data management in IoT systems, specifically query processing and transaction management. There are many lessons learned from traditional database systems, distributed systems, and sensor networks, however, traditional solutions are often inadequate to meet the needs of applications in such a complex ecosystem, namely IoT. In traditional database systems, for instance, query operations are usually local, and execution costs depend on the current processor power and other resource constraints (i.e. memory). On the other hand, transaction management mechanisms guarantee the ACID properties in order to provide overall data integrity. It is apparent that different types of IOT applications that operate on heterogeneous, streaming, real-time, and geographically distributed large data will significantly change the well-known aspects of querying and transaction management. Context-aware querying, distributed querying, MapReduce computing model and flexible transaction models such as web-based transaction handling are some of the current issues discussed in this paper. With the succinct yet comprehensive information presented in this work, I intend to provide a guide for researchers in the IoT systems, especially in the context of database systems.