Yesterday I had a chance to see a presentation by Mance Harmon & Tom Trowbridge, the Founders of Hedera Hashgraph. Hashgraph presents a potential for a next generation ledger solution, or in simple terms - The next evolution of Blockchain technology.

The actual Hashgraph algorithm was written by Leemon Baird, a highly gifted individual with a PhD in Computer Science from Carnegie Mellon University. Baird has a long history of Math & Algorithms, specifically around machine learnnig and more recently distributed consensus.

Hasgraph promises to tackle some of the biggest issues with Blockchain: Performance, Security & Governance.

Hashgraph Performance

The most critical performance metric for any Blockchain/Ledger is TPS - Transactions Per Second.
Here are some TPS numbers for the most popular Cryptocurrencies and traditional payment processors:
  • Bitcoin 2.5 TPS
  • Ethereum 15 TPS
  • Visa 1700 TPS (avg) to 4000 TPS (peak)
  • PayPal 115 TPS
And Hashgraph's promise?
  • Hashgraph 50,000 TPS - 250,000 TPS
If these numbers can be replicated on a public ledger, we're looking at a potential mass-market solution without any of the familiar growing pains for the foreseeable future.
Hashgraph also promises to integrate Sharding to allow for even larger distributed applications on top of the platform.
Mance Harmon, CEO of Hedera Hashgraph

Hashgraph Security

Hashgraph claims to have achieved the highest theoretical level of security in the crypto industry, the coveted "asynchronous Byzantine Fault Tolerance" or aBFT for short.
Without going too much into the details of aBFT, they present an elegant and efficient way to replace Proof-of-Work, which thus far has been the dominant BFT solution.
This potentially gets rid of the obvious energy-consumption and hardcoded slowdown issues of the system.

Hashgraph Governance - the council of 39

In order to avoid the immediate 1/3 dishonest nodes problem, two thirds of the initial tokens will be given to a council of 39 Major Corporations & Organizations from different industries and locations. Those entities will not technically hold the actual tokens, but they will have committees to eventually decide when to release those tokens to the public. All revenues from sale will go back to Hedera.
This event will occur when the token is of high enough value, such as that a 1/3 dishonest node attack will no longer be financially feasible.
It remains to be seen whether it's possible to determine when this will happen, as well as preventing a single actor from gaining too much power over the economy as a whole.
Hedera Hashgraph is definitely trying something unique here, and it's mostly a matter of time to see how it works in a live public setting.

A distributed Ledger that is fault-tolerant AND can do over 250K TPS can truly usher an era of a decentralized, user owned internet.