Recently, I saw someone ask about Ethereum's scaling solutions, so I might as well organize my understanding.

The core problem Ethereum faces is actually quite simple—only able to process 15 transactions per second, which is a joke for a thriving public chain ecosystem. As a result, gas fees skyrocket, making it unaffordable for ordinary users, and it feels like a playground for the rich. But this is clearly not the original vision of Ethereum.

So how to solve this? Directly expanding the main chain blocks? No, that would cause node operation costs to skyrocket, making the network more centralized. Therefore, the Ethereum community ultimately chose layer 2 scaling solutions, with Rollups being the most core technology.

The logic of Rollups is actually quite clever—moving computation off-chain to a "Rollup chain," executing transactions there, and then submitting the data back to the main chain for verification. This reduces the burden on the main chain and significantly lowers gas fees. In theory, TPS can be increased from 15 to 1000, and gas costs reduced from thousands to hundreds.

However, Rollups technology itself also has two main camps, each with its own characteristics.

One is Optimistic Rollups, represented by projects like Optimism and Arbitrum. Their logic is "assume transactions are valid first," then give others a period to report fraudulent transactions. If someone submits proof of fraud, the system re-verifies on the main chain. This mechanism relies on a bond system to prevent malicious behavior—you need to deposit ETH to submit transactions, and if caught cheating, you forfeit it.

The problem is this process takes time. From submitting a transaction to final confirmation, it can take days or even two weeks, which can be inconvenient for users who need quick results.

The other is ZK-Rollups, which are more technically challenging but more efficient. They do not rely on "fraud proofs" but use zero-knowledge proof technology to verify computations off-chain and then submit the proof directly to the main chain. Transactions can be confirmed almost instantly, and data on-chain costs are further reduced.

But ZK-Rollups also have a critical flaw—lack of compatibility with EVM. This means applications on the Ethereum mainnet need to rewrite their code to migrate to Rollup chains. The ZK-sync team is working hard to solve this problem, and if they succeed, ZK-Rollups could truly become game-changers.

Currently, projects like Loopring, Hermez, Aztec, Starkware, and others are exploring the possibilities of ZK-Rollups in different ways.

Ultimately, Ethereum's future roadmap revolves around Rollups. First, the Merge, which combines the beacon chain and execution layer, followed by data sharding, essentially to accelerate the implementation of Rollups. Once Rollups mature, Ethereum can not only solve scalability issues but also maintain decentralization and security—something unprecedented in blockchain history.

At that point, low gas fees and fast transactions will attract a large influx of new users and developers, leading to a real explosion in Ethereum's DeFi ecosystem. That’s why I’ve always believed that Rollups are the key to whether Ethereum can truly become the global settlement layer.