Not Just Blockchain: How Stateless Infrastructure Enables Efficient Decentralized Networks
Introduction to the Importance of Stateless Infrastructure by Polynya, an active member of the Ethereum community
What is Stateless Infrastructure?
Not emphasizing the consensus of the network
Stateless infrastructure only requires one honest person
Why do we need Stateless Infrastructure?
Blockchain efficiency is always low
Products cannot be fully built on the blockchain
Stateless infrastructure is more decentralized
Stop blindly pursuing blockchain technology
While the meaning of statelessness has been widely discussed in the technical field, in this article, Polynya considers blockchain (including Layer2) as a carrier for achieving global consensus, meaning that all nodes have the same latest state and consensus infrastructure. Infrastructure without consensus and with loose consensus rules is referred to as stateless infrastructure.
Recommended reading:
Introduction to global consensus
To better understand this article and know which applications are truly suitable for blockchain, this article can be considered as the next episode, explaining what alternative solutions there are for applications that are not suitable for blockchain technology – stateless infrastructure.
For example, Ethereum is a stateful infrastructure where all nodes record the same ledger, and interacting with any node will yield the same result. It forms a strong consensus based on the longest chain, ensuring global consensus on information in the network. On the other hand, IPFS is a typical stateless infrastructure where each node records different information, so there is no consensus. When a file is referenced, it requires a node that has stored the data to be online in order to successfully retrieve the data. Additionally, IPFS nodes selectively delete data that has not been used for a long time based on data access frequency to improve efficiency. Although the IPFS network lacks consensus, it has relatively low costs and is suitable for storing data that is not very important compared to blockchain storage.
Other common stateless infrastructures include BitTorrent, Farcaster Hubs, and ZK Coprocessors.
When discussing stateless infrastructure, it is necessary to change existing perspectives. In the world of blockchain, people have already adapted to the assumption of honest majority and therefore run as many nodes as possible. However, outside of blockchain, most things operate based on the assumption of honest minority – as long as there is one honest party, everything will function normally.
For example, in IPFS, as long as one operational node is willing to provide certain data, that data is accessible without worrying about whether other nodes are willing to provide it.
Only one honest node is needed for operation, which means that regardless of the number of nodes or the execution speed, the efficiency will be much higher than the design architecture of blockchain, whether it’s Layer1, Layer2, modularization, etc. Stateless infrastructure is fundamentally different.
Currently, blockchain is highly inefficient, requiring tens of thousands of times more expenses compared to centralized servers for the same computation. Although with the help of new technologies such as proof of validity and data availability sampling (DAS), efficiency can be improved thousands of times, the overall computational cost still cannot surpass the architecture of centralized servers.
Blockchain cannot match the efficiency of centralized servers, no matter how much the technology is improved.
By using stateless infrastructure, industries can go further. With stateless infrastructure, parallel execution across multiple machines becomes much easier, exponentially scaling decentralized networks to a larger scale and efficiency.
A competitive and user-friendly product cannot rely solely on blockchain technology. Even for applications that have the greatest demand for global consensus, such as finance and identity, there are many components that require other infrastructure, such as most computations and data storage generated during the process, which are more suitable for centralized servers and IPFS storage.
Future products will be a combination of various technologies, rather than solely emphasizing blockchain technology.
Stateless infrastructure and blockchain are not replacements for each other, but complementary in achieving competitive decentralized applications.
By combining all these technologies, it becomes possible to create applications that were previously impossible.
In fact, stateless infrastructure can be more decentralized than blockchain.
Achieving strict global consensus on the blockchain is an extremely difficult process, requiring an expensive mechanism to resist attacks and the assumption of honest majority. The cost of achieving global consensus is that the existing block production mechanisms are at best controlled by the rich (PoS) or corporations (PoW).
Blockchain mitigates the above situation through node operation, but it still requires the assumption of honest majority and flexible consensus formation process with thousands of nodes.
On the other hand, although stateless infrastructure can take various forms, most of them are suitable for the assumption of honest minority – as long as one node is honest. Therefore, efficiency can be improved, leading to more decentralization.
If strict global consensus is not needed, peer-to-peer stateless infrastructure is more efficient and decentralized than blockchain.
It is time to stop endless pursuit of blockchain technology and consider what tools are suitable for solving real needs and problems, or combine various infrastructures to create more complete and competitive products.
What truly matters is building useful Web3 applications, rather than the most efficient blockchain.
global consensus
Polynya
Stateless infra
无状態基礎設施
Further reading:
What is global consensus? Which products or features are suitable for building on Ethereum?
Web3 startups and the dawn of fundraising? Japanese Cabinet approves new law, pending parliamentary review to take effect.
