The birth of Solana is not only an innovation based on underlying consensus for Web3.0, but also a return to the original intention of building a decentralized "world computer".

Why do public chain projects emerge in an endless stream and continue to be popular? On the surface, the reasons are market demand and profit-seeking, but if you dig deeper, you will find that there is still a huge performance gap between existing blockchain projects and the huge market demand.

 

Blockchain is a solution: it provides a unique, standardized, and unchangeable ledger for orderly and valid transactions. In short, blockchain technology has three very basic characteristics: anti-censorship, permissionless, and trustless. However, many mature blockchain projects still have huge bottlenecks at the consensus level, or face three major problems, namely, security, decentralization, and scalability cannot be achieved at the same time. Many teams are exploring and trying to capture value at the consensus level of blockchain.

 

If we think from the perspective of changes in consensus protocols, we can see how existing public chains handle the relationship between time and state. Grisha Trubetskoy traced all the value attributes brought by blockchain technology back to the beginning of the invention of Bitcoin, and proposed that the essence of the underlying consensus is to solve: How to implement a timing system in a distributed system?

 

Blockchains represented by Ethereum and Bitcoin all use the clock system of traditional blockchains, that is, time and state are coupled, and only the emergence of new blocks can produce a globally consistent state. Solana's innovation is to separate time and state updates, provide a license-free, globally available and trust-minimized clock system for its smart contract platform, and optimize network operation before consensus is reached.

 

In Solana's clock system, time and state are not coupled or bound, the timestamps of transactions on the chain are encoded, and transactions flow between networks like water. Solana calls this core innovation "Proof of History" (POH).

 

On the one hand, the vision of a decentralized "world computer" has not yet been realized due to many technical bottlenecks; on the other hand, in the future, it is necessary to realize a decentralized Internet 3.0 that focuses on privacy and user ownership. An article written by Kyle Samani, co-founder of Multicoin Capital, tells us that these two visions are not logically contradictory, but complement each other, linking the past with the future.

 

The birth of Solana is not only based on the innovation of the basic consensus of Network 3.0, but also the initial return of building a decentralized "world computer".

 

First, Solana’s answer to the “world computer” problem

 

For blockchain developers, Solana is the closest thing to a “world computer,” an early concept in the crypto world. Solana is one of the most attractive Layer 1 platforms we have evaluated so far, and we encourage developers around the world to seriously consider Solana. — Kyle Samani, co-founder of Multicoin Capital

Solana proves that by separating time and state, a "world computer" with minimal trust and no permission can be achieved. As Kyle Samani, co-founder of Mulicoin, said, Solana is one of the most attractive layer 1 platforms. The following article will demonstrate from several angles that Solana is the "world computer project" closest to blockchain developers.

 

History proves that speed is paramount - underlying performance is the foundation for all superstructures

 

Solana wrote on its official website: History has proven that speed is paramount. The high-performance blockchain underlying platform provided by Solana is the foundation of all superstructures.

 

 

It is very important to understand the performance of distributed systems from the "time dimension". Time is everything. Through the new thinking of PoH "encoded time", Solana has created a trustless cryptographically secure time source, reduced the overhead of information transmission, optimized large-scale networks, and made the performance of unlicensed distributed systems comparable to that provided by current centralized cloud computing.

 

Specifically, network systems such as Bitcoin and Ethereum can support about 10 transactions per second (TPS). Taking Mint as an example, the PoS system based on practical Byzantine fault tolerance (PBFT) can support about 1,000 TPS when the number of nodes is between 100-200.

 

Solana has a throughput of over 50,000 TPS on a test network consisting of about 200 validators on five continents, with an average TTF (final time) of 1.5 seconds. This is basically comparable to the best global distributed database spanners, but Solana has considerable decentralization.

 

Layer 1 - Abstract complexity; Layer 2 - No fragmentation (reject fragmentation)

 

The scalable Layer 1 underlying network with simple logic and minimal trust actually logically abstracts complexity, allowing application developers to focus on application logic.

 

Expecting scalable solutions and optimized application scenarios for Layer 2 actually increases development complexity and friction between users, developers, and service providers.

 

Solana named its official podcast “Unsharded,” which sounds a bit radical, but Solana’s view on Hardin as a second-layer solution also reflects Solana’s attitude: do everything possible to make the first layer scalable.

 

  

Solana CEO Anatoly Yakovenko said that projects implementing sharding technology may introduce new security risks to their blockchains, which are more vulnerable to consensus attacks, and the risks of implementing sharding far outweigh the potential scalability advantages. Anatoly believes that if the network splits at Layer 2, additional attack vectors will be introduced. He also said that if a shard in the network is taken over by a hacker, it may cause a domino effect, affecting the token price, causing a large outflow of users and nodes, and allowing hackers to use decentralized strategies to destroy the entire network ecosystem.

 

When Vitalik Buterin presented Ethereum to the world in January 2014, he emphasized exactly this point: the point of the world computer is to abstract away everything that is not application-specific. However, Vitalik’s ideal remains unfinished. On August 21, 2019, Vitalik tweeted that he was pessimistic about the expansion solutions under the second layer chain because many application layers have to deal with incentive mechanisms and are difficult to apply on a large scale.

 

 

In fact, in most use cases, developers building smart contracts may not want to deal with Layer 2 or fragmentation, and Solana blockchain meets this need. Developers on the Solana blockchain do not need to consider how to scale the underlying logic, because Solana provides a more concise and practical option: without sacrificing the security and simplicity of the Layer 1 network, it can achieve higher performance processing speeds than Layer 2 scaling solutions.

 

Solana abstracts all the complexity of the underlying and economic systems to Layer 1 itself, allowing developers to focus on the application development logic that they should focus on.

 

The underlying blockchain platform for developers

 

For blockchain developers like DApp, Solana has all the potential attributes to help them develop successful software:

 

  • High throughput: Testnet reaches 50,000 Tps in a global network of 200 nodes

  • Low latency: ~1 second finality

  • Low transaction fees: usually measured in cents

  • Programming language: Rust is its flagship programming language, and it also supports C, C++, and Libra's Move language

  • Asynchronous algorithm: Asynchronous Byzantine fault-tolerant consensus algorithm Global state: Supports global unified state for composable smart contracts

It is worth mentioning that Facebook's Libra team has created a new VM and programming language Move. Although Libra will not be programmable when the mainnet is released in 2020, from the Libra team's open source code base, Solana developers have found that Move and Solana's Pipeline VM have many similarities.

 

 

Among them, the Move language introduces Resources and scripts as high-level concepts. Both can naturally adapt to the design of Solana pipeline runtime and local programs. Solana's goal is to use Move as the first language to support native Solana programs, and it can be interactively developed and combined through the Move language and Solana's native Rust language without affecting performance or security.

 

Solana previously announced support for the Move language, including BFT and parallel transaction processing on GPUs. This means that projects and applications built with the Move language are compatible with Solana, and developers can easily integrate applications written for Libra into the unauthorized Solana chain, sharing the transaction speed and strong scalability of the Solana network.

 

Solana can not only benefit from the development channel based on Libra, but also run Libra programs in a completely permissionless manner. According to the current process, Solana's mainnet will be launched in October 2019, and Solana is likely to become the first blockchain that truly supports applications based on the Move language.

 

Second, how does Solana lead the Web 3.0 wave through innovation based on the underlying consensus?

 

I’m not sure whether privacy is necessary for Defi. Maybe some people think that privacy is not a necessity, and what we need to do is to give users the right to choose. - Anatoly Yakovenko, CEO of Solana

 

 

In the Web 2.0 era, the ownership of user data belongs to the platform, which brings risks such as platform data leakage and user data selling. The vision of the Web 3.0 era is to return the rights to users and let users control their own data. The giants who monopolize data may be overthrown.

 

The concept of Web 3.0 was first proposed by Gavin Wood, former CTO of Ethereum and founder of Polkadot, in April 2014. Compared with the traditional Internet, Web 3.0 aims to create a new decentralized Internet solution that benefits from its decentralized core and has privacy and anti-censorship features.

 

Gavin Wood once said that privacy protection is the core of Web 3.0. When talking about the importance of privacy in the Web 3.0 era, Solana CEO Anatoly gave another answer in an interview: "I'm not sure whether privacy is necessary for Defi. Some people think that privacy is not necessary, but what we have to do is to give users a choice."

 

Kyle Samani, co-founder of Multicoin, a well-known American currency investment institution, wrote an article at the end of 2018, analyzing in detail the ecological panorama of the stack under the perspective of Web3.0, describing how various decentralized technologies are stacked layer by layer to form a complete system architecture. Solana was included in the core stack of Web 3.0 due to its unique innovation in the consensus protocol layer, which gave investors great expectations and attracted great attention and discussion.

 

A year later, as the Solana ecosystem continued to expand, the test network was fully launched. Kyle Samani once again released "Opening Web3.0 Unbinding Centration Vector" in 2019, in which he mentioned that he believed that Web 3.0 would become a paradigm shift and unlock trillions of dollars in value in the next decade. After a year of thinking and precipitation, he pointed out that the pain point of realizing the Web 3.0 vision lies in the separation of data ownership and application logic.

 

Unbinding centralized databases to solve the availability problem of large-scale data

 

In the Web 2.0 era, the platform stores and manages data for users, creating a large-scale storage room and a 24-hour online server. However, in the Web 3.0 vision, data is not moving towards centralized storage, and there is no centralized platform data provider. Therefore, the paradigm of data ownership needs to change, and the unbinding of data ownership is one of the basic pain points of Web 3.0.

 

This pain point coincides with Solana’s innovation, especially Solana’s efforts to solve data availability issues on a large scale through innovations such as Replicator.

 

In the last article, we also introduced how Solana innovatively created a division of labor between two nodes through validators and replicators, and generated lightweight proofs to optimize data storage.

 

The large-scale data of the blockchain power grid needs to be solved urgently. At present, the blockchain network will generate 4 PB of data for the general ledger every year at a rate of 1 GB per second. At this rate, the storage of blockchain data will soon become the main centralized carrier, and the cumbersome centralized database direction may become one of the stumbling blocks to realizing the decentralized vision of Web 3.0.

 

Validator node: responsible for verifying network data, but PoH historical proof and PBFT tower help it increase the verification speed.

 

Replicator nodes: Obtain computational weights from validator nodes and operate with minimal hardware requirements.

 

The hardware requirements for replicator nodes are not high and all of our laptops can perform them on a daily basis. However, replicator nodes play an important role in the network by optimizing distributed data storage systems aimed at solving the data availability problem of processing petabytes of data.

 

Solana's replicator nodes do not need to participate in consensus and store the entire data history, but use multiple replicator nodes to store small pieces of data history respectively, thereby generating lightweight proofs and performing erasure code functions, thereby dividing the entire state history into multiple parts. This proof can replicate all parts of the ledger and allow validators to batch verify across graphics processors.

 

The concept of PoRep used by Solana is mainly borrowed from Filecoin, using the timing source provided by PoH consensus to optimize the creation method of PoRep. Solana uses two types of node division of labor to generate lightweight proofs, untangling cumbersome centralized data storage and changing the paradigm of data ownership.

 

Decouple applications from architecture to achieve logic centralization + architecture decentralization

 

Vitalik, the founder of Ethereum Museum, explained in detail the concept of blockchain "decentralization". In his opinion, when people discuss decentralization, they are actually discussing three independent dimensions, but in some cases, these three dimensions are also interdependent:

 

Decentralization at the architectural level: How many physical computers does the system consist of? How many computers can the system withstand crashing at the same time? Decentralization at the control level: How many individuals or organizations have actual control over the system? Decentralization at the logical level: Do the interfaces and data structures presented and maintained by the system look more like a whole, or a non-whole?

Web2.0 solves all of the above problems through monopoly, logically relying on centralized storage. The problem with this model is that Web2.0 systems are coupled with all forms of centralization: they are not only logically centralized, but also politically and architecturally centralized.

 

Kyle Samani, co-founder of Multicoin, gave his solution in a recent article: Providing a single logically centralized interface for a politically and architecturally decentralized system, and concluded that the biggest challenge in the transition from Web 2.0 to Web 3.0 is to move from a coupled system with three centralization vectors (logically, architecturally, and politically) to a logically centralized but architecturally and politically decentralized system.

 

As we mentioned earlier, developers based on the Solana blockchain do not need to consider how to expand the underlying logic. Obviously, Solana abstracts the meaning of the complexity of the underlying and economic systems of the first layer itself through expansion. According to this logical thinking, the innovation of high-performance first-layer networks represented by Solana+Replicator to solve the data availability problem is actually an effort towards Web3.0 solutions (logical centralization + architectural decentralization).

Read 23