Beginner’s Guide to HashGraph – The Competitor to Blockchain

Beginner’s Guide / 03.05.2020

Blockchain – The Solution to Centralization

The advent of a decentralized, distributed, and self-fulfilling platform, i.e., Blockchain, has revolutionized how we exchange value and information with each other for good. In many ways, blockchain will help us move away from the world with central regulatory bodies needed for brokering trust and towards a decentralized trust-less ecosystem. A world where technology shoulders the responsibility of federating the trust.

Belief is that Satoshi Nakamoto, the anonymous inventor(s) of Bitcoin and Blockchain technology, wouldn’t have imagined that their novel invention would go on to break traditions and institutions in the way it has done today. Blockchain technology came to a renewed interest in the idea of a utopian decentralized world where everyone works for everyone, and no single authority is given a chance to take advantage.

With Pros Come Cons

But then, as we always say, not everything is perfect. Sure, blockchain might end up changing the way we transact value and information for good. But with the innovation in decentralized technologies came problems that were unique to it and it alone. A blockchain is a disruptive technology, but it has its fair share of issues. The proof-of-work algorithms needed to validate transactions are slow and resource-intensive. A few transactions per second is snail-paced compared to the millions of transactions we do over the internet.

Since 2010, innovators have been trying to build upon the Bitcoin blockchain by identifying its flaws and trying to solve them by building clever and innovative infrastructures. No matter how much one tries to improve on a blockchain project by bringing in another blockchain project, problems inherent to blockchain technology cannot be changed. Many realized that there would be an imminent update to blockchain technology that would borrow its basic principles and innovate in its way.

Introducing Hedera HashGraph – The Blockchain 2.0

Blockchain and Hashgraph data structures

Blockchain and Hashgraph data structures. Source: Hackernoon

As per Wikipedia, “HashGraph is a distributed ledger technology developed by Leemon Baird, the co-founder and CTO of Swirlds, in 2016. It is an asynchronous Byzantine Fault Tolerance (aBFT) consensus algorithm that they consider capable of securing the platform against attacks. It does not use miners to validate transactions and uses directed acyclic graphs for time-sequencing transactions without bundling them into blocks.”

HashGraph is a DLT platform, which employs Gossip protocol for consensus building. It was made to provide a fast, fair, and secure infrastructure to initially run Decentralized applications (dApps) but can increase its scope in the future.

HashGraph works on a gossip-based protocol, which is fundamentally not a new idea. In a gossip-based system, nodes communicate with each other and tell each other information that they know. For instance, communicating the current state of the network, communicating nodes trusted by a node, and so on. Upon receiving this gossip, these nodes will gossip this information to other nodes, forming a network where every node eventually knows about every other node.

The innovative idea that distinguishes HashGraph from traditional gossip systems is that rather than just gossiping arbitrary data, nodes gossip about gossip, i.e., the data structure in the image on the left is stored in memory and sent to others. Intuitively, if you imagine each node in the diagram representing a new transaction being added to the graph, then rather than just sending that node, they will send a whole tree of nodes starting from the new one when host gossip.

Delving Deeper into the Properties of HashGraph

To achieve consensus, HashGraph uses two techniques – Gossip about gossip and Virtual voting.

Gossip about gossip means sharing information about the information being spread. This is achieved by adding a tiny amount of additional information, a couple of hashes about the people spreading the gossip (the parent nodes). This helps in creating a historical graph of information being disseminated.

The information transmitted from a node is disseminated as gossip through a random selection of peers to share what you know. The idea is to share the information with as many participants as possible, if not all. Interestingly, epidemics spread in the same way (boy oh, boy!). It has been mathematically proven that gossiping is the fastest way of spreading news, and HashGraph builds on it.

Virtual Voting is used to build a consensus on the gossip’s ordering and timestamp (the transactions). In simple words, the participants take turns voting on what they know and when they came to know about it. It’s virtual, as each participant simulates the voting, using an algorithm, to arrive at a consensus on ordering and timestamps. Actual voting is unnecessary, as all participants already have a copy of the same data-set and are supposed to arrive at the same voting results, which is also mathematically proven. This increases the speed with which a consensus is arrived at.

Blockchain vs. Hashgraph

By many standards, HashGraph looks like the perfect replacement for the Blockchain. HashGraph improves upon many of the flaws of the blockchain infrastructure by solving them from the ground up. This allows applications built on HashGraph to be wary of the blockchain system’s fundamental flaws, thereby being more efficient and optimized.

  • Fast – more than 250,000 transactions per second and very low latency (in seconds)
  • Fair – it ensures fairness in consensus on ordering and timestamps and fairness in access, which is key to any distributed ledger system.
  • Secure is tolerant against behaviors like lying, collusion, and selective non-participation, which is essential for distributed ledgers. In technical works, it is Asynchronous Byzantine Fault Tolerant, which is the gold standard for security in distributed systems.

HashGraph is an innovative update to a blockchain. It uses a traditional piece of technology called the Gossip Network to improve the blockchain’s fundamental flaws. In many ways, HashGraph sure does look like it should have replaced Blockchain already.

But then, just like Blockchain, HashGraph too has its fair share of trade-offs. What it gains in efficiency and optimization, HashGraph loses in decentralization and openness. HashGraph has to work in a relatively closed environment to work at full capacity compared to a blockchain. Blockchain thrives in distribuends, meaning that the greater the number of participants, the more its strength. Whereas HashGraph can function in a distributed environment, the network’s efficiency declines with an increase in the number of participants.

And so, HashGraph should not be viewed as a replacement for blockchain, as many “market experts” have confidently claimed. HashGraph and Blockchain are like Pepsi and Coca-Cola of the technology world. Each side has its advantages and trade-offs, which gives prospective users the ability to weigh their options and choose the best infrastructure that suits their needs.

Sudarshan M is a long time crypto-enthusiast. Pulled in by bitcoin early on, it did not take long for Sudarshan to divert all of his academic attention from business studies to blockchain by doing his Masters and eventually pursuing his PhD in the subject.