DeFi Opportunities on High-Performance Chains

Author: Derek Walkush

Published on: 03.08.2024

Due to the growth of L2, the Ethereum ecosystem remains the battleground for most DeFi choices. However, the recent successes of Solana, Aptos, and other high-performance chains indicate that they will play a significant role in the development of DeFi.

However, the differences between high-performance chains and Ethereum may be something that experienced DeFi builders in the Ethereum ecosystem cannot fully comprehend.

Below, I will discuss what I believe are the four most important technical differences relevant to DeFi projects. Builders who understand these characteristics of high-performance chains will be able to seize opportunities that may not be achievable in the familiar Ethereum playground.

#1: Throughput

A defining feature of high-performance chains is their extremely high TPS. This technological improvement unlocks a range of DeFi applications that cannot be realized on the Ethereum mainnet. Achieving fast confirmations on current Ethereum L2s is extremely difficult, as it requires not only proof generation (zk) and fraud proof dispute (op) time but also a 12-second final confirmation on the mainnet. Therefore, for the time being, high-performance chains are the only viable solution for fully on-chain order books. With dynamic parallel approaches like Block-STM, newer CLOBs (such as Econia) are capable of minimizing conflicts and maximizing throughput.

To be "fully on-chain," an order book must not only support trade settlement (asset transfers between wallets) but also support a matching engine (where traders place bids). The latter is more challenging to host on-chain because the order book must absorb a large amount of junk information. While order books require better liquidity and throughput compared to AMMs, they do not require passive LPs, thus providing a more capital-efficient market structure for trading.

New projects also have another opportunity: while spot DEXs price assets on order books, DEXs with synthetic derivatives often require external infrastructure to provide prices, namely oracles. Although high-performance oracles are emerging, many traditional oracles cannot meet the low-latency demands of high-performance chains (with confirmation times under 1 second).

#2: Gas Fees

Lower gas fees, combined with high throughput and low latency, offer numerous advantages for DeFi application builders on high-performance chains, especially DEX aggregators.

Aggregators source optimal prices from various on-chain (and even off-chain) liquidity sources. However, due to high gas fees and greater latency on the Ethereum mainnet, prices are often retrievable only from a single DEX or off-chain source.

In contrast, on high-performance chains, the advantages of low fees and low latency enable aggregators to find the best prices for assets across different liquidity venues. In other words, an aggregator can swap tokens on one DEX, but if another path offers a better price, it can choose to swap again on another DEX or through a more complex route. Thus, pricing for aggregators on high-performance chains is more competitive than on the Ethereum mainnet.

The story of Solana illustrates this well. Jupiter is Solana's leading aggregator, controlling about 80% of trading flow (after filtering out bots). This starkly contrasts with the Ethereum mainnet, where a significant amount of trading flow still goes through DEX frontends, with multiple aggregators controlling ~40-50% of order flow, and no single frontend controlling more than 30-40%. This means that liquidity and order flow are naturally decoupled on high-performance chains, and aggregators will always seek the best price for a swap across multiple venues, introducing a natural incentive not to build their own liquidity (as it may not provide the best price). Liquidity sources (e.g., AMMs, order books) can attempt to establish their own proprietary order flow, but they may struggle to compete on price with leading aggregators.

Additionally, advancements in asset bridge infrastructure and intent-based protocols make the emergence of higher-level solutions possible: cross-chain aggregators on high-performance chains. As activity increasingly spreads across different L1s and architectures that are entirely different from the Ethereum ecosystem, a single cross-chain aggregator could be very promising.

Thus, aggregators on high-performance chains represent a huge opportunity as well.

#3: Validator Selection

To support extremely high TPS, many high-performance chains compromise on decentralization in their consensus protocols. They achieve higher scalability through mechanisms like DPoS and validator clusters, effectively increasing transaction propagation speed by not broadcasting to a large network of nodes.

As part of their consensus protocols, many high-performance chains use deterministic leader schedules, which are a set of predefined validators responsible for ordering transactions in blocks. Therefore, transaction ordering is highly dependent on the validators chosen as leaders, as they occupy exclusive and established positions. This is in stark contrast to Ethereum, where validators are pseudo-randomly selected.

The selection methods for leaders or representatives vary, but the result of a deterministic leader schedule is that certain validators can build more blocks and earn greater rewards, creating a self-reinforcing cycle where these validators typically accumulate more stakes. Moreover, given the higher demands for state access (to support high TPS, parallelization, etc.), high-performance chains often have higher hardware requirements, meaning that the entry barrier for validators is higher (in terms of upfront costs).

This dynamic creates a range of advantages for widespread staking, especially for liquid staking tokens (while validator selection on Ethereum is very important, it typically does not have a significant impact on overall staking yields). Essentially, some liquidity tokens (LSTs) that can access the best validators can gain structural advantages through higher yields. There is also an opportunity for high-performance chains to attempt something similar to Ethereum's shared security/re-staking layers, allowing projects that can access the best validators to further enhance yields through re-staking.

#4: Block Building

The block building process varies by chain, but broadly speaking, there are significant structural differences between high-performance chains and the Ethereum mainnet, which greatly impact MEV.

Using a reserved leader schedule or delegation to validate transactions can pose challenges for packing transactions. A first-in-first-out (FIFO) model sometimes requires searchers to send junk transactions to the chain in an attempt to include them first, rather than orderly packing sorted transactions. Different gas fee market designs (especially Solana's native fee market) also mean that priority fees do not always increase the chances of transactions being included. These architectures make executing MEV strategies (like sandwich attacks) challenging.

Furthermore, some chains (especially Solana) lack public mempools, making it harder for external MEV searchers to retrieve and order transactions to execute MEV strategies. Pipeline flow and parallel blockchains like Aptos can disrupt sandwich attacks by randomly reordering transactions before the parallel execution processing stage.

Thus, MEV opportunities on high-performance chains are structurally different. Subtle differences in blockchain architecture can significantly impact MEV. For example, Jito block building auctions on Solana still present a massive opportunity for builders. These architectures create opportunities for new types of MEV infrastructure, but the topic is still under research and is not yet as well understood as MEV on the Ethereum mainnet.

In Conclusion

Most high-performance chain ecosystems are still in their infancy and often crudely follow the developmental path of Ethereum DeFi. However, the slight divergences mentioned above (and many others I have not discussed) could greatly influence the trajectory of certain industries and create significant opportunities for DeFi builders.

Thanks to Mert (Helius), Lucas (Jito), and Avery (Aptos Labs) for reviewing this article.