Tuesday 26 May 2020 9:20 am

Minimum Effective Blockchain

Many enterprise IT teams feel there is no justifiable use case for blockchain. At the other end of the spectrum are those who believe in storing every last bit, byte and nibble of data on the blockchain. Neither of these is overly helpful in supporting the adoption of blockchain networks in industry.

A foundational premise of applying blockchain to solve business problems is Minimum Effective Blockchain (MEB).

Perfection is achieved, not when there is nothing more to add, but when there is nothing left to take away.

Antoine de Saint-Exupéry

Tim Ferriss, in his book The 4-Hour Body, refers to the principle of Minimum Effective Dose (MED). The definition is straightforward: The MED is the smallest dose that will produce the desired outcome. Of course, in Tim’s context, it was related to sports nutrition and supplements. This principle can also apply in the context of blockchain.

Minimum Effective Blockchain (MEB) is the smallest combination of data and nodes to create value for the broadest range of stakeholders.

Minimum Effective Blockchain (MEB) is the smallest combination of data and nodes to create value for the broadest range of stakeholders. 

Good data architecture design is critical. Blockchain networks are not designed for storing large amounts of data. Neither are they intended for storing high volumes of transactions. Blockchain networks do not perform as efficiently as relational databases. The data stored in the blockchain should be the minimum set of information necessary to ensure trust between counterparties and satisfy the requirements around tracking and tracing assets.

And what of the rest of the data?

A blockchain project will often require more data than that which is on the blockchain. The additional data may exist in a distributed database or filesystem. These data stores can be designed to prevent unauthorised access using encryption and other established technologies. 

In these cases, the information stored on the blockchain can contain time-stamped records of the original data. Any tampering with the data becomes immediately evident. The blockchain can also store records of access to data and granting/revoking access to information. Inevitably, the distributed data will have a central point of control. However, so long as the underlying blockchain is decentralised, counterparties can be confident that the data integrity is maintained.

What is the minimum number of nodes necessary to ensure the integrity of the network?

With the data architecture planned, what is the minimum network architecture required? Or, to put it another way, what is the minimum number of nodes necessary to ensure the integrity of the network? (A node in a blockchain network is a computer which contains a copy of the blockchain and participates in the process of adding new data to the blockchain)

A vital aspect of the Bitcoin network’s security is a large number of nodes. As a public permissionless blockchain, the Bitcoin network has an estimated 10,000 nodes. To compromise the network, an attacker must take control of more than half (the 51% attack problem). 

In enterprise blockchain projects, there are a range of additional factors to consider.

The minimum number of nodes necessary for network integrity varies based on how the network achieves consensus related to adding data to the blockchain. The Bitcoin protocol uses something called Proof of Work. A large number of nodes is critical for this protocol. Other protocols like Hyperledger Fabric, use a different consensus mechanism and can operate with only a few nodes. 

A further consideration for enterprise or industry blockchains is private versus public. If a blockchain network is private, then all members are known to each other. The closed nature of the network affords a trust which arises from every member knowing every other. Any node that attempts to behave like a bad actor is caught out and potentially thrown off the network. 

Operating and maintaining a blockchain node is one of the costs of the network. It is not likely that every member will want to run their node. Many smaller stakeholders will have neither the competence nor the desire to run their nodes. The decisions of who will run the nodes – and how the cost will be absorbed – is part of the design.

In applying the MEB principle to enterprise blockchain projects, determine the minimum number of nodes necessary for the network to be fully trusted and to operate securely.

As with many things, the MEB approach – if taken to an extreme –  can go too far.

As with many things, the MEB approach – if taken to an extreme –  can go too far. In the case of blockchain it is possible to reduce the number of nodes and data to such a small amount that it is no longer blockchain – but only a DLT (Distributed Ledger Technology approach) – or possibly not even DLT. If stripped back far enough, the solution has value only to one stakeholder. And for that, you can use a database.

Today’s primary objections to the use of blockchain technology focus on issues of scalability, performance and transparency. These objections are genuine and can happen with flawed blockchain design from the start. 

When looking at how to solve a multi-enterprise industry problem with blockchain technology, begin by deciding the minimum amount of information which must exist on a blockchain network that is distributed and decentralised where data is immutable. Then look at what data should reside in a decentralised filesystem or database with traditional technology solutions for scalability, performance and access control. 

Not everything belongs on the blockchain. Not every stakeholder should run a node. Blockchain cannot solve every problem. To avoid problems in the future, apply the principle of Minimum Effective Blockchain (MEB) to the design in the beginning.

Get in touch with us info@blockchainrookies.com / Twitter @igetblockchain