The titles vary and the main theme is renewed, but scalability manages to show off its omnipresence, always finding a place to sneak into our blog as it continues to unveil the crypto world. Today, it is my turn to analyze what Constellation (DAG) is and how it intends to provide a solution to one of the great problems of our time.
We quickly notice that the interesting thing about this solution is that it leaves blockchain technology behind to work with Acyclic Directed Graphs. This is a mathematical-computational construction, which allows the creation of distributed systems and networks with similar characteristics to our beloved blockchains.
But, let’s not dwell any longer on long introductions and let’s quickly attack the topic that summons us today.
Getting to know Constellation
When we talk about Constellation, we refer to a “Layer 0” or “layer 0”, which is not based on blockchain technology, but, as I mentioned in the introduction, on Acyclic Directed Graphs.
Evidently, the opening paragraph, has not been clear, even for me. Let’s understand it step by step. First, a “layer 0” is one that comes before a “layer 1”. Therefore, these are solutions that allow us to create main chains, which fulfill the same functions as our famous blockchains. Let’s look at an example:
- Bitcoin and Ethereum, are “layers 1” or core networks.
- A “layer 0” would be one that allows us to create main chains
This is what Constellation is all about. It allows us to create “state channels”, with the business logic we want to imprint on them.
At this point, we can draw a parallel between the Cosmos ecosystem and Constellation. Like Constellation, Cosmos allows the creation of interoperable blockchains that are adaptable to the needs of their creators. Of course, the big difference lies in the technology used by Constellation’s “state channels”. Let’s see what this mathematical-computational theory of DAG is all about.
What are directed acyclic graphs (DAG)?
DAG technology, Directed Acyclic Graph, finds a definition similar to the one we can give to blockchain. It is a distributed accounting technology. But, in them we add the concept of asynchronicity. These networks are able to combine asynchronous accounting with asynchronous accounting.
The graphical representation of this technology is a set of interconnected data. The first thing we notice when we look at one of these directed acyclic graphs is its nodes, represented by circles. These are connected by lines that simulate the flow of data between the different points within the graph. At the same time, these points represent a subset of data within their own group, each one. Looking at the image below, it will be simpler to unravel this technical hieroglyph.
The last detail, which we can highlight about a DAG, is that the larger its size, the higher its efficiency, which we can understand here as speed. Instead of achieving consensus according to the “longest chain” principle, priority is given to the “heaviest”.
Properties of Directed Acyclic Graphs
The performance of these graphs is determined by the following properties:
- It is not possible to change the information that a node holds without modifying the data contained in all the nodes of the graph. In other words, making changes to the relationship between nodes would imply rewriting the entire DAG.
- They have an origin and a destination. In this way we ensure that the information path always goes from an origin point to an end point. That is, it is impossible to start from a node of the graph, go completely through the rest of the nodes and end the trip at the same node where we started.
- A DAG can process information in parallel and connect different information “routes” between different nodes at the same time.
- They are feasible to reduce. By this definition we mean that the structure of a DAG can be reduced to an optimal point where its path complies with all the ratios specified in it without any loss. From this comes its quality of “lightness” in terms of computational weight.
Advantages of a DAG vs. a blockchain
When comparing DAG to a blockchain, as we know them, we find the same differences as when comparing a graph to a chain. Blockchains have a single entry point and a single exit point. Each new block is “connected” or “placed” ahead of the previous one and will continue to do so for as long as it continues to operate.
However, in Directed Acyclic Graphs, versatility is the rule. In them, nodes can have different entry and exit points. At the same time, these nodes work in an acyclic way, allowing to perform an infinite amount of processing at the same time.
We can summarize the characteristics of these graphs as follows:
- Infinite scalability. Setting up a node does not require any kind of permission and anyone can do it. As the number of nodes increases, so does the processing capacity of the network.
- Extreme speed. Since it is possible for each node in the graph to process an increasing number of transactions in parallel, the number of transactions per second grows at the same rate as the number of nodes.
- High-level security. The large number of inputs and outputs, present in the graphs, prevent malicious manipulation of the nodes, since, like a blockchain, by modifying one, all should be modified.
How does Constellation work?
Well, we already know the technology that is presented as the solution to each and every one of the problems that our beloved blockchains still cannot solve, without shattering the “blockchain trilemma”. Now, let’s get to know how the Constellation project is composed.
Constellation and its application of DAG technology
Constellation, embodies DAG technology through its decentralized protocol called Hypergraph Transfer Protocol (HGTP). It is a system that allows transferring, validating, auditing and managing data from devices connected to the Internet at high speed.
Adding these features to the decentralization through the creation, without the need for any permission, of new nodes capable of sharing information, they fulfill functions analogous to those developed by a blockchain.
With a simple, inexpensive and impenetrable implementation, Constellation aims to allow anyone to create their own “state channels”, and adapt it to their own objectives.
The DAG L1 network
But Hypergraph not only allows each user to create their own “state channel”, it also has its own, known as the DAG network. In it, we find the native currency of the ecosystem, identified with the ticker $DAG. Its main function is to be the means of payment in which those who set up a node in Constellation must pay.
This network is in charge of validating the transactions of its native cryptocurrency. A remarkable feature is that in DAG L1 transactions are free. As in an Ethereum L2, the transactions made here will be sent to “Layer 0” where they will be finally hosted.
States channels
State channels can be understood as sub-networks that carry out their own validation and consensus of their transactions, prior to sending them to “Layer 0”, as is the case with DAG L1. Layer 0 is where the final consensus and final storage of transactions and their data takes place.
A state channel is the equivalent of a conventional blockchain, since it processes the data of the interactions that take place within its environment, and unlike Layer 0, they require access to an exclusive entry point.
However, unlike a traditional blockchain, they have a so-called “own-state isolation” approach. This allows them to independently process unrelated state updates, something that is impossible in a blockchain. Another difference with a typical Ethereum L2 solution is that in a state channel, there can be different consensus capabilities.
Clearly, the Constellation community is looking to grow its internal ecosystem of state channels in order to attract more users.
Disadvantages of this technology
Those of us who staunchly defend blockchain technology tend to ask ourselves, when we take a close look at Directed Acyclic Grids, is it a truly decentralized technology?
The reality, answers us that our question, beyond the intention of defending our blockchains, is right. At the moment, the nodes of the solutions that have implemented DAG are almost entirely managed by those who gave birth to these networks. The defense of the defendants argues that this is a short-term solution, in order to give rise to the adoption of the technology. Undoubtedly, they are confident that, once a certain degree of adoption has been achieved, users will start running their own nodes.
Secondly, DAG-based networks have not yet had large-scale production trials. Thus, their performance according to expectations has not yet been “tested”. This lack of “testing” in a large environment, which generates the necessary frictions to know the response of DAG technology when faced with its own limits, generates certain uncertainty and is pointed out as one of the factors holding back its definitive explosion.
The future of our ecosystem
After a first reading of this article, we can conclude that the theory of Directed Acyclic Graphs is a rival to be taken into account for those of us who work in the blockchain ecosystem.
Like any new implementation, its future depends not only on its goodness, but also on issues related to the timing of its appearance, its ability to present itself as a solution to real problems and the blessed chance that does so much for us.
As it takes its first steps in the competitive world of distributed information storage networks, it is still struggling to achieve end-user adoption.
Will DAG be the one to send blockchain technology to museums? Voices, and rationales, are rising both ways. It will be a matter of following closely, as this story is written….
