The Importance of Decentralisation in Smart Cities

What is decentralisation? What does it have to do with smart cities? Is it important and if so, why?

What is decentralisation?

Although I’ve looked at tons of definitions, I’ll briefly highlight the two I felt were most relevant — the Cambridge dictionary defines it as: “to move the control of an organisation or government from a single place to several smaller ones”, but my personal favourite has to be Merriam-Webster’s which is “the dispersion or distribution of functions and powers”. While the context of both definitions seems to gravitate towards government/institutions, the definition of the latter can be extended to fit other scenarios such as computing. A great example of this is Blockchain; a decentralised network in which computing power, data and storage is distributed to prevent hijacking and malicious use of the network.

Decentralisation in computer/digital systems typically takes place through distribution of resources such as processing power, data storage or network bandwidth etc depending on the system function or overall goal.

What does it have to do with smart cities?

Well, in theory, smart cities will generate massive data footprints. Sensors will be embedded in almost any and every item; from toilet seats and roads to fridges and postboxes — everything, literally every damn thing, will effectively become a data point. This in itself is a problem. Every datapoint will generate data and eventually all this big data will need to be stored and processed. This may present an even bigger problem.

In a system with, potentially, tens of millions of distinct data points generating hundreds of terabytes of data; decentralisation could be key. Storing all this data in a single place (i.e. a centralised system) is a major risk as it threatens data security, integrity and availability. A single breach would compromise all captured data. It would would also mean that a system has a single point of failure.

CIA Triad

Decentralisation can overcome the storage problems centralised systems face by dispersing captured data to multiple points instead of a single point.

This provides a critical boost to data integrity. For instance, let’s assume that we have a system S1 where the same data is captured and dispersed to 5 locations simultaneously and we also have a system S2 where the data is captured and stored directly in 1. For a breacher to manipulate data in S1, they would need to gain majority power in the network i.e. they would need to gain access and manipulate the data in 3 out of 5 locations simultaneously to bring the system to a ‘consensus’. While in S2, as mentioned previously, a single breach would compromise the entire system. Aside from this, there would be cryptographically secure protocols in place to further secure data in both systems. The conclusion is data manipulation becomes more infeasible as the network grows larger and data integrity is made stronger in S1. The ‘consensus model’ aims to deliver a system that provides true equity.

Confidentiality of data can be boosted by designing, developing and applying anonymity algorithms to data. Data availability depends on ‘who’ has access to ‘what’ data and overall data transparency. A public system in which everybody can read but not write data might not be appealing in the health sector but might be encouraged in the public services to see how taxes are being spent.

In order to process the vast amounts of data captured by smart cities, a decentralised approach should be encouraged for fast, scalable and fault tolerant data processing. By distributing compute power resources, excessive pressure is taken off multiple systems, such as servers and even the national grid, and this provides a more efficient ecosystem. For example, if smart city M needed to process some non-sensitive data but didn’t have enough compute resources, then they could share the compute resources of neighbouring smart city N.

Is Decentralisation important in Smart Cities?

Imagine how effective the emergency services or public transport systems would be if they made use of a decentralised system. No wasting of time with bulky and lengthy processes/red tape. Communication between the police, fire brigade and NHS would be transparent and instantaneous — cutting down casualties from any incident and eventually leading to better standards of living for all inhabitants of any smart city.

A decentralised system would also mean decommissioning of old legacy systems. There would be massive savings across the board; only one system needs to be maintained, no need to train staff on using 49,252,120 different systems, security and compliance will be improved as it’s focused on a single system. This would result in faster, improved services and happier citizens.

Decentralisation could also encourage more transparent politics. By making all data public and available to every citizen, politicians will be made accountable and can be challenged for any unreasonable expenses incurred using the the taxpayers money.

Decentralisation could make the entire smart city experience seamless and seems to be the best way to push forward and develop the cities of tomorrow.