By
Dimitar Bogdanov
February 19, 2024
4 Min Read
The long-anticipated Dencun upgrade for Ethereum is set for a March 13 release, promising to bring big changes to the world’s second-largest blockchain protocol. Of course, the highlight of the upgrade is expected to be the implementation of EIP-4844, which is set to introduce blobs of data designed to make rollups even more efficient and cost effective. We’ve already talked about EIP-4844 and its significance on these pages. But the upgrade will include several other improvements that are not that well known.
In this article we’ll be exploring that other side of the Dencun upgrade, formally known as Cancun-Deneb. In fact, the name seems like a perfect starting point for this piece.
So before we delve into the technical details, let’s have a brief high level overview of the Dencun upgrade. First, let’s unpack the name of the upgrade.
As mentioned above, the official name of the upgrade is Cancun-Deneb, so Dencun is simply a somewhat clunky portmanteau of the two parts of that name. So, why Cancun-Deneb, then?
Well, because these are actually two separate upgrades that will launch simultaneously. Cancun is an upgrade for Ethereum’s execution layer, while Deneb is its counterpart targeting the consensus layer. Because they’re launching at the same time we can think of them as two parts of a larger upgrade. So Dencun it is!
The standout feature of the upgrade is undoubtedly proto-danksharding, which will be introduced with the implementation of EIP-4844. You can check out our detailed piece on the matter, but here’s a quick refresher. Proto-danksharding will introduce data ‘blobs’ that will store rollup transaction data for a set period of time. This is a form of temporary storage for rollup transaction data that is expected to make rollups even more efficient at scaling Ethereum.
Beyond EIP-4844, Dencun is set to bring several other improvements, including some aimed at tidying up things at the opcode side, as well as improvements to staking and validating and communication between the two layers. In the following paragraph we’ll be taking a closer look at some of those less talked about features of Dencun.
Apart from EIP-4844, the biggest change Dencun is set to introduce is probably EIP-4788, which is expected to significantly improve communication between the consensus layer and the execution layer of Ethereum.
The execution layer and the consensus layer are essentially two separate chains linked together via 1:1 mapping. This means that for every consensus layer block (also known as beacon block) there is only one corresponding execution layer block and vice versa. Despite this, facilitating communication between the two layers has historically been far from straightforward, requiring either a complicated voting process or intermediaries such as trusted third-party oracle solutions.
This is about to change with the implementation of EIP-4788. The planned changes include inserting the root of the parent beacon block into the current execution layer block. This root is a succinct cryptographic representation of the header of the previous (rather than the current) consensus layer block. Thus, by including it in an execution layer block, we can have information about the state of the consensus layer directly on the execution layer. This essentially creates an oracle that provides information about the consensus state in a largely trustless manner.
In addition, a small history of these roots will be stored in a smart contract, allowing Ethereum dApps to make queries about the state of the consensus layer.
Next, we have two Cancun-focused improvements that are designed to improve opcodes.
EIP-6780 is aimed at addressing potential issues with one particular opcode - SELFDESTRUCT. The opcode is viewed as problematic by the community, but it’s used in a lot of existing smart contracts, so it cannot simply be removed. EIP-6780 is designed to ‘disempower’ SEFDESTRUCT, effectively making it obsolete for future uses, without removing it completely.
Meanwhile, EIP-5656 adds a new opcode called MCOPY that is designed to make memory copying in the EVM more efficient. Currently, the preferred method to do this involves referencing two opcodes - MSTORE and MLOAD. MCOPY eliminates the need to use multiple instructions, which is not only more convenient, but more efficient, as well.
This pair of EIPs is aimed at improving the staking and validator experience on Ethereum. In particular, EIP-7044 is set to make it better for ETH stakers to use delegated staking solutions.
Following Ethereum’s transition to Proof of Stake, users can become validators by staking a minimum amount of 32 ETH. However, people who do not want to perform the validator role can still benefit from staking via specialized solutions for delegated staking. Some solutions that allow stakers to use them non-custodially, meaning that they do not need to cede control of their funds to the delegated staking provider.
However, even with these non-custodial solutions, there are other instances where the staker is reliant on the honest behavior of the provider. Namely, when a staker wants to withdraw from a particular validator. This process requires a voluntary exit message that has to be signed by the validator. This means that the staker needs to be able to trust the delegated staking provider to be honest.
To provide stakers with sufficient assurances delegated staking providers typically offer pre-signed voluntary exit messages that their stakers receive when committing to a particular validator. However, presently, these pre-signed VEMs are only valid for two Ethereum upgrades.
EIP-7044 will change that by making VEMs valid in perpetuity. This means a much greater level of security for stakers.
Meanwhile, EIP-7045 is all about attestations. Under Ethereum’s consensus, blocks need to receive attestations from validators in order to be considered valid and added to the blockchain. These attestations need to be submitted within a specified timeframe, which is currently one epoch, or 32 12-second slots. EIP-7045 will increase that timeframe so that validators will have the rest of the current epoch + the next one - in other words, up to 62 slots - to submit attestations for a given slot.
Naturally, this change will allow more validators to compete for block rewards. Even more crucially, however, increasing the attestation submission window will allow more attestations to be accumulated, which will help speed up the block confirmation process.
These are some of the new features coming to Ethereum as a result of the Dencun upgrade. We also have EIP-1153, which is set to introduce opcodes for ‘transient storage’ that is expected to further reduce costs and improve efficiency. And then, there is the highlight of the upgrade, proto-danksharding.
As a whole, the Dencun upgrade will mark another important step in the evolution of Ethereum and is expected to set the stage for even more improvements in the future.
