Pectra hard fork explained — Will it get Ethereum back on track?

The Ethereum roadmap has been under attack from all sides recently, with even ETH researcher Dankrad Feist and Base creator Jesse Pollak agreeing the existing scaling plan is “unambitious” and not “even close to aggressive enough.”

The community consternation appears to have lit a fire under some leading figures.

Over the weekend, Ethereum creator Vitalik Buterin released a new post calling for the roadmap to accelerate and quickly scale up from the current three data availability blobs per slot to 128 blobs and beyond. L2s use blobs to store data and need as many as possible to grow affordably. Crypto investment firm Paradigm laid out its own plan that basically says: “Stop dicking around and ship scaling upgrades faster.”

Separately, the leaders of L2s including Base, Arbitrum and Optimism have agreed on the need to unify the ecosystem by implementing based and/or native rollups for their projects.

Ethereum’s next upgrade is called Pectra and has just been scheduled for March. So how far along the scaling road will it take us, and what other new functionality does it bring for everyday users?

Pectra upgrade split in two

Initially planned to be the largest upgrade in Ethereum history with up to 20 Ethereum Improvement Proposals (EIPs), the fork has since been split into two. As a result, Pectra is bringing just a temporary reprieve from scaling issues by doubling the number of blobs used for data availability from three to six. 

Mallesh Pai, the senior director of research at Consensys, points out that the idea of doubling blobs “was not on the roadmap like six months ago,” but the upgrade is still not going to be able to meet the astonishing demand from L2s.

“The current estimates based on L2 rate of growth are that this upgrade will only be enough until the end of the summer,” says Pai.

The good news is that the Fusaka fork, slated for later in the year or early next year, is set to include an upgrade called Peer Data Availability Sampling (PeerDAS), which will be the most important change for scaling; Buterin wants a laser-like focus on it.

PeerDAS: Like torrents for blobs

PeerDAS is the next evolution of “proto-danksharding” which enabled L2s to store data in blobs and not compete with L1 transactions for block space. But EIP-4844 still requires full nodes to download all blobs from a block and verify that no data is missing from a particular blob.

“This is wildly inefficient and affects smaller node operators because downloading blobs (and even broadcasting them, in the case of a proposer) takes a lot of bandwidth,” says Emmanual Awosika, Head of Research at 2077 Collective.

PeerDAS instead divides up the task of storing and verifying blob data into manageable chunks — a bit like how peer-to-peer Torrenting software can divide up and access a bit of each movie file on a few thousand different computers.

Buterin says it will increase the blob count immediately by 2x to 4x, growing to 8x or more over time. 

Just a week or so ago, Pollack was saying the existing roadmap (below) was too slow, and the 2027 targets needed to happen by 2026. If Buterin’s estimates are correct, PeerDAS alone may enable 48 blobs per slot sometime next year.

However, Pectra brings notable improvements beyond doubling the blob count, says Awosika. They include enabling account abstraction for all accounts and changes to the maximum staking limit, both of which he says are a “big deal.” Another mooted change called EIP-7623 is still line ball as to whether it will be included or not.

“If they actually implement EIP-7623, then increasing the gas limit further in the future is possible — so it’s a big deal. The rest are just improvements and iteration.”

One reason there’s so much bearishness about Ethereum’s roadmap is that so few people actually understand the jargon and technical concepts the devs and researchers use to explain it. We’ve tried our best to translate the elements of the Pectra upgrade into English for regular people.

Everyone gets a smart account (EIP 7702)

The biggest usability improvement enabled by Pectra is EIP-7702, which allows existing Ethereum accounts to upgrade to smart accounts (also known as “account abstraction”) with a single click.

This will finally fix the long-detested requirement to sign two transactions for “approve” and “swap” to make an Ethereum transaction.

It also enables “social recovery, native multisigs, and all sorts of other functionality that allows us to bring a huge amount of new value and experiences directly to users,” says Pai. 

If this all sounds familiar, it’s because smart accounts already launched on Ethereum in a blaze of glory with ERC-4337 back in early 2023. However, that was a bit of a workaround to avoid a hard fork and required users to go through the painful process of starting up a new wallet. At its peak, around 3.5 million wallets used it.

“Widespread adoption has been held back by a key limitation: users couldn’t upgrade their existing wallets to smart accounts,” explains John Rising from smart account platform Stackup.

Also read: ‘Account abstraction’ supercharges Ethereum wallets: Dummies guide

Incredibly, Pai says that Buterin knocked up the much more elegant solution in Pectra in under an hour.

“I know for a fact he was, like, on a safari in Kenya, at an off-site on a truck […] and came up with a path that everyone could get behind and do it. So yeah, he is amazing.”

Rising says that by allowing existing wallets to simply upgrade, “we’re removing the last major barrier to mainstream adoption of account abstraction.”

“Features we take for granted in traditional finance — automatic payments, account recovery, merchants covering transaction fees — can finally become standard across Web3,” he says.

“We’ll see subscription services without pre-approval, gas fees paid in any token, biometric security replacing seed phrases, autonomous agents operating safely within wallets and seamless crosschain interactions.”

While that all sounds great, complex smart account features are relatively expensive on the L1. Once EIP-7702 is live on L2s, however, smart account capabilities can work seamlessly across both.

Validator staking limits increase (EIP-7251)

One big improvement for the health of the network itself is increasing the maximum validator stake limit from 32 ETH to 2048 ETH.

At present, if you were lucky enough to own 2048 ETH and wanted to stake it all, you’d need to run 64 separate validators. That puts 64 times more strain on the network than needed for not much gain in terms of decentralization, given that the same person controls ETH.

“The resulting networking overhead (each of these validators has to sign blocks, the signatures have to be aggregated, etc) lowers the performance of the network as a whole,” explains Pai.

Awisoka says that if millions of validators tried to join the Beacon Chain it would break, so upping the limit is a wise move. “It removes one of the root causes for why people like Lido and Coinbase run so many validators if they have few node operators.”

Lowering network overheads has other benefits too. “This upgrade unlocks other much-needed upgrades like single-slot finality,” says Awosika. Single-slot finality allows blocks to be proposed and finalized within 12 seconds rather than 15 minutes and means transactions are confirmed almost instantly. “SSF is a big blocker for rollup interoperability,” he says.

Blob target will double 

As previously noted, the blob target is doubling from 3 per block to 6 to help the L2s scale up. The maximum number of blobs is already six, but whenever L2s post more data than the target, fees shoot through the roof and the L2s become uneconomical.

“This has already happened several times this month. Pectra increases the number of blobs per block to a maximum of 9, with a target average of 6. This is therefore effectively a 100% increase,” Pai explains.

That said, increasing the number of blobs even faster comes at the cost of increasing block sizes, and imposing higher bandwidth demands on Ethereum solo stakers. The higher the requirements, the fewer solo stakers.

“Trading off the needs of Ethereum’s L2s with its core values of decentralization will be a key theme for developers over the coming year,” he says. 

Separately, validators have been calling for an increase in the gas target to 36 million per block, which will help scale the base layer of Ethereum. It doesn’t require a hard fork and is likely to be enabled as soon as the vote hits 51%.

Blobs may increase further (EIP-7623)

EIP-7623 is being considered but has not yet been confirmed for the Pectra upgrade. The proposal could increase the number of blobs per block to as many as 18 by repricing call data — which was the original method of storing data on Ethereum.

“If they actually implement EIP-7623, then increasing the gas limit further in the future is possible,” Awosika says.

The proposal essentially increases gas costs for transactions using lots of call data to incentivize them to use blobs instead. This makes call data and block sizes more predictable, meaning the network can safely increase blob numbers without overloading the system.

“If you can make it so that you know what max blocksize will be (meaning: rollups trying to use call data instead of blobs won’t blow up the block), then I think you can be more confident about adding more blobs,” Awosika explains.

Exiting delegated Ethereum staking without permission (EIP-7002)  

This is another complicated upgrade that makes delegated staking your ETH more secure. It both enhances Lido and RocketPool’s control over rogue node operators, as well as the ability of staked ETH tokenholder’s to withdraw funds without the permission of the node operators that run validators on their behalf.

Currently, you can be in a situation where you can’t withdraw the funds you’ve staked because you don’t have the validator key needed for withdrawals.

“This is bad for obvious reasons, namely that someone can just tie up your funds and refuse to agree to the withdrawal,” says Awosika.

The new upgrade separates the validator key used by node operators from the withdrawal key, which stays with the ETH staker. This allows the staker to recover their balance and withdraw any time they choose. Stakers submit a request as a regular transaction, and after a bit of technical malarky, “the staker withdraws staked funds once they are sent to the withdrawal address without ever needing the operator’s approval.”

The upgrade also benefits pooled staking protocols like Lido and Rocketpool. “Currently, node operators funded by the DAO-controlled withdrawal addresses can refuse requests to exit a validator or may deliberately commit offenses to trigger slashings of staked ETH,” says Awosika.

“The protocol DAO is left helpless in the status quo as those validators can only be exited by the validator key controlled by node operators. EIP-7002, however, empowers the DAO to force the exit of one or more validators controlled by a malicious node operator.”

It won’t happen overnight, but it will happen

Pai says that it’s perfectly understandable the community is calling for scaling improvements to happen faster — and he is too. “Everyone wants stuff done today,” he says, adding, “This stuff will get done.”

Consensys boss and Ethereum co-founder Joe Lubin says the roadmap is probably going a lot better than many people think. “I don’t think we’re absolutely optimizing, but I’m really happy with how things are going,” he says.

“There are details that some people will hold that some elements of it are not going that well, but I think we’re on track.”

Lubin points out that upgrading a $400 billion blockchain like Ethereum requires balancing speed with caution.

“You’ve got to make sure that we don’t break the network because there’s a lot of value on it and we want to make sure that it is sufficiently decentralized for participants that have a certain level of equipment. And so that’s a debate. I’m in favor of upping the base requirement for bandwidth and hardware. I think that’s very reasonable.”

And as far as Pectra goes, Pai says it’s a significant upgrade that will help solve issues that have been plaguing the Ethereum ecosystem.

“The interoperability and fragmentation issues that have hurt user and developer experience will be much easier to solve after the upgrades in Pectra,” he said.

This article first appeared at Cointelegraph.com News