Hashgraph is a consensus algorithm that aims to provide a distributed and secure method for achieving consensus in a network of nodes. It was created by Leemon Baird and introduced in 2016. Unlike traditional blockchain-based consensus algorithms, Hashgraph does not rely on proof-of-work or proof-of-stake mechanisms. Instead, it uses a directed acyclic graph (DAG) structure to achieve consensus.
In Hashgraph, each node in the network maintains a record of all transactions and the order in which they occurred. These records are stored in a data structure called a “hashgraph,” which consists of two main components: events and edges. An event represents a transaction or an action taken by a node, while an edge represents the causal relationship between two events.
The consensus algorithm in Hashgraph operates in two main stages: gossip and virtual voting.
- Gossip: In this stage, nodes in the network exchange information about the events they have witnessed. Each node randomly selects another node and shares the information about its events and their order. Through this process, nodes gradually build a shared understanding of the history of events in the network.
- Virtual Voting: Once nodes have exchanged enough information, they can determine the order of events through virtual voting. Nodes do not send actual votes to each other; instead, they use a virtual voting algorithm. This algorithm considers the information received during the gossip stage and assigns each event a “fame” value, indicating the confidence level that the event is valid and should be included in the consensus order. The virtual voting algorithm also considers the voting weight of each node, which is determined by the number of events created by that node.
As nodes continue to gossip and perform virtual voting, a consensus order is gradually determined. Once consensus is reached on the order of events, the hashgraph can be used to generate a tamper-proof record of the history of transactions and actions in the network.
Hashgraph offers several advantages over traditional blockchain-based consensus algorithms. It claims to achieve high throughput, enabling a large number of transactions per second. It also claims to provide fairness, as no single node has an advantage in determining the consensus order. Additionally, it claims to provide high security, as the hashgraph structure ensures that any tampering with the order of events is detectable.
Overall, Hashgraph is designed to provide a robust and efficient consensus algorithm for distributed networks, with potential applications in areas such as cryptocurrencies, supply chain management, and decentralized applications.
