Initially, blockchains were a simple network of nodes running software to process, validate, and store transactions. Now, there are modifications of nodes that run a variety of functions for L1s, L2s, MEV, Provers, Eigen AVSs and more. Ethereum has just under 6,000 nodes, second only to Bitcoin with over 18,000 nodes. Blockchains are considered valuable because of the economic capital used to securely facilitate the distribution of compute over a peer-to-peer network. Over time, the computation that nodes are responsible for has expanded.

EigenLayer AVSs have demonstrated that a blockchain node can support far more than simply verifying and processing blockchain transactions. Instead, it is becoming the fundamental infrastructure to power services ranging from decentralized AI inference to Zk Co-Processors. 

Reth rebundles the various services and functions mentioned above into a single framework where they can all be modified, run, and maintained via one node. Two primary impacts in the near term: First, with Paradigm’s recent backing of EigenLayer’s rival Symbiotic, Reth acts as a developer funnel enabling all Ethereum nodes to run decentralized services for Symbiotic. Secondly, it provides a performance module to increase the TPS for L2s, allowing chains like Base and Optimism to become more competitive with the likes of Solana by supporting cheaper transactions. 

Background on Ethereum Nodes

Nodes are foundational infrastructure for a blockchain, and their primary responsibility is to always run the software for a network. Node software is analogous to Linux in the Web2 space, where it powers most of the servers behind cloud service providers like Google, Microsoft, and Amazon. Similarly, Ethereum nodes are used to power dApps, staking, exchanges, and other major Web3 products and services. 

Ethereum nodes run two primary forms of software: the execution client responsible for processing and storing transactions and the consensus client which enables validators to come to an agreement on the transactions processed by the execution client. 

Ethereum has a culture of client diversity. There are six different variants of the execution client and five for the consensus client developed by different teams. The idea is that if there are multiple implementations of the software, in the event that there is a bug in one of them, Ethereum as a network can continue to operate since there are other implementations. Therefore, running multiple clients is important for network security and resilience.

A similar phenomenon is found in web2 cloud services where companies like Amazon take advantage of the open source and modular nature of Linux to create their own version tailored to the needs of their users. Having multiple large companies maintaining different Linux implementations improves the software’s robustness and range of use similar to Ethereum’s client diversity. 

As discussed previously, the role of a node has expanded over the years, and many entities will modify Ethereum nodes and clients to specialize in certain functions. For example, Ethereum stakeholders are not the only ones who run nodes using Ethereum software. BNB chain forked its own version of Geth (Ethereum client software) and modified it to increase the amount of transactions that can be processed. Most Ethereum rollups such as Optimism have also forked Geth and added certain modifications in the form of OP-Geth. But this is not limited to L1s and L2s, Flashbots created a fork of Geth called MEV GETH for MEV bots and frontrunning protection. However, forking is a delicate process since one must actively maintain it when new upgrades are made. There is the risk of making unsound changes that break certain features, which can result in a security or performance issue. This added risk and developer overhead will persist as more people run new use cases for Ethereum nodes such as High-Throughput L2s and AVSs. 

Reth Overview

At a high level, Reth allows Ethereum, Alt-L1s, L2s, and offchain services to be run, maintained, and modified within a single-node framework. However, there are many different types of nodes and Reth has different goals for different parts of the stack:

Ethereum L1 Nodes

For Ethereum mainnet, Reth’s sole goal is to improve the stability of the network by diversifying nodes that are running different execution clients. Currently, 40-50% of nodes are running Geth, followed by Nethermind. Reth’s team aims to have 20%-30% of Ethereum L1 nodes running Reth. 

L2 and Alt L1 Nodes

Beyond mainnet, Reth aims to create more performant nodes that are easy to maintain and modify by Alt-L1 and L2 teams. Many Alt-L1s and L2s modify the EVM, from raising the Gas Limit in order to support cheaper transactions to including quality-of-life upgrades for developers. As discussed in the Ethereum Nodes section, modifying the EVM requires developers to fork and maintain an Ethereum node which introduces some overhead to teams. 

Reth instead introduces the ability to run L2 node software as a sidecar process for the L1 node, making it easier for individuals to run an L2 node and allowing teams to modify the software easily. Rather than creating a copy of Ethereum node software and creating custom features, Reth has the ability to support the creation of modules that teams can import. An example of this is a performance module that enables higher TPS for EVM chains which will be discussed in a later section. 

Execution Extensions

Reth’s node framework extends to offchain services via “Execution Extensions” that repurposes existing blockchain nodes for decentralized AI, Zk Co-processors, DePin, etc. All of this is achieved in a similar form factor to Reth’s other products by allowing these additional services to be run as a “sidecar” process to existing L1 and L2 nodes. 

This is the high-level anatomy for Reth’s Execution Extensions:

• High-fidelity data stream: Building decentralized onchain or offchain applications requires real-time crypto data to ensure that services are not lagging behind and provide a sub-optimal experience to the end consumer. The main pain point preventing real-time crypto data streams is block reorganizations, when validators disagree on the most accurate version of the blockchain and have to remove or change a block. Ethereum often has over fifteen block reorgs daily, so Reth created a high-fidelity data stream that is both reorg-aware and easily accessible to developers.
• Orchestrating concurrent services: Reth enables additional decentralized services to run on top of Ethereum L1 nodes but this introduces the complexity of ordering and orchestrating the execution of these tasks which results in both higher engineering overhead and risk of downtime. Reth created a “manager” modeled after web2 Extract, Transform and Load (ETL) frameworks to relieve the risks and overhead on protocol and dApp teams. ETL frameworks automate the orchestration and execution of complex data tasks that companies like Spotify, Lyft, and Airbnb use. 
• One API to rule them all: Managing multiple services, such as an Ethereum node and an AI Co-Processor, could be tedious. Reth provides a single API that can be used to download the Execution Extension like an app on the app store and start running it alongside additional services.

Implications For L2s And Restaking

As discussed above, Reth is able to consolidate different pieces of L2 infrastructure such as Sequencers and Nodes and allows for them to be run and maintained in the same environment. With projects such as Base, Optimism, MegaEth, Worldcoin, Taiko and Immutable testing out Reth, there is a possibility where a significant portion of Ethereum's L2 ecosystem along with its surrounding infrastructure runs on Reth. Additionally projects like Base and Optimism that are in the top three amongst L2s when it comes to onchain volume have expressed interest in using Reth’s performance module to increase their TPS, making them more competitive with high-throughout ecosystems such as Solana. 

As more projects begin to utilize Reth, this also serves as a funnel towards restaking protocols like Paradigm backed Symbiotic. Symbiotic has mentioned their alignment with Reth as a framework to develop AVSs which can prove to be a catalyst for the number of projects building on Symbiotic compared to EigenLayer. 

Creating High-Throughput L2s

Reth provides performance upgrades to EVM L2s or Alt-L1s to increase the amount of computation per second. The Reth team prefers to use the term Gas per second (GPS) which encompasses all compute done on a blockchain per second including processing transactions. 

To put performance in perspective, Ethereum L1 does 1.25M GPS, translating to roughly 13 TPS. Reth seeks to provide a performance module that Alt-L1s and L2s can use to give a parallelized virtual machine that can reach 100M GPS, translating to roughly 1,000 TPS. Their goal is to get to 1 giga GPS or approximately 10,000 TPS in the medium term, which projects like the Base team have expressed public interest in. 

Reth is looking to achieve this ambitious goal by making key changes to the EVM: This includes parallelizing the EVM, introducing a more efficient compile method, and processing state roots differently. 

• Parallelized EVM: Instead of processing one transaction at a time, doing multiple concurrently increases the overall throughput. 
• Efficient Compiling of the EVM: The EVM provides instructions on how to process transactions and then translates this into a format that the nodes can understand: web assembly. This process is time-consuming and the Reth team has found a way to compress it.  
• Efficient State Root Processing: Ethereum and most blockchains store their state as a Merkle tree. When creating new blocks, rather than copying the updated tree that has the entire blockchain’s history it calculates something called a state root which is just a compressed proof of the entire updated tree. The process of doing this makes up over 75% of the total time it takes to process and validate a block of transactions and the Reth team has found a way to make this process more efficient. 

As a test, the Paradigm team released an alpha deployment of a high throughput Reth rollup via Conduit using OPstack. This makes sense because of the ease of deployment along with the fact that both Conduit and Optimism are Paradigm portfolio companies. With that in mind there is an overlap between Paradigm backed/aligned L2s and L2s that already have a significant amount of onchain activity. The networks at this intersection have the highest likelihood of leveraging Reth as a catalyst for their ecosystem.  

Base, Optimism and Blast fall into this sweet spot. The Reth team has expressed that it will currently not support Blast due to certain modifications made by their team. Already in the top three in terms of onchain volume amongst Ethereum L2s, Base and Optimism saw a further surge in transactions post EIP-4844. This shows that these ecosystems could see a surge in activity with an implementation from Reth reducing transaction fees and increasing throughput.

When onchain activity surges on L2s, so does the revenue generated by sequencers. EIP-4844 introduced a 4x increase in sequencer revenue for Base of which 2%-15% of it is paid to the OP treasury, and in the case of OP mainnet, 100% of its sequencer revenue is paid to the treasury. During that time period, Base was generating over $272M in annualized revenue which would yield over $40M to the OP treasury. This is alongside OP mainnet’s almost $17M in annualized revenue for the OP treasury during the same time period. Over the past 12 months, the OP Superchain ecosystem has generated over $25M in Eth for the OP treasury. With cheap DA costs and an upcoming significant reduction in execution costs, it will be interesting to see how much additional sequencer revenue is generated from the next catalyst for L2 onchain activity.

Onboarding Funnel For Symbiotic

Frameworks such as OPstack and CosmosSDK have proved to increase the number of projects building on a particular substrate. For example, OPstack has led to over forty three optimistic rollups, and CosmosSDK has launched over eighty Cosmos app chains. Reth is positioning itself to be a similar catalyst for Symbiotic by providing a simple framework to create decentralized services. There are already projects such as Hyperlane that are an EigenLayer AVS but are also opting to use Symbiotic potentially due to their use of Reth.

Sreeram, the founder of EigenLayer, acknowledged that developing AVSs is no trivial task. His team is currently focused on making it possible to develop AVSs, and once that is achieved, they will work on making the process easier. Symbiotic will be launching with the support of Reth right out of the gate making it easier for individuals to develop AVSs for them and also opening the possibility for existing Reth users to be onboarded to their restaking protocol. Some examples of projects dabbling with Reth are Astria, Avail, Axiom, Hyperlane, MegaEth, Taiko, Berachain, Base, Worldcoin, Immutable, Aave and many others. 

As discussed in an episode of the Bell Curve Podcast EigenLayer, AVSs have soft caps on how much capital can be deployed to them. This cap is determined based on how much security the AVS actually requires along with what it is willing to pay for. This means that large capital aggregators such as LRT providers have a significant amount of capital that is idle and waiting to be deployed. This dynamic is due to the rate of capital flowing into restaking exceeding the rate of AVSs being developed and deployed. If Reth is able to accelerate the development and deployment of AVSs on Symbiotic this can serve as a catalyst to drive idle capital from LRTs into Symbiotic as opposed to waiting for the next EigenLayer AVS to deploy.  

Assessing Risk

Reth has reached just under 3% adoption amongst Ethereum mainnet nodes with Geth as the clear leader with over 50%. As mentioned in earlier sections, Reth is not only used for L1s but this is an indicator that Reth is still in an early stage and will require further testing to establish its place in the node client landscape.

Additionally, Reth’s parallelized transaction processing comes with difficulties due to the inherent inefficiencies that come with processing blockchain transactions concurrently in the EVM. There are currently two ways that virtual machines process transactions at the same time. The first is in a pessimistic approach seen in Solana. In this system, transactions have labels, allowing them to be sorted to avoid conflicts. This avoids the risk of processing multiple transactions at the same time that depends on the outcome of one another. The second approach is the optimistic approach found in Aptos and currently the only way to address parallelization in the EVM. In optimistic environments, transactions are executed in parallel without an upfront label. Instead of avoiding conflicting transactions upfront, the system assumes all are compatible and will re-execute in the event of a conflict. 

Concurrent transactions that depend on the outcome of one another are typically re-executed in chronological order. However, too many conflicting transactions could result in a backlog that needs to be re-executed, resulting in potential complexities and performance downgrades. To avoid having too many conflicting transactions, Aptos leans on the Move programming language that is object oriented and allows for transactions and contracts to be labeled up front similar to Solana’s pessimistic approach. This is something that is lacking on the EVM, and Reth as a byproduct, although may be implemented in the next major Ethereum upgrade Pectra, slated for Q1 2025. Further, Optimism is positioned for an increase in activity amongst the Superchain L2s, which could drive value to the token treasury and therefore OP. However, with less than 30% of the token supply circulating, it remains unclear how the unlocked supply will affect future liquidity and growth.

Conclusion

Ethereum and its adjacent infrastructure is becoming increasingly complicated, and Reth seeks to consolidate the infrastructure into a single framework to decrease developer overhead and make it easier to add modifications. There are two potential tailwinds from Reth that are worth paying attention to:

• Reth accelerates the rate of AVS deployments on Symbiotic driving idle capital from EigenLayer to Symbiotic, making it attractive for teams wanting to develop AVSs and for LRTs and Ethereum stakers to deploy their capital.  
• Reth’s performance module that unlocks higher throughput EVMs will benefit high-growth L2s found in the OP Superchain ecosystem such as Base and Optimism. The Paradigm team is already running an alpha testnet on an OPstack rollup along with projects like Optimism and Base expressing open interest in Reth. This catalyst could lead to higher revenue for sequencers in the Optimism Superchain increasing the OP token treasury and driving value to OP. Blockworks has covered AERO being tightly coupled with the growth of Base by serving as a gas token proxy which could be another beneficiary of Base’s implementation of Reth. 
  
  
  The information contained in this report and by Blockworks Inc. and related affiliates is for general informational purposes only and is not intended to provide legal, financial, or investment advice. The report should not be construed as an offer or solicitation to buy or sell any security, token, or financial instrument and does not represent any recommendation or endorsement of any investment or financial product or service. Blockworks Inc. and related affiliates are not registered as a securities broker-dealer or an investment advisor in any jurisdiction or country.