Sequencers: Enhancing Blockchain Scalability

How Sequencers Manage the Flow of Transactions

A sequencer is specialized software acting as an intermediary node between the L1 and L2 of the Ethereum blockchain. It organizes and groups transactions before they are added to the mainnet. 

Imagine every information transfer in a decentralized network as notes on a musical score and L2 blockchains as musical instruments; the sequencer is akin to an orchestra conductor, synchronizing musicians to produce a unified melody.

Here’s how it works in practice:

1. Users initiate transactions on any L2 blockchain (such as Optimism, Arbitrum, Polygon, etc.), which then go into a mempool (a waiting list).
2. Transactions from the mempool are passed to the sequencer—a specialized node that prepares them for execution.
3. Inside the sequencer, transactions are verified for accuracy and arranged in chronological order.
4. Once organized, the sequencer sends the transactions as a single block to the Ethereum network.

Illustration of Data Flow in the Ethereum Ecosystem. Source: jarrodwatts.com

Sequencers don't store information; they only process it. By taking some of the workload off the main blockchain, sequencers significantly improve network performance. 

They fulfill several critical tasks:

• They process more transactions, enhancing network throughput.
• They offer additional protection against double-spending by independently verifying the transactions.
• They facilitate consensus among network nodes by synchronizing transactions.

Types of Sequencers

Sequencers vary in type and capacity, with each L2 network employing its own unique intermediary node to verify and relay transactions to the main blockchain.

1. Centralized sequencers are controlled by a single entity and are known for their high-speed data processing capabilities. Nodes of this type are present in blockchains like Optimism and Arbitrum. There are two significant risks associated with centralized sequencers. The first is known as Miner Extractable Value (MEV), which suggests that validators of these nodes might theoretically manipulate transaction orders for their benefit. The second risk is a single point of failure, where if a centralized sequencer fails or malfunctions, the entire transaction verification system could collapse. 

Arbitrum Sequencer Operation Diagram. Source: kroma.network

2. Decentralized sequencers are spread across numerous nodes. They increase the security and robustness of the network but might be less efficient. Networks such as Metis use such sequencers, where several randomly chosen P2P validators independently verify and sequence transactions, thus upholding network consensus. This decentralized approach is resistant to single points of failure, manipulation attempts, and is nearly immune to MEV. However, the downside is that transaction processing times are longer.

Decentralized Sequencer Operation Diagram. Source: Metis

3. Shared sequencers aggregate transactions from various L2 blockchains, enhancing their compatibility and improving the liquidity of pools. Essentially, they form a separate network of decentralized nodes for verifying and “packaging” transactions, accessible to every user in the Ethereum ecosystem. An example is the Espresso platform, which offers independent decentralized transaction synchronization as its main service. The advantages of such networks include lower fees and the possibility for cross-arbitrage across different DEX platforms. However, the challenges include integration complexity and compatibility issues between different L2s.

Espresso Shared Sequencer Operation Diagram. Source: HackMD

As blockchain technology continues to strive for maximal scalability, the importance of sequencers will inevitably grow. Although new technologies for optimizing transaction processing will emerge, choosing between them will likely always involve balancing decentralization with speed.