Centralized computing systems have been commonplace for a long time. In a centralized system, many endpoints connect to one central point, which processes and connects them to each other. In recent years we have seen the rise of distributed computing, which is where there is no central hub, and each connection point interacts directly with each other, creating an interconnected network. For example, banks’ account ledgers operate on a centralized computing system, while bitcoin’s peer-to-peer ledger is a distributed computing system.
Centralized systems have several advantages and disadvantages: all of the information is in one place, and with powerful hardware you can process the information effectively. However, if the load is too heavy, or if there is a hardware issue, the central hub can go down, severing the connection to all the devices on the network. Think of what happens when a wifi network crashes or when there are too many connections and the wifi becomes slow. Centralized systems are also great targets for hackers — all of the information is in one place, and all that is needed for a major breach is to infiltrate a single point. Additionally, there can be incompatibility issues between one central system and other devices or add-ons, especially with proprietary systems.
Distributed networks are not subject to many of these difficulties. Because the system is not centralized, it is much more difficult for hackers as they must now hack a multitude of devices. Often, these devices will not even contain useful information, since information is usually widely dispersed on distributed networks. Additionally, distributed networks are not dependent upon a single point because the interconnectedness of the distributed network inherently protects it; if one point goes down, the network will operate around it. Another advantage of distributed networks is that they can often support a large amount of devices. As you increase the number of devices you create a larger network without increasing the stress on any one point.
Distributed computing systems are becoming more widespread, but are still evolving. In 2017, Bluetooth® released Bluetooth® mesh, a distributed network system that uses Bluetooth® Low Energy to connect devices. These networks are at a distinct advantage when compared to other solutions that use centralized systems. The network is not at risk of going down due to a failure at a single point, and is more secure. Additionally, the network will not suffer any increased strain from adding devices, since new devices simply add to the network. BubblyNet is an example of a system that is operating on the Bluetooth® mesh network, focusing on building automation.