用大白话讲清楚 Rollup 怎么让以太坊变快
TechFlow Selected TechFlow Selected
用大白话讲清楚 Rollup 怎么让以太坊变快
Rollups can ultimately enable Ethereum to have a large enough market (a big enough pie) and the ability to consume that pie: they exponentially unlock significantly more on-chain capacity without compromising the network's decentralized security.
Navigating Web3 tides with focused insightsAuthor: Benjamin Simon, Researcher at Mechanism Capital
Translated by: Perry Wang
Let’s start with an analogy.
Imagine we're back in middle school math class. Our teacher—just as ruthless as the one you once had—hands us a list of one hundred long division problems, each involving very large numbers. We’re told our task is to solve as many as possible. But here's the catch! We must write down both the final answers and every single step of calculation on just one sheet of paper. Minutes later, with a sinking feeling in our stomachs, we realize this task is absurdly impossible. Only a few complete solutions can fit on a single page.
What does this Kafkaesque story have to do with Rollups?
In this analogy, that single answer sheet represents an Ethereum block, and the math problems are smart contract transactions. Ethereum today is extremely congested. There are simply too many transactions to fit into each block. Worse yet, most of these transactions are computationally intensive—like flash loans or trades routed through aggregators. The days of simple transfers and payments from the pre-DeFi era are long gone.
And here lies the crux. Just as the math teacher forces us to squeeze every grueling line of long division onto a single sheet, Ethereum must process and record every line of computation for every transaction.
This is where Rollups come in.
Rollups move computation off-chain while storing only minimal transaction data on-chain. In this fundamental sense, they act as scratch paper for Ethereum. Rollups handle all the messy computational work, enabling exponential scaling so that thousands of smart contract transactions can be batched into a single Ethereum block.
This analogy helps illustrate the problem Rollups aim to solve (“network congestion due to computational overload”). It even gives us a vague idea of what Rollups need to do (“move computation off-chain”). But to understand how Rollups actually work, what they look like, and why they excite people like me so much—we need to dig deeper.
What Is a Rollup?
A Rollup is essentially a separate blockchain, but with modifications. Like Ethereum, Rollup protocols have a “virtual machine” (EVM) capable of executing smart contract code. The Rollup virtual machine operates independently from Ethereum’s own VM, but it is governed by smart contracts on Ethereum. This connection enables communication between Rollups and Ethereum: Rollups execute transactions and process data; Ethereum receives and stores the results.
At a technical level, the key difference between a Rollup chain and more traditional blockchains lies in how new blocks are produced.
Typically, blockchains are maintained by a distributed network of participants (“miners” or “validators,” depending on the type of chain). These parties collectively produce blocks through consensus. Simply put, they vote on how to process a set of transactions—or how to construct the next block—and the block receiving majority support becomes permanently written onto the chain.
By contrast, Rollup chains don’t operate under majority rule. Instead, any party monitoring the state of the Rollup can send a so-called “assertion” back to Ethereum stating how transactions should be processed. Crucially, Ethereum can accept or reject this assertion regardless of whether a majority of other Rollup participants support it. In practice, this usually means one designated party within the Rollup is responsible for processing transactions and generating blocks.
Wait… Are Rollups Centralized Solutions?
The centralized nature of Rollup block production is part of what makes them so efficient at processing transactions. But it also raises an obvious and troubling question: Without majority consensus, how does a Rollup ensure correctness in block production? What happens if the block producer turns out to be malicious?
This centralization confuses crypto users accustomed to consensus-based blockchains. Indeed, if the story ended here, we might (incorrectly) conclude—as Avalanche co-founder Kevin Seqniqi recently suggested on Twitter—that Rollups are nothing more than “a database copied by a single entity.”
In reality, as we’ll see, the accusation of centralization is less wrong than it is logically misleading. Normally, a centralized blockchain without strong consensus mechanisms is highly vulnerable to corruption and malicious takeovers. But in the unique case of Rollups, lack of decentralization doesn’t compromise security or integrity. To understand why, we need to go deeper.
The Critical Role of Data Availability
Recall our earlier math homework dilemma. Our lifeline was having scratch paper available for calculations, so we only needed to write the problem and its answer on the final answer sheet. Rollups serve exactly this role: we said they “move smart contract computation off-chain while storing minimal transaction data on-chain.”
In fact, this last piece—the on-chain storage of transaction data—is crucial to how Rollups function. In a Rollup, only computation (data processing) occurs off-chain. Every transaction processed by the Rollup still has its input data (formally called “calldata”) stored on Ethereum.
Why is keeping transaction data on-chain so important? In our math analogy, the final answer sheet we hand in includes both the long division problems and their answers. Our teacher can verify our work afterward—even though the individual steps aren’t shown. Similarly, the persistent availability of on-chain data means Ethereum’s base layer can re-execute any computation performed on the Rollup.
In short, Rollups’ on-chain data availability enables a built-in auditing process. Ethereum can “double-check” the integrity of transactions processed on the Rollup before permanently writing them to the ledger—much like the U.S. Supreme Court’s power of judicial review.
The Power of Constraints
Thus, the real value of Rollups lies in their constrained nature. Rollups can only push transactions to the base layer—they cannot force acceptance, because Ethereum can overturn any Rollup transaction when necessary. Because they are subject to this audit process, Rollup transactions do not have true finality.
Given this constrained nature, we can now return to the unresolved charge of centralization. A single block producer in a Rollup might attempt to maliciously process transactions. But if so, Ethereum would simply reject that batch after the audit process completes, and the block producer would be penalized.
Now, the exact mechanism of the “audit process” depends on whether the Rollup uses Optimistic or zero-knowledge technology. But for both types, the audit process is exponentially more efficient than Ethereum processing those transactions natively without a Rollup.
In summary, the Rollup system operates on principles of “checks and balances” to ensure Ethereum remains the sovereign chain—Ethereum’s own consensus is the ultimate arbiter of truth. But critically, this differs from sidechain solutions. Lacking the same audit process, sidechains process transactions via their own fully independent consensus mechanisms. Sidechain transactions are “final,” whereas Rollup transactions are not (until confirmed on Ethereum). Therefore, sidechains require stronger trust assumptions, as they cannot benefit from Ethereum’s decentralized security. In fact, I believe sidechains resemble EVM-compatible Layer 1 solutions more than they do Rollups.
Conclusion
To summarize: Rollups scale by moving computation off-chain to free up on-chain space. On-chain data availability is essential because it allows Ethereum to verify the integrity of Rollup transactions. This audit capability acts as a “check” on Rollup block production, eliminating the need for internal consensus mechanisms.
Ultimately, Rollups allow Ethereum to have its cake and eat it too: they exponentially increase on-chain capacity without compromising the network’s decentralized security. At least in my view, this is about as elegant a scaling solution as we could hope for.
Thanks to t11s and Hasu for their insightful feedback and suggestions.
Join TechFlow official community to stay tuned
Telegram:https://t.me/TechFlowDaily
X (Twitter):https://x.com/TechFlowPost
X (Twitter) EN:https://x.com/BlockFlow_News
