Why Blockchain will never kill the database
A fundamental difference in the way the data is handled and stored indicates that the technologies are complementary, not competing.
The hype blockchain is out of control. Although blockchain is a surprising technology that makes ecosystems data more secure, reliable and verifiable, it is not a panacea. The hype blockchain includes in particular a false statement, that is, because blockchains can serve as verifiable records systems, databases are no longer the right technology to achieve that goal. This is misguided. Blockchains and databases are different types of recording systems and, in fact, they are free.
Blockchain Advantages and challenges
There are several blockchain technologies and networks and all share a basic feature: a "transaction" record is not stored in a single database. Instead, a consensus of the transaction is recorded between an entire network of participants in an ecosystem.
Blockchain is a record of immutable and distributed transactions. It uses cryptographic algorithms to reach a consensus between a group of parts in a secure manner, with the result that each part of the transaction chain has an accurate record of each transaction. There is no central repository protected by a single part that can be induced to modify the database for its own interest. The blockchain is reliable by virtue of its distributed model, how the blocks are connected to the chain and its consent algorithm which makes the cost of altering it prohibitive.
Blockchains are computationally expensive. By design, the cryptographic algorithms used to obtain a consensus require substantial work. As a result, there are many efforts focused on reducing computational spending, spending on cryptocurrency and on energy expenditure. An approach, called anchor, reduces the amount of data stored in the chain where transactions are grouped, submitted to hashes, and organized into timestamp blocks for inclusion in the blockchain. A receipt indicating where on the blockchain the data has been anchored is then stored in databases or other durable archives, making any transaction verifiable.
A key aspect of this approach is that the data involved in the transaction are not "stored" in the anchor. Only one cryptographic data hash is stored. The anchor is used to check the original data against the hash and to determine when the blockchain has been committed, but it is not used to store the data. This is really a registration system because it records a hash of transaction data whose integrity can be verified by anyone at any time. This provides an independent source of trust, while maintaining the confidentiality of confidential data, even on public blockchains.
What applications does a blockchain support? They are subdivided into three categories:
- Smart Contracts ensure the consistent transfer of assets according to predetermined rules
- Smart resources ensure that the ownership status of any tokenised resources can be monitored, verified , and established between the parties
- Smart IoT ensures that the signals generated by the devices have not been tampered with and reflect the true values detected
The databases differ from the blockchain because they explicitly store the data , not just hash. Databases feed two types of workloads: operational workloads and analytic workloads.
Operational databases, called online transaction processing (OLTP) systems, power some applications. For example, consider a fraud dispute resolution system that allows a call center agent to help customers review financial transactions and file disputes related to those transactions in a second or less. This requires special data structures and algorithms that can process data from many users at the same time very quickly.
Online Analytical Processing (OLAP) systems review historical transactions and take insights or generate predictive models of machine learning. These systems are specialized for data sorting and calculation metrics like sums and averages. This requires high throughput.
New databases are now emerging that can combine OLTP, OLAP and machine learning in a single platform called Online Predictive Processing (OLPP). [Editor’s note: The author’s company, Splice Machine, provides an OLPP platform.]
For example, consider these three use cases:
- Customer service call centers: call center agents that respond to customer requests on channels such as telephone, web, or mobile apps often after orders have been placed
- Customization: machine learning models that predict what actions to take with a customer at a time
- Predictive maintenance: machine learning models that predict when equipment outages in the field are likely to occur
All of these use cases require a database – a blockchain simply can not perform these functions.
A final word
The death of the database is extremely exaggerated. Blockchains can revolutionize transaction integrity, but databases will always remain to power mission-critical applications, analyze those applications and serve as the heart of the learning AI. Together, they offer a powerful combination to many verticals.