Understanding Solana: PoH and Base Station Byzantine Fault Tolerance Algorithms

DeFi is expanding rapidly, and the “impossible triangle” problem plaguing the public chain sector is getting more and more prominent, how to break it?

Understanding Solana: PoH and Base Station Byzantine Fault Tolerance Algorithms

On one hand, the competition for decentralization and security is getting hot, on the other hand, scalability is also crucial.

In this dimension, Solana has emerged from the crowd.

Solana’s scalability is positioned at the network level and oriented to web3 to provide a more “silky” future experience for the blockchain world.

Another highlight of Solana is its low cost, confirming 1 million transactions for only $10.

The extreme performance is achieved thanks to Solana’s strong technical DNA and three-dimensional technical innovation.

Solana was founded in 2017 by engineers from Qualcomm, Intel and Dropbox.

The engineers wanted to achieve the vision of a decentralized network of nodes that could match the performance of individual nodes through technology. But all of this was predicated on the engineers properly optimizing the communication between nodes in the network. At the time, there was no effective solution to this problem, so Solana was created to achieve this goal.

After intensive research and development, the Solana team introduced eight key innovations.

Proof of Work History PoH

Base Station Byzantine Fault Tolerance (Tower BFT)

Turbine (block propagation protocol)

Gulfstream (memoryless transaction forwarding protocol)

Sea Level (Parallel Smart Contracts)

Pipeline (Validated Transactions)

Cloud Scatter (horizontally scalable account database)

Archive (distributed ledger storage)

Today’s article will focus on Proof of Work History PoH and Tower BFT (Byzantine Fault Tolerance), explaining from the underlying logic how Solana achieves a performance jump without sacrificing decentralization and security.

Proof of History of Work (PoH)
Under the reality of Layer1’s congestion and high gas fees, Layer2 seems to be a “different approach”, but its essence is to directly abandon the historical dilemma and introduce a new thing by removing the dross.

But considering the countless projects built on Layer1 have already taken root, the cost behind bridging, licensing, expansion and migration will be huge. Therefore, if the heyday is to continue, the launch of Layer2 will not solve all the problems.

Not only that, the launch of something new may also be accompanied by other hidden problems.

V God has expressed his concern about Layer2 on his Twitter, the difficulty is that for the incentive needs to do a lot of application layer aspects of processing, large-scale application is also one of the difficulties.

“Projects implementing sharding technology (sharding) may introduce new security risks in their blockchain, become more vulnerable to consensus attacks, and the risks of implementing sharding far outweigh the potential scalability benefits.” said Anatoly Yakovenko, CEO of Solana.

If the network is split at Layer 2, an additional attack vector is introduced. Hackers have more targets to control, and if one slice of the network is taken over by a hacker, a domino effect can easily occur, affecting not only the price of tokens, but also the outflow of users and nodes.

In order to solve the performance problem, Solana chooses to make every effort to promote the development of Layer1 scalability, purify the underlying and economic system for developers, and clear the obstacles. In the Solana ecosystem, developers will be able to focus on development, technology and applications.

So, how does Solana maintain high TPS without sacrificing decentralization and security?

Compared to centralized networks, decentralized trading networks are inherently “inert” in terms of performance. In a centralized network configuration, the core nodes have maximum authentication and unified rights, and the entire network marches at the same pace, autocratically but efficiently. The latest throughput figure for Ether is 25 TPS, while the centralized Visa is about 1500 TPS, and this is where such a significant gap comes from.

In a decentralized network, countless nodes with the same authority but in different time logics need to be transformed into the same “time zone” before they can reach a final agreeable conclusion, which naturally consumes more energy. As the number of nodes grows, congestion occurs.

Solana’s unique Proof of Work History (PoH) is the solution.

Solana’s PoH finds a verifiable, shared time, and SHA 256, Solana’s Verifiable Delay Function, is irreversible and can only be computed in one direction.

In PoH, the last output is used as the current input to SHA256, and the data to be written in is appended to the input, and so on, periodically recording each SHA 256 output and count, and the verification node gets the required time interval by verifying and repeating this computation process. Regardless of the local time generated by transmissions and changes, they are unified in SHA 256.

Solana’s average block out time is compressed to 400 milliseconds, with high speed without Layer2 and negligible fees.

With a new mindset of PoH, Solana creates a cryptographically secure and authentication-free time source.

With simple logic and a high degree of unification through decentralization, Solana reduces messaging consumption and optimizes the entire network as a whole, making it possible to surpass the performance of today’s best centralized systems.

Base Station Byzantine Fault Tolerance (Tower BFT)
While freeing up processing time from the underlying logic, Solana’s second innovation aims to leverage historical proof to significantly increase the throughput of transactions that the network can handle.

This is the base station Byzantine Fault Tolerance (Tower BFT) algorithm.

Being resistant to Byzantine failures and improving fault tolerance in practice is something that all public chains are thinking about. Benefiting from working history proofs, each Solana node can know both the elapsed time and the chronological order of events that occurred on the network. This makes the proof of history a “clock” before consensus is reached in the application, so that even if a broken node does not complete its “communication”, the transaction will still proceed according to consensus.

The following scenario occurs in operation.

Nodes with proof-of-history registers will achieve the same computation without communicating with each other.

Each hash of the PoH identifies a unique version (branch) of the register.

During the verification process, a vote is valid if there is a hash in the registry that the node has voted on.

This allows Tower BFT to guarantee that each verifier’s vote becomes valid by choosing a locking system, where there is a locking period (locking – measured in 400 ms time slots) before the result of the vote, during which the verifier also agrees to vote only on the fork of the fork he voted on, with a strong penalty mechanism of “cutting a portion of the verifier’s shares for non-compliance”.

In addition, Solana’s Tower BFT encourages verifiers to vote for legitimate blockchain forks, and with the rollback, reward and punishment mechanism makes it a unique asynchronous consensus algorithm.

Solana prefers vividness over consistency.

Because of Tower BFT, once the majority of nodes vote for a historical proof hash, the hash is “normalized” and cannot be restored to its previous state. At the same time, the asynchronous voting of nodes supports more derivations.

A simpler asynchronous consensus algorithm driven by an original proof-of-work history allows Solana to achieve record blockchain throughput.

Posted by:CoinYuppie,Reprinted with attribution to:https://coinyuppie.com/understanding-solana-poh-and-base-station-byzantine-fault-tolerance-algorithms/
Coinyuppie is an open information publishing platform, all information provided is not related to the views and positions of coinyuppie, and does not constitute any investment and financial advice. Users are expected to carefully screen and prevent risks.

Like (0)
Donate Buy me a coffee Buy me a coffee
Previous 2021-06-08 07:29
Next 2021-06-08 08:14

Related articles