The narrative logic of "infinite scalability"

Written by: Beichen, Chain Teahouse

Recently, many people have been paying attention to Near, and there are many reasons for this. Raising $350 million from top institutions, continuous financing for ecological projects, launching the algorithmic stablecoin USN… these actions at different levels all point in the same direction—Near is starting to take off.

We are no longer in the era of public chains, but Near's narrative is that of Web 3.0, often mentioned alongside technologies like Ethereum 2.0 and heterogeneous cross-chain that have not yet truly arrived, so Near does not need to align its pace with other public chains.

But can Near really take off smoothly? After all, the narrative of Dfinity, which is even grander than Near's, serves as a painful lesson.

1. Sharding as Direction

On April 22, 2020, Near's genesis block was officially launched. Compared to those Ethereum sidechains, this is a blockchain network built from scratch based on sharding technology, focusing on high performance and high security, with infinite scalability. Of course, these have not been fully realized yet.

Near's high performance is due to the use of the PoS proof mechanism, while infinite scalability is achieved through sharding technology.

The advantages of the PoS proof mechanism need not be emphasized; the high-performance emerging public chains like BNB Chain and Solana that have taken off in the past two years have risen precisely because of this. This efficiency, which delegates all computations to a small group of nodes, can be very high, but they do not possess infinite scalability because the computational capacity of nodes has an upper limit. In fact, both BNB Chain and Solana have experienced incidents where they had to pause operations due to node failures.

Thus, sharding technology is particularly necessary, with the solution being to divide a single chain into countless chains that can independently process their own data, thereby alleviating the pressure on a single chain.

The idea behind Ethereum 2.0 (now called "Ethereum Upgrade") is consistent with Near's, combining the PoS proof mechanism and sharding technology, but it is still far from implementation (most likely not achievable in 2023), and the specific technical roadmap has yet to be determined.

Near has already been implemented, although it is still in the second phase and cannot achieve the initially claimed infinite scalability and TPS in the millions, it is indeed progressing steadily according to plan.

2. Rapid Iteration of Product Roadmap

In fact, this is precisely what makes the Near team commendable—they have not been trapped in grand narratives and delayed their launch, but instead followed the MVP principle (Minimum Viable Product), prioritizing the launch of the most basic functionalities and adjusting the product in a timely manner based on feedback.

Thus, Near first built a simple internal test network, attracting developers to test through hackathons and seminars, and later gradually launched the public test network and mainnet.

The Near team received a lot of feedback from developers during the testing phase and realized that the initially designed complete sharding solution was not realistic, as a public chain does not need such a large capacity right after launch (in fact, this is still true today). Therefore, the adjusted new plan is more pragmatic—divided into four phases for completion.

The mainnet was officially launched on April 22, 2020, but at this time, there was no sharding, making it no different from other public chains.

It wasn't until November 15, 2021, that the first step of sharding, Phase 0, was taken, also known as "Simple Nightshade." The sharding in Phase 0 of Near is incomplete, more form than substance, as it merely divides the Near blockchain into four shards, and it is only state sharding, meaning each shard only stores all data within its shard, while actual computations are still handled by all nodes. This is no different from a single chain, as true sharding should have its own nodes for each shard.

In early 2022, Near entered Phase 1, introducing a new role called Chunk-Only Producers (pure block producers), who only provide validation services for specific shards, equivalent to shard nodes/miners. Note that at this time, only state validation is provided, and they are not responsible for actual computation. Therefore, the entry threshold is relatively low, and it can run on hardware with a 4-core CPU, 8GB RAM, and 200GB SSD storage. Additionally, the staked NEAR is relatively low, allowing mining by directly purchasing the corresponding services on AWS.

Near is still in Phase 1, but plans to launch Phase 2 in the third quarter of 2022, at which point not only state sharding will be implemented, but computation processing will also be sharded. Only then will Near be a true sharded network, but it still has not achieved the initially claimed infinite scalability, as it still requires dynamic allocation of nodes among the various shards.

It is expected that by the fourth quarter of 2022, Near will enter Phase 3, which will truly realize dynamic sharding, meaning shards will be dynamically split and merged based on resource utilization. In other words, the number of shards in Near will be adjusted according to the situation.

3. Replacing Beacon Chain with "Blocks"

Near's sharding is not like Ethereum 2.0's model of beacon chain + shard chains (similar to Polkadot's relay chain + parallel chains), but directly divides the mainnet into countless shards.

Therefore, it can also be said that all shards constitute a chain (called the main chain), and the "blocks" produced by the main chain contain the "trunks" (the blocks verified by pure block producers) of each shard. The "block" is the global state composed of all "trunks."

The reason Near did not choose a beacon chain to act as a coordinator is that a beacon chain itself is a chain, and as long as it is a chain, there is an upper limit to computational capacity, so the number of shard chains is limited (the limited slots of Polkadot and the limited shard chains of Ethereum 2.0 are due to the same reason).

Since Near does not have a beacon chain as a coordinator, how do shards communicate with each other? The "blocks" mentioned earlier play a role similar to that of a beacon chain to some extent.

Each "block" of the main chain has corresponding validators (pure block producers) on each shard. When validation is needed, the pure block producer on the shard can directly verify the "block" on the shard against the "block" of the entire main chain.

Looking at the sharding solution from the perspective of contracts may be more intuitive—if an operation involves multiple shards, it needs to be validated sequentially on each shard. First, a new "block" is formed on the first shard, then passed to the second shard through the "block," forming the second "block," and so on.

In summary, to solve the efficiency bottleneck of blockchain networks, sharding technology allows multiple nodes to validate transactions across multiple shards. Near's sharding technology is similar but goes further—communication between shards does not require a master chain, but rather nodes directly verify through "blocks," thus not being affected by the performance limits of a master chain.

Additionally, nodes on the shards only need to validate the portion of the main chain "block" that corresponds to their own "trunk," thus improving efficiency.

4. Replacing "Functional Sharding" with "Dynamic Sharding"

The biggest selling point of Near's sharding technology is infinite scalability. The earlier mentioned replacement of the beacon chain with "blocks" leads to no upper limit on scalability, which is a necessary condition but not a sufficient condition, as having no upper limit does not mean it can truly break through the limit, just as no one stopping you from becoming the world's richest person does not mean you can actually become the world's richest person.

The true technological innovation that brings infinite scalability is dynamic sharding, but before formally discussing dynamic sharding, let's take a look at other sharding solutions. Polkadot is essentially also a sharded network, and this article will discuss it as well.

Polkadot's solution separates state and applications with a relay chain + parallel chains, where the relay chain is responsible for maintaining network state, while other parallel chains focus on more customized application chains.

Ethereum 2.0 also follows the beacon chain + shard chain model, with the confirmed plan being that the Ethereum mainnet will become the beacon chain, serving as the consensus layer coordinating the network. Additionally, the first phase's goal has been set—to be data availability-oriented, with shard chains essentially playing the role of the current ZK-based Layer 2, achieving TPS of up to 100,000, with security equivalent to that of the Ethereum mainnet.

The second phase's goal for Ethereum 2.0 is code execution-oriented, where each shard chain can store and execute code and process transactions, and each shard will contain a unique set of smart contracts and account balances, akin to Polkadot's current solution.

However, there are divergences in the community regarding the second phase's goals, and the specific technical solutions have not yet been determined. Vitalik summarized three possibilities:

1. Shard chains do not need state execution and merely act as databases, essentially remaining in the first phase;
2. Adding state execution functionality to only a few shards, with a total of 64 shards planned. This is a compromise between the first and second phase goals;
3. To be determined later… The specific timing depends on when "succinct non-interactive zero-knowledge proofs" are validated, as they can bring smarter sharding, which is clearly still under research.

Thus, it is evident that Ethereum 2.0's sharding is still far from actual implementation, even though the official timeline is set for 2023.

Now, looking at the innovations in Near's sharding technology becomes very clear.

Near's paradigm is completely different from any of the aforementioned ones. The commonality between Polkadot and Ethereum 2.0 is that each shard is an independent blockchain, while Near breaks down shards, shard nodes, and smart contracts, dynamically recombining these fragments.

Again, looking from the perspective of contracts may be more intuitive—first, Near's contracts will be divided into many fragments, which will be allocated to different shards (dynamically assessed based on different shards' transaction volumes and storage usage). When the resource quantity of a shard exceeds a specific threshold, new shards will be added (so theoretically, the number of shards is infinite), and conversely, they will also be reduced.

Thus, in Near's sharded network, there is no fixed number of shards, no fixed nodes, and certainly no fixed nodes on the shards. If a chain is divided into 10 shards, the efficiency might be 1×10 times, while Near's sharding technology achieves an efficiency of 1×10×10 times.

5. Half-Pushed, Half-Pulled Operational Logic

Near's positioning determines that it will not be content to become an Ethereum sidechain, but the current state of the blockchain ecosystem must acknowledge that Ethereum has accumulated the highest quality developers, users, and funds.

The rise of emerging public chains relies on capturing the value overflow from Ethereum; whether it is Tron or the later BSC and Solana, none have missed this wave of dividends (of course, long-term operations follow another logic). Thus, the newly awakened emerging public chains have also launched Ethereum EVM compatibility, which, compared to the initial grand vision, has essentially devolved into Ethereum sidechains, a necessary compromise.

Near seems to have only recently begun to confront the gap with Ethereum, launching a cross-Ethereum bridge in April 2021 and the Ethereum Virtual Machine (EVM) Aurora in May, which is also a Layer 2 built on the Near protocol.

The reason Aurora is based on the Near protocol is that its underlying layer uses Near's validators, thus its speed and gas fees are consistent with Near.

For Near, the role of Aurora is to bring Ethereum's resources to this Layer 2, seamlessly transplanting them into the Near ecosystem. Additionally, with the support of the Rainbow Bridge, the multi-chain experience between Ethereum, Aurora, and Near becomes even smoother.

Although Aurora missed the biggest wave of Ethereum's overflow, its performance after launch, whether in terms of ecological projects or recognition from investment institutions, has been quite impressive, mainly due to the team's operational capabilities. If Near's sharding progresses smoothly in the future, Aurora's technical advantages will continue to grow.

The only question is, is Aurora really an Ethereum Layer 2? To be honest, it resembles an Ethereum sidechain more, just like Polygon, which also claims to be an Ethereum Layer 2 but is essentially a sidechain disguised as Layer 2.

Although Near has begun to focus on developing its ecosystem, the rhythm of the ecosystem must follow the rhythm of technology, and according to Near's technological pace, it is clearly not yet time for a real push. However, based on Aurora's performance, we have reason to believe that by then, Near's operational level will at least be comparable to that of Aurora, and only then can it bring about Near's true ecosystem.

6. Web3-Oriented Octopus Network

If the launch of Aurora was a necessary move for Near to cope with the current situation, the Octopus Network is a proactive layout for the future's "backup."

Like Aurora, Octopus is also based on the Near protocol, using Near's validators at its core, thus being almost identical to Near in terms of speed, gas fees, and security.

Octopus does not directly serve projects within the Near ecosystem but is a network that exists parasitically on Near, with its own ecosystem and logic.

To put it simply: Octopus attempts to create a Polkadot (or Cosmos) and treats Near as a relay chain (or Hub).

First, the services provided by Octopus are similar to those of Polkadot, offering a series of out-of-the-box technical solutions based on Substrate, allowing developers to launch chains with one click, and these are customized application chains (governance structures, economic designs, underlying consensus algorithms, etc.).

Octopus Relay plays a role similar to that of Polkadot's relay chain slots, responsible for the security of these customized application chains, and requires payment for rental. However, Octopus Relay is not a chain; it is essentially a set of smart contracts running on Near.

Having solved the security issue, there is still a need to address the cross-chain requirements between these customized application chains and other ecological blockchains. The medium provided by Octopus is various rainbow bridges, which will fully connect the Ethereum, Near, and Cosmos ecosystems in the future.

As a competitor to Polkadot, Octopus's advantage is that it is cheaper (Octopus's application chains cost less than $100,000, while Polkadot's cost is 40 times that), and there is no limit on the number.

Additionally, considering the Octopus team's extensive cross-chain experience, the future ecosystem may be more open than Polkadot, although this is a bonus; the core remains the overflow value of its own ecosystem.

It is not difficult to see that Octopus has its own logic, and its narrative runs parallel to Near's core narrative. Near clearly saves Octopus a lot of trouble (a set of smart contracts can act as a relay chain), and Octopus also brings potential greater space to Near, a space much larger than that of Aurora, as the latter only captures the value overflow from the Ethereum ecosystem.

Conclusion

Near's core narrative revolves around infinite scalability, which is brought about by its unique and even disruptive sharding technology, Nightshade Sharding—uniquely manifested in replacing the beacon chain with "blocks," and disruptively breaking down the common concepts of sharding, shard nodes, and smart contracts, dynamically reshaping a new type of blockchain network in an atomic manner.

Beyond sharding technology, Near's product roadmap is also very clear—without getting trapped in grand narratives and delaying the launch, it is achieving its blueprint through rapid iteration in phases.

In addition to the product roadmap, operational capability is also crucial. Near's operational rhythm follows the product roadmap, and it is not yet time for a real push, but we can already sense the team's operational level from Aurora's development.

Moreover, Near is also proactively laying out a grand cross-chain network that has yet to arrive, with not only Aurora, which can draw traffic from Ethereum, but also the Octopus, which can benchmark against Polkadot, and it will target all Cosmos (based on IBC) ecosystems.

Of course, Aurora and Octopus are not Near's core narrative; Near's long-term value ultimately depends on its core narrative—the infinite scalability arriving by the end of 2022 and the subsequent operational performance.

If Near's core narrative is realized, then Aurora and Octopus can draw traffic; if it is not realized, then Near will form a reverse siphoning effect with other ecosystems, becoming the one being drained instead.