Polygon zkEVM is an EVM-compatible, zero-knowledge (ZK) rollup for Ethereum. EVM compatibility means developers can easily port Ethereum smart contracts with minimal code changes. ZK rollups are scaling solutions that increase throughput by moving computation and state storage off-chain while using zero-knowledge cryptography to maintain high security guarantees.
The goal of this article, however, is to highlight where Polygon zkEVM fits in a broader market context - so for a holistic technical dive, check out the project documentation.
Let’s start by framing the blockchain stack. Blockchains consist of four layers. We’ll use Volt Capital’s definition of each layer.
We’ll also use Celestia’s definition of modular blockchains.
If a blockchain handles all four functions, it is considered monolithic.
Within the blockchain stack, Polygon zkEVM is considered a modular blockchain and, more specifically, an execution layer. It serves as a platform for users to build applications and transact directly while outsourcing settlement, consensus, and data availability to Ethereum.
Let’s touch on how Polygon zkEVM interacts with the other layers - without getting too into the weeds.
As mentioned above, the execution layer is where users transact. When a user signs a transaction, the transaction is queued and processed - the same way it would be on Ethereum.
Once a “batch” of transactions is processed, three important things happen:
ZK proof is generated
ZK proof is posted on Ethereum
Transaction data is posted on Ethereum
ZK proofs get posted on Ethereum, which acts as the settlement layer. Ethereum smart contracts verify the ZK proofs, effectively checking the work of Polygon zkEVM nodes and making sure all computation was done correctly.
Transaction data also gets posted on Ethereum, which acts as the consensus/data availability layer. This gives all nodes the ability to see account balances and the opportunity to recreate the state themselves (very important).
Note: In one of the articles shared above, you’ll see mention of Polygon zkEVM’s “consensus mechanism” called Proof of Efficiency. This, of course, begs the question of whether or not execution layers do consensus themselves. You could argue decentralized execution layers do have some form of consensus, but the industry seems to be reserving the term for base layer ordering and finality. As an alternative, we can use the term “leader election.”
So how much Ethereum security does Polygon zkEVM inherit?
To gauge this, we’ll look at a graphic below that places Polygon zkEVM along a security spectrum. Solutions on the left are not secured by Ethereum, and solutions on the right are fully secured by Ethereum.
Polygon zkEVM (formerly known as Polygon Hermez) sits to the far right - at the same point along the spectrum as other ZK rollups and Ethereum data shards.
You might already know that Polygon has made a big strategic bet and financial commitment to ZK solutions - allocating $1 billion of the treasury to ZK R&D. As a result, there are now four ZK rollups in the Polygon suite: zkEVM, Miden, Nightfall, and Zero.
Each ZK rollup has a different design (e.g. Nightfall is a hybrid optimistic-ZK rollup), but they all fundamentally use the same ZK primitives and fully inherit Ethereum security.
Data shards are also fully secured by Ethereum as (once implemented) they will be enshrined in the core Ethereum protocol.
Side note: Rollups and data shards are way better together.
We keep saying that rollups inherit Ethereum security. You might already have a sense of how economically secure Ethereum is, but let’s make it concrete.
Because the merge is so close (at the time of this writing), let’s evaluate Proof-of-Stake (PoS) Ethereum.
There are currently ~13.3 million staked ETH on the Beacon chain, which at current token prices total ~$22 billion. Under a 2/3rd honest majority assumption, it would take ~$15 billion to corrupt the network. This is by far the best in the crypto industry.
Another aspect of economic security that should not be overlooked is monetary policy. Ethereum has implemented a very novel approach that has the potential to be more sustainable than that of Bitcoin.
As much as ZK cryptography has been around for 30+ years, practical applications like ZK rollups are still in their infancy.
It will likely take several more years before ZK rollups are considered robust like the EVM and Tendermint - technologies that have been in production and at the center of Ethereum / Cosmos development for the better part of a decade.
There is a chance that Polygon zkEVM doesn’t technically start off as a rollup but instead a validium or volition. In the documentation, this is referred to as the “hybrid mode for on-chain data availability.”
This means that Polygon zkEVM would not be fully secured by Ethereum. When validiums and volitions publish transaction data off-chain, they inherit additional trust assumptions. For example, if Polygon zkEVM were to post ZK proofs on Ethereum (use it for settlement) and post data on Celestia (use it for data availability), it would be reliant on the honest majority of both Ethereum and Celestia validators. If there’s a failure mode on either chain, the entire system is at risk.
To be clear, the Polygon team has not yet decided whether Polygon zkEVM will be introduced as a rollup or use a hybrid schema. And even if they decide on the latter, a hybrid schema is an understandable place to start. Posting transaction data on Ethereum is very expensive and cuts into the rollup value proposition. There is also a clear path toward making transaction data cheap to post, and once implemented, Polygon zkEVM could transition into being a true rollup.
Regardless - there are associated risks worth acknowledging.
Polygon zkEVM competes on a few key dimensions:
Performance and cost are pretty straightforward. Developers want their applications to be fast and cheap to provide better experiences for users.
You could argue that Polygon zkEVM competes with EVM-compatible L1s (e.g. Avalanche) on performance and cost. In the near term, this is likely true. However, we won’t spend time on this comparison because any successful L1 that remains optimized for high performance and low cost will need to dramatically increase fees or risk unsustainability.
Polygon zkEVM’s true competition is other EVM-compatible rollups - both ZK (e.g. Scroll) and optimistic (e.g. Optimism). There are many articles comparing the two roll-up types, so we won’t go into depth, but an easy way to think about it is that ZK rollups have higher performance (due to faster finality, which we’ll discuss soon), and optimistic rollups have lower costs (due to ZK computational intensity).
Note: It’s unclear how Polygon zkEVM measures up to other ZK rollups in performance and cost.
When it comes to security, Polygon zkEVM is far better than L1s and largely the same as other rollups (ZK and optimistic alike).
Lastly, there’s developer experience, which we’ll split into finality and EVM compatibility.
Finality is “the time it takes for a transaction to be considered final from the view of the base layer contract.” ZK rollups have faster native finality than optimistic rollups. This is because the optimistic rollup design relies on “watchers,” or nodes that monitor the chain for fraud. Because watchers need sufficient time to catch fraud and post proofs, finality is typically delayed a week (typically minutes for ZK rollups) - translating to longer wait times and degraded user experience. Optimistic rollups have workarounds for liquid tokens to achieve faster finality, which Alex mentions in the article above, but generally, ZK rollups have the advantage.
EVM compatibility can mean different things. Vitalik recently published an article categorizing solutions by Type (1-4), with Type 1 being the most EVM-compatible. He referred to Polygon zkEVM as a Type 3 building towards a Type 2, which we agree with.
So where does Polygon zkEVM stand today?
Through a technology lens, Polygon zkEVM “has already passed 60% of the Ethereum test vectors suite, which is used to establish EVM-equivalence. Polygon zkEVM’s zkProver is now able to process 500,000 gas within five minutes on a single CPU.” Impressive stuff - considering ZK proof generation is a computationally intensive task.
Through a go-to-market lens, it’s very close. Based on the timeline Mihailo outlined in his recent Bankless interview, testnet should be here any day now, and mainnet should be a few months away.
One must acknowledge how much ZK teams have already accomplished. ZK rollups use state-of-the-art cryptography (moon math, as some would say) and went from theory to practice years ahead of schedule. Making an EVM-compatible ZK rollup is even more challenging, and the fact that multiple zkEVM teams have mainnet in sight is mind-boggling.
Which zkEVM teams will win is yet to be seen, but one thing is for sure - the fact that we’re at the starting line is a massive win for web3.
Hopefully, this article gave you a frills-and-bias-minimized view of Polygon zkEVM and helped you see where it fits in the ecosystem.
With that said, this article is based on a snapshot in time, and web3 moves very quickly. We strongly encourage you to stay tuned because technology and even social consensus (which is, at the end of the day, what blockchain systems are rooted in) will change.
Polygon DAO is the centerpiece of Polygon’s long-term strategy of progressively decentralizing the ecosystem's growth to the community. Polygon DAO aims to represent and support various areas of interest within the Polygon ecosystem. Through Polygon Village, developers gain access to everything they need to build, grow, and showcase their Web3 initiatives; projects building on Polygon receive comprehensive, quality support.
Stay in the Loop