Ethereum's Quiet Upgrade: When Infrastructure Actually Improves

If You Missed It

The Ethereum network (ETH/USDT) underwent the Fusaka upgrade on 3 December 2024. In contrast to prior upgrades, which were launched with significant hype, the Fusaka upgrade did not have any associated price predictions. Rather, it was focused on providing users with ideal upgrades to the Ethereum network using Layer 2 rollups, where the vast majority of Ethereum network usage currently occurs.

With the Fusaka upgrade, the Ethereum network has increased its data capacity eight times over, reduced transaction fees for Layer 2 users by 40-60%, and decreased the bandwidth requirements for validators by up to 85%. This isn't just theoretical; within 24 hours of the upgrade taking effect, Layer 2 transaction fees fell to the lowest level they've been at in years, in addition to seeing 190,000 new wallets created on the Ethereum network, or approximately 35% above the average number created each month.

With the Fusaka upgrade, Ethereum marked its first upgrade since its successful transition to a Proof of Stake Consensus mechanism, and it has also shifted from large-scale upgrades to smaller incremental improvements. The network now appears to have transitioned from "Can we make this work?" to "What can we do to improve this?" The development of PeerDAS will assist validators with data verification while not requiring them to download all of the data required to support Layer 2 growth, thereby eliminating the bottlenecks that currently limit Layer 2 expansion.

What Fusaka Actually Does

Fusaka is based on an earlier version of peer-to-peer data availability sampling of decentralized data (EIP-7594) known as PeerDAS or PDSA (Peer Data Simplified). Prior to the introduction of Fusaka, validators on Ethereum had to download tin each of the new blob (data bundle) each time one of the many Layer 2 networks (L2s) created by a number of different companies, such as Arbitrum, Optimism, or Base, published a top transaction, which created a huge amount of traffic, causing the layer to congest, forcing every validator to do a lot more work; this caused the fees for using all the L2s to increase significantly during peak times and made it difficult to scale the solution of validating layers 2 with larger amounts of transactions.

With PeerDAS, instead of downloading an entire blob of data, validators sample random chunks of the data with cryptographic proof of the data's existence and availability as a way of verifying it. One can think of the verification process of a batch of data as similar to the way you would do with a manufacturing line, taking a statistically significant (i.e., random) piece from a batch for verification rather than taking all of the items in the batch.

PeerDAS greatly reduces the amount of data a validator has to process (the amount of data being downloaded is reduced by about 80%-85% for full nodes on L2 networks). Thus, validators' costs for running validators are much lower than they would be without PeerDAS, which makes operating a validator much more feasible for individuals or small organizations.

The most advantageous layer 2 networks were the first to witness a large increase in savings on transaction fees during the early days after Fusaka was launched. For example, on Arbitrum, Optimism and Base, all three are Layer 2 networks, the average fee for sending a transaction dropped between 40-60% as a result of Fusaka's launch, from what used to be about $.50 to under $0.20 to $0.30. Furthermore, if you were experiencing high levels of congestion on the main net, the savings would be even greater.

The blob capacity for these networks has also been dramatically increased since Fusaka was introduced. Originally, the blob target was 6 blobs per block, but this has now increased to a target of 10 blobs. Additionally, they want to increase that number to 14 blobs per block. With the introduction of Fusaka and the addition of the PeerDAS protocol, the rollup data throughput (the capacity to create rollups) on the Layer 2 networks is up to 8x greater than it was prior to the introduction of Fusaka.

Why This Matters Beyond the Technical Details

Ethereum's Layer 2 network has found its place and is gaining traction. Arbitrum typically processes about 2.5 million transactions each day while Optimism routinely processes over 1 million transactions each day. Coinbase's Layer 2 solution Base has drawn significant interest due to its lower costs and quicker speeds compared to the Ethereum mainnet. The finality and security of all these networks occurs when they post the bundled transactions to the Ethereum mainnet, while it is the technology companies that are responsible for their operational support systems.

The costs associated with Layer 2 solutions are combined with blob fees (the cost to upload the data), either absorbed by the Layer 2 operators or passed through to their users when and if the amount of blob capacity becomes limited, and the corresponding costs increase. The blob was originally introduced via the Dencun upgrade implemented in March of 2024 and initially caused Layer 2 prices to drop by somewhere between 90-95%. As demand for blob capacity increased, so did the corresponding prices of blob capacity. Fusaka has addressed this scaling challenge by allowing an immense number of blob capacities (hundreds) to be included in each individual block.

As such, reduced operating expenses convert into greater margins, or lower pricing for Layer 2 DeFi protocols including Uniswap, Aave and Synthetix among others. Consequently, when the surplus cost savings go to end-users, DeFi will be able to compete better with centralized finance firms.

Additionally, deploying on Layer 2 is more affordable for developers due to improved cost efficiency of deploying and utilizing Layer 2 projects through Fusaka and other platforms. Thus, projects that were previously unable to afford the high Layer 2 fees or mainnet gas costs can now take advantage of cheaper options for building upon Ethereum and thus expand Ethereum's overall ecosystem in profitable ways.

A Broader Development Strategy

The Ethereum blockchain is seeing another large upgrade with the release of "Fusaka" as part of their ongoing development strategy. In May of 2024, a large portion of The Ethereum blockchain was improved/relaunched by changing how validators can validate. Along with this update was the announcement of a new feature for all Ethereum users called "Account Abstraction". These changes are both examples of a change in the development philosophy on how Ethereum works. Instead of spending years on enormous efforts like the Merge, Shanghai, and Dencun, the team at Ethereum will now focus on small updates throughout the year, thereby avoiding the large level of risk and complexity that accompany launching a new version. If another blockchain rolls out an upgrade before Ethereum rolls out theirs the frustration of users will increase.

The aim of Ethereum now is to provide more frequent updates to its platform, missing some features in those updates, rather than large-scale releases. The Fusaka update was prepared for launch ahead of other updates. Therefore, less risky changes will be made, with the more complicated or higher-risk changes coming later.

According to Ethereum core developer Gabriel Trintinalia, all features that pose a risk to the fork were de-prioritized and dropped from the PeerDAS roadmap (i.e. we're delivering feasible improvements while continuing to research longer-term projects). More specifically, there's been a shift in focus on providing immediate improvements rather than being "the world's computer" with total decentralization. Fusaka does not change the rules, but instead, improves on what is currently available.

What the Fusaka Upgrade Doesn’t Solve

Fusaka isn’t going to change Ethereum’s base layer / Core Architecture. Layer 1 will continue processing approx 15-20 Transactions/Second. Therefore, any High Throughput Applications will still require the use of Layer 2 Solutions. Fusaka helps to reduce Layer 2 Operations Costs and increase Layer 2 Capacity; however, it doesn’t negate the need for using Layer 2s entirely.

Large Stakeholder Candidates won’t see the same amount of potential reduction in operational costs as Smaller Candidate Validators and New Node Operators due to the reduction of bandwidth, storage fees, and Validator margin fees associated with running on Layer 1.

Institutions that have already optimized their infrastructure for Ethereum will not see much impact from Fusaka as the reduction in marginal operating cost would be negligible compared to current operational expenses.

It remains unclear how quickly Layer 2s will take advantage of their expanded storage capabilities. Time to market for Layer 2 growth is determined by developers building Layer 2 solutions as well as user interest. While Layer 2 networks can support significantly greater amounts of blob data, actual usage is uncertain and unlikely to see immediate increases in transaction volume.

Ethereum Foundation developer Marius Van Der Wijden stated that the advantages of increased blob capacity will take time to develop and become apparent, as they build out the blob network slowly to allow for safe and effective scaling of the layer from current throughput capacities, and that this is an indication that the developers will take their time with this rollout.

The Longer Road Ahead

"In this passage, the author discusses the future of Ethereum and the urgency that Vitalik Buterin sees in the continued development of its roadmap.

Danksharding is regarded as Ethereum's ultimate scaling solution, allowing the Ethereum network to support thousands of blobs per block. It is viewed as a long-term step beyond Fusaka. PeerDAS, which will be further developed in future releases, serves to confirm that Ethereum can implement data availability sampling.

Verkle Trees, which will be added in 2026, are more compact than conventional ways of creating cryptographic proofs. As a result, it eliminates the storage requirements of validation nodes to a fraction of what they currently require.

Verkle Trees can help reduce entry-level barriers by reducing synchronization times for Validators from days to just hours, according to Vitalik Buterin.

In addition to advancements in technology, researchers are also investigating how the Ethereum base layer can more effectively handle ever-increasing data throughput without sacrificing decentralization, as well as what security measures Layer 2 will require for greater sovereignty and how to improve interoperability between Rollups."

The Mature Network Thesis

The Ethereum ecosystem has reached a level of maturity thanks in part to the many enhancements implemented during its evolution, according to Nick Johnson, the Ethereum name service's lead developer. No longer are upgrades about promises or the ability to create additional revenue streams; now the focus is on producing the best possible service for all users of Ethereum through increased efficiency, decentralization and resiliency.

Fusaka provides an excellent opportunity for developers of all types to streamline their processes to create more robust applications. Fusaka allows developers to create dApps (decentralized applications) that are used by thousands of end-users daily, conduct billions of transactions every day, and protect hundreds of billions of dollars in assets. Fusaka improves the efficiency of the entire Ethereum ecosystem.

One of the best indicators that Fusaka is a successful component of Ethereum is the fact that the upgrade occurred without any disruption to the network. This is the same way that mature infrastructure continues to function. There may be a change made, there are some benefits to using the new version, and the Users can continue to operate on the existing version as they have always done.

With the introduction of Fusaka, the direction of Ethereum is clear: more security and more throughput, increased capacity and efficiency in Layer 2, reduced cost of running a node, and continued delivery of incremental improvements every six months. Fusaka proves that the strategy of continuous improvement for Ethereum is a sound business strategy, and as the next upgrades come in the future, we will see more evidence that this strategy is working.