AI- and Blockchain-Driven Databases

"From store of value ... toward value of storage"

Databases have been a critical component in the development of applications, across all generations of the web. While the advent of databases started from a centralized design pattern, there have been several iterations of innovations over the past three decades. These new patterns address key pain points of centralized databases.

In this chapter, we will dive deep into a new generation of decentralized databases and filesystems based on innovative design patterns. Some of these design patterns inspire traditional applications to blend Artificial Intelligence (AI) and blockchain technologies. We will observe different types of decentralized databases and understand how they help to perform better AI analysis alongside blockchains.

In this chapter, we will cover the following topics...

Technical requirements

This chapter assumes your acquaintance with the basics of database design and the application of AI techniques in related scenarios.

Centralized versus distributed data

Databases have been primarily consumed in a centralized manner since their earliest applications, dawning in the mid-1960s. Databases were meant to provide direct access to the information requested by either users or client applications. This centralized approach was influenced majorly by the client-server architecture introduced in the early days. This design paradigm was popularly followed by the market with successful products in commercial- and consumer-level databases such as DB2 and dBASE, respectively. Relational Database Management System (RDBMS)-based databases followed the client-server model. These centralized databases managed data redundancy by making regular copies of the data on disks and magnetic tapes.

However, the dawn of NoSQL in the 2000s is credited with distributed databases that scale horizontally, with higher tolerance to failures and less chance of data corruption. NoSQL databases are able to manage data without schemas and...

Blockchain data – big data for AI analysis

As you may be aware, blockchains generate enormous amounts of data due to their transactional nature. At the time of writing, the size of some of the prominent blockchain networks is as follows:

Blockchain	Total size of the blockchain (approx. in GB)
Bitcoin	323
Ethereum	4,233

Some experts in the industry have speculated that the size of blockchains will soar 10 times more, due to an increase in the number of users and the adoption of public networks in the business-to-business (B2B) landscape.

The growing size of blockchain data enables new avenues of growth for data science. The application of AI and analytical practices on this giant heap of transactional data in the blockchain can create a large impact on most of the current blockchain products. Analytics derived from qualified data sources such as blockchain can also lead to new digital transformation projects. In order to facilitate this, we need a secondary source of...

Global databases

In this section, you will be introduced to some of the most popular decentralized databases. These databases use innovative cryptographic and networking techniques to address some of the key issues such as censorship, surveillance, and permissioned access to confidential information. Several efforts are being made by the projects outlined next to bring a new order to how data can be treated in the public sphere as well as the enterprise sphere.

Let's now understand some of the top global decentralized databases.

IPFS

IPFS is a distributed filesystem that allows users to host and receive content in a peer-to-peer (P2P) manner, eliminating any need for intermediaries, for storing or accessing data from any corner of the world. IPFS allows users to store and serve data in a censorless manner. The data remains persisted in the network as long as somebody in the network values the data. Although there may not be a monetary incentive for users who persist the data on their...

Data management in a DAO

A Decentralized Autonomous Organization (DAO) is a computer program representing a group of stakeholders and entities and is not influenced by external environments. A DAO is programmed by a set of rules and governance protocols to ensure that transactions occur between parties without the chance of any conflict. Dash and BitShares are some of the earliest implementations of a DAO. In the past few months, many more DAOs have been launched on blockchains such as Ethereum and Bitcoin.

Aragon

Aragon is an open source DAO running on the Ethereum blockchain network. Aragon leverages Solidity smart contracts for business logic, and IPFS for decentralized files and governance record management, thereby creating a truly P2P operating system for a whole new generation of organizations, called aragonOS. Users can perform operations and govern their DAOs using the Aragon Network Token (ANT).

Aragon has integrated IPFS very closely into its command-line interface (CLI) program...

Emerging patterns for database solutions

Very few companies have been able to both converge technologies and address key issues in the respective industries. In this section, we will explore the key issues of respective domains and explore patterns to solve them, along with ideal examples.

Let's now understand the current issues in the enterprise software domain and explore the emerging patterns applicable.

Enterprise

Enterprises and large organizations have been successfully scaling, thanks to scalable systems such as Enterprise Resource Planning (ERP) software, Knowledge Management Software (KMS), and Inventory Management Software (IMS), to name a few. However, growing requirements and a groundbreaking revolution in how data is managed have led to a path of blockchain applications and AI technologies.

Technical impediments

The enterprise software domain is facing new challenges in managing huge amounts of data among stakeholders in a reliable manner. Here are the top three challenges...

Summary

In this chapter, we have looked into the realm of storage by introducing you to the concept of centralized, distributed, and—finally—decentralized databases. Further, we have also contrasted the data consumption patterns between Web 2.0 and Web 3.0 apps. We have also understood more about the core motivations and the need for using decentralized databases in applications and DAOs. Finally, at the end of this chapter, we have explored various emerging patterns that can be analyzed and applied.

In the next chapter, we will observe how these emerging patterns are applied to build smart applications for the decentralized economy, with the help of blockchain and AI.