Brain Trip: Hedera Hashgraph Distributed Registry Platform

Brain Trip: Hedera Hashgraph Distributed Registry Platform
Consensus algorithm, asynchronous tolerance for inexplicable errors, directed acyclic graph, distributed ledger - about what unites these concepts, and how not to dislocate your brain - in an article about Hedera Hashgraph.

Swirls Inc. is:
Distributed ledger platform Hedera Hashgraph.

Starring:
Lemon Baird, mathematician, creator of the Hashgraph algorithm, co-founder, CTO and chief scientist of Swirlds Inc.;
Mance Harmon, mathematician, co-founder and CEO of Swirlds Inc.;
Tom Trowbridge, President of Hedera Hashgraph, Hashgraph technology evangelist.

The project is attended by:
Financial holding Nomura Holding;
Telecommunication company Deutsche Telekom;
International law firm DLA Piper;
Brazilian retailer Magazine Luiza;
Swiss corporation Swisscom AG.

I still don’t understand why all the information about Hedera Hashgraph is presented so confusingly, whether this is a consequence of a conscious policy of the developers or it just happened by chance. But in any case, writing a coherent text about Hedera Hashgraph turned out to be very difficult. Every time it seemed that this was it, I finally understood everything, almost immediately, again and again, it turned out that this was a deep delusion. In the end, it seems to have turned out to be something meaningful, but anyway - read carefully, the danger of dislocating the brain has not gone away.

Part 1. The task of the Byzantine generals and gossip
At the heart of this story is the so-called Byzantine Fault Tolerance (BTF) problem, a thought experiment designed to illustrate the problem of synchronizing the state of systems in the case when communications are considered reliable, but nodes are not. Anyone who is interested can study the issue here or here in more detail.

The algorithms of the Hedera Hashgraph platform are based on a special case of Byzantine Fault Tolerance, the asynchronous Byzantine Generals Task, or aBFT. In 2016, for the first time, the mathematician Lemon Baird proposed a solution for it and, don't be a fool, he immediately patented it.

The Hedera Hashgraph platform is characterized by the sharing and synchronization of digital data according to the consensus algorithm, the physical decentralization of data storage nodes and the lack of a single control center. However, the Hashgraph protocol (in this case, Hedera is an eco-environment, Hashgraph is a protocol) does not belong to blockchains, but is a digraph devoid of sequential cycles and consisting of parallel sequences starting at the same node and reaching the end node in different ways.

Roughly speaking, if the classical blockchain can be visually depicted as a strict sequence of links (which, in fact, is its main property), then Hashgraph visually resembles a bonsai with a huge number of branches. Since the number of simultaneous cycles is almost unlimited, Hashgraph allows you to simultaneously carry out a huge number of transactions (the developers say about 250 thousand per second, which is five times higher than even Visa, not to mention the bitcoin network), and there is usually no transaction fee.

The next fundamental difference between Hashgraph and the classic blockchain is the gossip subprotocol (gossip protocol). Within a distributed ledger, each transaction does not mean the transfer of all data, but only information about information (Gossip about Gossip). The node reports the transaction to two other arbitrary nodes, each of which, in turn, broadcast messages to two others until the moment when the number of notified nodes is sufficient to reach a consensus, and this happens when most of the nodes are informed (and it is due to this that the declared number of transactions per unit of time is achieved).

Part 2. Blockchain killer or not
Hedera Hashgraph is currently under development. In particular, we are testing our own cryptocurrency with support for micropayments, a distributed network storage of files and scripts that allow you to create smart contracts based on the languages ​​of the Ethereum environment.

Opinions about this project are rarely polarized. Some sources simply call Hashgraph a “blockchain killer”, others rightly point out that there are no examples of working decentralized applications in the Hedera environment, others are embarrassed by the fact that the basis of the platform is patented, and its development is controlled by a supervisory board, which includes representatives of a number of companies from the Fortune 500 list (although the latter just says that the project has a non-illusory potential, and is definitely not a scam). By the way, some time ago the project was spun off into a separate company, Hedera Hashgraph, which also indicates its priority for developers.

The developers, without much fuss, first raised $18 million for operational needs at a closed sale of tokens, and, after some time, another $100. No specifics about the ICO are also reported, and in general, the Hedera Hashgraph roadmap is distinguished by rare incomprehensibility, which does not prevent the company from conducting active activities aimed at popularizing this consensus algorithm, the company is actively working on the formation of various professional communities - from programmers to lawyers, project representatives have already held more than 80 meetings with interested citizens around the world, they even reached Russia - on March 6, a meeting was held in Moscow with Hedera Hashgraph President Tom Trowbridge, which, as they say, brought together a lot of representatives of our IT and financial circles.

Mr. Trowbridge said that at least 40 decentralized applications based on Hedera Hashgraph are expected in the near future, and in general there are more than 100 of them in operation, so in the foreseeable future everyone will have a chance to see how this economy works live.

Total
In general, several things can be said with certainty. Firstly, the project is not trivial and has already aroused the most serious interest of representatives of large corporations. Secondly, for a non-specialist, he is frankly incomprehensible, which, apparently, explains the lack of data about him in the public domain (and also, judging by the video with Mr. Limon, the fact that this smart uncle has never been a speaker at all). Thirdly, it is unlikely that it will become a “Bitcoin killer” or anything equally pathetic, but its declared advantages look weighty enough to follow the project very closely.

Moreover, there are rumors that soon the organizers are going to attract another tranche of investments, it is quite possible that it makes sense to take part in it.

Source: habr.com

Add a comment