Anatoly Yakovenko, CEO and Co-Founder of Solano in a Webinar with Blockmatics expressed, covered some of the innovations that make SOLANA a high-throughput real-time blockchain, proof-of-history, tower byzantine fault tolerance, the proof-of-stake model used and many other topics in SOLANA. SOLANA is incredible stuff.
Anatoly Yakovenko, pointed to physics classes and that it is important to remember that radio transmissions interfere when they are in the same frequency. If two towers transmit at the same time and the two frequencies gets the noise, then the receivers don’t know what the information is and what was actually transmitted. So, this is a similar problem on blockchain. If two block producers produce a block at the same time there is a fork. And, nobody knows what the state of the network is and it is in a noisy state. So, you have to kind of reconcile what happened.
How does this solve in radio is? The first idea that people had was why don’t we give each one tell us by a clock but synchronize and they alternate by time. If you can keep the clock synchronization that is really tight. Imagine, if they start to drift, they will start interfering again. And, if you can keep that really tight, they just take turns by 1 second, 20 milliseconds 5 milliseconds whatever the tolerance is. And, you can increase the number of transmitters per second and that kind of eliminates the problems when you have one of them fail. Like if one of them is failed and the next one goes in two seconds. That is worsened when the next one goes in 20 seconds or 5 seconds or 5 milliseconds. As the clock alternations and the number of producers per second improves, it is going to start eliminating what appears like dead zones or dead drops for users, which they don’t even noice.
Yakovenko agreed on how leader election is same thing like consensus. He spoke about the Solana Cheat Code. Proof of History is the cryptographic process called the verifiable delay function. When you look at the data structure you know that it takes a certain amount of time to generate it. The real time may be different, but you know that some amount of time. He spoke about their logical clock, which he says the domain of how big this data structure is. You almost think of it as a water clock or sand dial. Like the more sand falls you can see the level of data grows. You don’t care what the data is, but it is like a cryptographic sand clock.
