DLT, DAO, dAPP, ASIC, PoS, P2P, PoW. Is the future of decentralised systems all alphabet soup and acronyms? It seems so. If you haven’t added ‘DAG’ to your list, you’re going to want to.
Let’s start with DLT, as in ‘Distributed Ledger Technology’. This is a set of records (a ledger) that is held by multiple entities (distributed). For example, a cryptocurrency is typically a blockchain DLT using an electronic ledger of transactions held by multiple computers. When a new transaction is added to the ledger, it is copied to other computers. In that way, multiple, identical copies of the latest version of the ledger are always available.
The Directed Acyclic Graph (DAG) DLT is a more recent solution that offers the benefits of blockchain with better performance.
The idea of a Directed Acyclic Graph is by no means a new one. In mathematics a DAG is a graph that travels in one direction without cycles connecting the other edges. This means that it is impossible to traverse the entire graph starting at one edge. The edges of the directed graph only go one way. The graph is a topological sorting, where each node is in a certain order.
Imagine a collection of individual transactions where each transaction is linked to at least one other transaction in the following way:
The links point in the same direction with earlier transactions linked to later transactions, and so on.
Loops are not possible. A transaction cannot loop back on itself after linking to another transaction.
The mesh of connected transactions can be represented as nodes in a graph network, in which nodes are joined to each other by links.
An example of a DAG DLT is IOTA’s Tangle. Each new transaction in IOTA must validate at least two previous transactions before it can be validated. An algorithm in IOTA ensures the random selection of transactions for verification, effectively preventing network members from only validating their own transactions.
Similar to how blockchains tend to differ from project to project, DAGs also differ. peaq is building on what’s been done by the likes of IOTA by creating smart-contract capabilities on top of their DAG. The peaq team are also researching consensus algorithms that would make it possible to have a fully-decentralized DAG — a feat not yet conquered in the DLT world.
Practical differences between a DAG and a blockchain typically include greater scalability and lower transaction fees for a DAG. They offer much higher validation speeds. In fact, whereas Bitcoin-style blockchain slows down as traffic increases, a DAG works faster as the number of transactions to be validated increases, making them very scalable.
There is no mining on DAGs and therefore no need to incentivise miners through mining reward systems and thus, no need to charge users high transaction fees. No mining also means no mining equipment, the sort that you’ve undoubtedly heard uses more energy than entire nations.
With little or no transaction fees, DAGs can be used in situations where blockchain would not be feasible. A prime example is micro or nano-transactions between devices and sensors.
It’s not all sunshine and rainbows, though. DAGs have issues reaching secure decentralized consensus, whether it’s through using a centralised coordinator or limiting the number of validators, the code is yet to be fully cracked.
DAG Distributed Ledger Technology is still very, very young and needs to be rigorously tried and tested. The enormity of the promise it holds however, adds that extra layer of excitement to projects in the DLT space opting to go with DAGs over blockchains.