Centralized/Decentralized Systems

Centralized systems

From the beginning of the era of computing, centralization was the most preferred strategy in software system design as it comes up with a provided possibility of control and authority over the individuals and used resources. Consequently, all the individual components are connected to one central element that controls the whole system. The centralized approach is typically implemented with a server as a central node with low-end computers as client terminals.

Definition

Centralized systems are systems where a single authority controls the system and is exclusively responsible for all operations on the system (Imran Bashi (2017)Imran Bashir. Mastering Blockchain. Packt Publishing, Limited, 2017.).

Asharaf et al. (2017)S. Asharaf and S. Adarsh. Decentralized Computing Using Blockchain Technologies and Smart Contracts: Emerging Research and Opportunities: Emerging Research and Opportunities. Advances in Information Security, Privacy, and Ethics. IGI Global, 2017. refer to centralized computing as the “allocation of all computing/storage resources to a single unit which controls and facilitates all the computing/storage services for the organization”. This centralized computing paradigm relies on the processing at the central server and all client terminals can act as clients with having very limited or no computing capability and rights, and request services from the server. This ensures data integrity and avoids data redundancy.

But this approach leads to problems since all data passes through a central node, hence the server can become a bottleneck as the data volume grows with the number of users and applications. Moreover, the failure of the central server can make the entire system inoperable. Thus, a centralized network is highly vulnerable as the destruction of the single central node undermines communication between the clients. This gives motivation to the idea of decentralization.