The directed acyclic graph (DAG) vs. blockchain debate brings new life to the cryptocurrency industry. Ever since Satoshi Nakamoto first published the Bitcoin (BTC) white paper, distributed ledger technology (DLT) has exploded in popularity. Many crypto enthusiasts don’t know blockchain systems aren’t the only decentralized network.
What is a directed acyclic graph? Are DAG networks truly better equipped to handle real-world use cases?
What Is a Directed Acyclic Graph (DAG)?
A directed acyclic graph, or DAG, is a distributed digital ledger that records transactions and stores cryptocurrency. Like a Layer 1 blockchain, some DAG networks support smart contracts and host innovative dApps and DeFi products.
To the untrained eye, using a DAG isn’t very different from using a traditional blockchain. However, looking under the hood, you’ll find that DAG-based networks use a slightly different data structure.
How Do Directed Acyclic Graphs Work?
In a blockchain, new blocks are validated by nodes or miners and added to the network. Nodes validate new transactions by confirming their data against the recorded history of previous transactions in the last block.
If a blockchain resembles a chain of blocks, a DAG looks more like a tree with unclosed vertices and edges. Every node in a DAG-based model can have more than one parent root, meaning multiple new transactions can be validated simultaneously. Instead of referencing only the last block, DAG nodes reference previous transactions from any node in the network.
In a directed acyclic graph, interconnected nodes build off each other and reference multiple transactions. This theoretically makes them more expansive and eases network congestion.
Like traditional blockchains, nodes achieve network validation by ‘agreeing’ on the network state through consensus algorithms. Generally speaking, DAG networks use the Proof-of-Stake (PoS) consensus mechanism due to its low energy consumption.
Which Crypto Projects Use DAG Technology?
Despite being a relatively new DLT system, DAG-based networks are proving popular within the crypto market. Crypto projects like Hadera Hashgraph and Fantom (FTM) are built using DAG technology to great effect. They are supporting thousands of users within their thriving ecosystems.
Other projects include IOTA, a directed acyclic graph designed to support Internet-of-Things (IoT) applications, and Nano, a decentralized payment network.
Distributed Ledger Technology (DLT) Compared: DAG vs. Blockchain
While it’s important to understand how directed acyclic graphs work, most people are more interested in results. How does DAG technology fare against traditional blockchain standards?
DAG networks were invented to solve some of the scalability issues faced by legacy Proof-of-Work blockchains, including high transaction fees and low throughput.
On paper, a directed acyclic graph can reach consensus and process transactions faster than blockchain networks. Why? Blockchains are only able to create one new block at a time, and cannot begin producing a new block until the previous one has been completed.
In contrast, nodes in DAG-based networks can reference multiple other operators simultaneously. In Hadera Hashgraph’s Gossip protocol (pictured below), nodes share information with each exponentially, meaning large amounts of data flow freely across the network quickly, helping nodes reach validation quickly.
While this sounds great in theory, DAG-based networks like Hadera Hashgraph and Fantom still can’t achieve the same transaction speeds that we see in top PoS chains like Solana (SOL) or Aptos (APT).
Compared with legacy traditional blockchains like Ethereum and Avalanche, DAG-based networks are more affordable. Transaction fees on Hadera Hashgraph cost as little as a fraction of a penny, while a similar transaction on Ethereum might set you back a few dollars.
The competition gets a bit tighter between Fantom and Avalanche, with both networks demanding only a few cents to process transactions. It’s also important to mention that network demand plays an important role here. While Ethereum, Avalanche, and Fantom have all witnessed spikes in gas prices due to congestion, Hadera Hashgraph’s scalability hasn’t been truly tested yet.
If there is an area where DAGs shine, it is their sustainability. Directed acyclic graphs typically boast low energy consumption rates, making them an ideal candidate for global adoption.
To give you an idea, the Fantom network uses between 0.000024-0.000028 kWh to process a single transaction. While it has made huge strides in reducing its environmental impact by moving to a PoS consensus, it still uses 0.03 kWh to execute a transaction.
Due to slow adoption rates, decentralization is a weakness for DAG networks. For example, the Hedera Hashgraph is governed by the Hedera Council, a central committee of 39 node operators. Compared to Ethereum, which has over 500,000 validators, you can see how easy it would be to convince a majority of validators to take over the network.
Additionally, DAG networks are less secure than their blockchain counterparts. Blockchains benefit from a ‘global state’ or a universally agreed-upon condition of the network with the creation of each new block. In a DAG, this global state is altered every time a new transaction is processed.
If communication between nodes isn’t fast enough to ensure every node is referencing correct information, there are increased chances of vulnerabilities or inaccuracies.
DAG Technology & The Ethereum Virtual Machine
Despite the architectural differences behind the scenes, most users won’t be able to tell the difference between using a DAG or a blockchain. This is due mainly to the simplicity and practicality of the Ethereum Virtual Machine.
Top DAG-based networks like Fantom and Hedera are both EVM-compatible. Their smart contracts are generally written in Solidity and compatible with EVM wallets like Metamask. Thanks to the EVM, developers can easily ‘copy and paste’ Ethereum-based applications directly onto Fantom and Hedera.
Directed acyclic graphs are fast and affordable networks far more scalable than legacy blockchains like Bitcoin or pre-merge Ethereum. However, they still face some decentralization and security issues that might discourage some users.
Moreover, modern Proof-of-Stake blockchains like Solana and Aptos are theoretically more scalable than the leading DAG networks, meaning DAGs still have some work to do to compete with top Layer-1s.
That being said, blockchain technology has been in development for much longer than DAG networks. With time and resources, DAG-based networks might still overcome their current limitations and push new boundaries in distributed ledger technology.
On the Flipside
- From an end-user perspective, it’s almost impossible to tell whether you’re using a directed acyclic graph or a blockchain. The average person doesn’t care about the underlying infrastructure as long as the final product is functional and serves its purpose.
Why You Should Care
Directed acyclic graphs are a creative method of building a distributed ledger. Just because blockchain is the most common and best-understood kind of DLT network, that doesn’t mean that innovative new systems can’t emerge and improve the industry.
DAG is an acronym for directed acyclic graph. Like a blockchain, a DAG is a network architecture for storing data and digital assets on a distributed ledger.
No, Bitcoin does not have a DAG. Bitcoin uses blockchain technology to record a distributed ledger of accounts and transactions.
Yes, in most cases DAG-based networks are decentralized. However, the top DAG crypto projects like Hedera Hashgraph and Fantom are less decentralized than leading blockchains like Bitcoin and Ethereum.
DAGs are considered to be more scalable than blockchains because nodes can validate new transactions simultaneously. On a blockchain, the network validates new blocks one by one.
No, Cardano (ADA) does not use a DAG system. Cardano is a blockchain that uses a Proof-of-Stake consensus mechanism to secure the network.