Transaction Ordering Inside the Execution Layer

The Merge decentralized consensus but centralized execution. The transition to Proof-of-Stake created a robust, decentralized finality gadget for the L1 chain. However, the transaction ordering process within the block—the mempool mechanics and block builder selection—remains a centralized, opaque, and extractive market dominated by entities like Flashbots.

MEV is the execution layer's core dynamic. Maximal Extractable Value is not a bug; it is the primary economic incentive for block construction. This creates a zero-sum game between users and searchers, where transaction order directly determines profit. Protocols like CowSwap and UniswapX now build intent-based systems to bypass this adversarial landscape.

Proposer-Builder Separation (PBS) is the proposed architectural fix. PBS formally separates the roles of block proposer (validator) and block builder (searcher/MEV firm). This creates a competitive builder market but introduces new trust assumptions and centralization vectors in the relay network, a problem actively researched by teams at Flashbots and the Ethereum Foundation.

The MEV supply chain is now the dominant force. Validators propose blocks, but specialized actors like Flashbots builders and Jito Labs construct them, competing to extract maximum value from the order of transactions.

Proposer-Builder Separation (PBS) is the formalized market. It creates a two-sided auction where builders bid for block space, separating the power to build from the power to propose, a design now central to Ethereum's roadmap.

Execution clients are the new battleground. The client software (e.g., Geth, Nethermind, Erigon) that validators run determines which builder's block they accept, making client diversity a critical vector for censorship resistance and market fairness.

Evidence: Over 90% of Ethereum blocks are built by a handful of entities like Flashbots, with builders paying validators over 500,000 ETH in MEV-boost payments since the Merge.

The mempool is irrelevant. A transaction's final order is not set in the public mempool but by the block builder's private orderflow. Builders like Flashbots and bloXroute reorder transactions to extract maximum MEV, creating a separate, opaque market for execution priority.

Sequencers centralize ordering. On L2s like Arbitrum and Optimism, a single sequencer node receives, orders, and batches transactions. This creates a trusted bottleneck where the sequencer's software and economic incentives, not a decentralized protocol, dictate final transaction sequence and latency.

Proposer-Builder Separation fails. PBS on Ethereum only separates block building from proposing. The builder's execution client (Geth, Erigon, Reth) still processes the ordered list sequentially, making client software optimization and state access patterns the ultimate performance limit.

Evidence: Arbitrum Nitro's sequencer submits batches every 0.25 seconds, but internal ordering is a black box. This centralization is the trade-off for achieving 40k TPS in theory while relying on a single operator for liveness.

Enshrined Proposer-Builder Separation (PBS) is Ethereum's endgame for MEV. It bakes the builder role directly into the protocol, eliminating the need for trusted relay networks like Flashbots. This creates a credibly neutral auction for block space at the consensus layer.

SUAVE's decentralized mempool is the complementary execution-layer vision. It proposes a shared, cross-chain mempool where users submit intents, and specialized solvers compete to fulfill them. This shifts competition from block building to intent discovery and execution, similar to UniswapX or CowSwap.

The core conflict is centralization. Enshrined PBS risks builder cartels, while SUAVE's solver network requires robust decentralization. The winning design will be the one that best isolates trust and maximizes liveness, preventing a single entity from controlling transaction flow.

Evidence: Flashbots currently dominates >90% of Ethereum blocks, proving the demand for PBS. The success of intent-based protocols like Across and UniswapX demonstrates the market shift away from simple transaction broadcasting.