Private versus Public versus Consortium Blockchains
The subject of a perpetual debate is why public blockchain is a terrible idea because it is slow and cannot confirm so many transactions per second. Another side of the debate is that the private blockchain is nothing but a glorified database. Therefore, consortium blockchain is a compromise between the two. These debates are based on technical attributes distinguishing between the three blockchain types.
In addition to their technical dissimilarities, you should know why one is more suitable than the other for the application or use case one is intending to construct. The thought process here is that a thorough business case and pain points analysis should be used to define which style of blockchain to implement. Otherwise, you might be better off with a centralized database.
In simple terms, private blockchain restricts access to its network of nodes and decides who can view and transact within the network. Access control is the biggest differentiator between private blockchain and public ones. Access to blockchain is restricted to certain individuals or entities, and mining is performed on a limited set of nodes. As shown in Figure 5.1, there may be one or more validator nodes. Hence, the deployment footprint is small compared with a public blockchain.
Figure 5.1 Limited number of mining and validator nodes with access to white-listed participants
Private blockchain is not decentralized from the viewpoint of both participation and governance. Also, mining rewards are not necessary because the network is controlled and managed by a single trusted entity or authority. Mining to reach consensus only guarantees to the participants that transactions have not been tampered with.
Use of private blockchains gained traction because of the needs of enterprises. They expressed the need to manage access to transactions and the underlying ledger. We have seen many implementations of private blockchains in which one company tells its vendors and subcontractors that they must submit transactions to an application running on private blockchain in order to achieve some objectives such as provenance and traceability.
Public versus Private Blockchains
In a public blockchain, anyone can read and add transactions if they are valid against the underlying protocols. In addition, anyone can participate in the consensus process of determining what blocks get added to the chain. Hence, these blockchains are global, decentralized, and censorship resistant. Also, there is less expectation of confidentiality with public blockchain.
Private blockchains then entered the scene, reading and adding transactions to an underlying ledger performed by a single entity.1 Hence, that entity can alter transactions without requiring consensus with others who are participating in reading and writing transactions.
Benefits of private blockchain are speed of transactions and higher levels of confidentiality because it doesn’t require consensus between multiple independent nodes. A primary weakness is that there are also no validator or mining nodes.2 Speed of transactions is higher compared with public blockchain because a small number of nodes is enough to maintain the state of the ledger. Another benefit that proponents of private blockchains promote is that they provide privacy, unlike public blockchains. On the other hand, it is not privacy that is the question, but rather permissions to read and write transactions. Public blockchains do not necessarily reveal confidential information but, rather, who added the transaction and who transacted with whom.
Blockchain maximalists argue that private blockchains are nothing but a glorified database with cryptography because an implementing agency can dictate the terms of using the system and, in some cases, force other participants to use it. For example, a large retailer requires its vendors to use the system if they want to remain a vendor. It can be easily argued, “Why does the retailer need a blockchain instead of a centralized database?” Hence, a need for blockchain for this kind of vertical collaboration is questionable.
Debate about the Private Blockchain
Enterprise communities, especially IBM and banks, have been pioneering the adoption of private blockchain in a variety of use cases through well-known proof of concepts. Private blockchains are often presented as an alternative to public blockchain for enterprise use cases that require access, control, and privacy. Truths are revealed, but so are misconceptions because private blockchains by themselves do not have privacy built in.
Privacy is added to it using access control, by which only individuals with proper credentials can see the transactions added to blocks that are typically deployed in a controlled cloud server environment. But once somebody gets into the system and because there are a small number of nodes, tamper resistance of the private blockchain can be very quickly compromised. In a use case where a private blockchain is being shared with multiple known and trusted parties, you wouldn’t put information sensitive to your organization or to individuals in the organization. The same applies to public blockchain. Instead, sensitive information is stored off-chain.
The other argument presented about private blockchain’s benefits over public blockchain is transaction throughput or transactions per second (TPS). Yes, it is obvious that private blockchains can process more TPS because their mining, or census, process is faster owing to the small number of nodes and “miner” nodes that do not compete for block rewards.
High TPS is a requirement for use cases with high-frequency transaction processing such as real-time stock trades. However, there are use cases that do not require a high TPS, such as real-estate transfers and international supply chains, where transaction settlements that take minutes or hours to complete are acceptable.
Use of private blockchain for vertical collaboration use cases is weak. For example, you are a big company that buys billions of dollars worth of auto parts from dozens of independent vendors, and you want to ensure parts can be tracked and traced for provenance, recalls, and so forth. You go and tell all your vendors they must use this new system on private blockchain. What is the merit of using blockchain in this environment other than a centralized cloud database? You’ve already forced your suppliers to use it. The vendors must trust that you will not tamper with the ledger (Figure 5.2).
Figure 5.2 Centralized database versus private blockchain
In this example, the blockchain that you own, deploy, and maintain is no different than a trusted central database. The blockchain, in this case, cannot function as a trust layer when suppliers are forced to trust; otherwise they’ll lose business.
Hence, private blockchains are much inferior to public blockchain when it comes to security, tamper-proofing, and censorship resistance. They are certainly not decentralized. Some have even argued that private blockchains are nothing but glorified, distributed databases. I wouldn’t go that far, but it certainly has a strong use case for consortium-type blockchain in which peers must share the database containing strong access control.
Here Comes Consortium Blockchain
Multiple companies can form a consortium with a common purpose. Among those purposes are interbank fraud prevention, threat intelligence, and asset sharing, implementing a platform with a ledger deployed on several nodes with equal access to the consortium members. The ledger is fully transparent within the consortium members and can even maintain nodes with implementation of byzantine fault tolerance. Figure 5.3 shows members with access to a blockchain maintained and operated exclusively on behalf of the members.
Figure 5.3 Membership-based consortium blockchain with equal access to the members
Nodes verify transactions/blocks as per the consortium rules. In order to add data to the blockchain, a node sends a transaction request to the network of nodes in the consortium. Block validators, or miners, which are special nodes in the consortium network, validate transactions and add them to a block. Hence, validator nodes are known to the network and must be trusted by the consortium. Consortium blockchain is thus a hybrid between private and public blockchains.
Consortium blockchain deployed with a private blockchain utilizing participating peers as nodes and validators poses problems of tampering and 51 percent attacks.3 It does contain an upside of higher transaction speeds and higher confidentiality because the node network is small, which means transaction confirmation is much faster than public/open blockchain. Consortium blockchains, in a limited sense, can be called semi- or partially decentralized because no single entity has the power to tamper with the blocks and transactions, assuming it is set up that way along with governance rules.
The probability of consortium members colluding to reverse and modify transactions is also real. One way to prove that the data has not been tampered with and preservation of auditability is to periodically publish the hash of a block onto a public blockchain. By doing so, one can be assured that blocks in the interval of two published hashes have not been modified.4
The most challenging aspect of forming a consortium blockchain is deploying a sustainable governance structure that oversees the system with regard to keeping the system adequately funded. Doing so may require creating a separate entity to operate and maintain the system on behalf of the consortium, which is then governed by a board.5
_______________
1 V. Buterin. August 6, 2015. “On Public and Private Blockchains,” Ethereum Blog. https://blog.ethereum.org/2015/08/07/on-public-and-private-blockchains/.
2 Hackernoon. August 1, 2018. “Blockchain Architecture Analysis: Private vs Public vs Consortium,” Hackernoon. https://hackernoon.com/blockchain-architecture-analysis-private-vs-public-vs-consortium-65eb061b907b.
3 V. Buterin. August 6, 2015. “On Public and Private Blockchains,” Ethereum Blog. https://blog.ethereum.org/2015/08/07/on-public-and-private-blockchains/.
4 O. Dib, K. Brousmiche, A. Durand, and E. Thea. 2018. “Consortium Blockchains: Overview, Applications and Challenges,” International Journal on Advances in Telecommunications 11, no. 1 & 2.
5 R. Wilson. March 3, 2019. “The Right Way to Do Blockchain Consortiums,” Coindesk. https://www.coindesk.com/the-right-way-to-do-blockchain-consortiums.