Little Learnings #4
MegaETH public mainnet is soon, but what is MegaETH, and how does it work?
Welcome to Little Learnings, a series of educational posts I release every Friday. The intention behind these is to break down some new interesting topic or development happening in the crypto space. God knows there’s enough happening in crypto, and it’s often complicated — so I’m here to try and simplify it a bit.
This post is sponsored by Open Campus
A fitting sponsor for this free newsletter, Open Campus is building the onchain financial layer for education.
They recently announced a partnership with the Government of Madhya Pradesh (a state in India) to bring 50 MILLION academic records onchain.
Every graduate in the state will have a verified digital credential on EDU Chain.
This is a pretty incredible achievement and will push forward the future of the role of blockchain technology in education.
Learn more here (check it out, it’s genuinely a very cool partnership).
MegaETH public mainnet is soon, but what is MegaETH, and how does it work?
Disclaimer: I invested in the Sonar round for MegaETH.
I’ve written about MegaETH before, but haven’t really broken down how it works. That’s what today is for. The timing feels appropriate, as they just announced that their Public Mainnet is going live on the 9th of Feb.
Starting from the basics, what is MegaETH? It is an Ethereum Layer 2 designed for real-time performance. Not the “fast” performance you see on Base or Solana where you submit a transaction and it’s approved in like half a second. Real time means real time, and the only way to do that is by being able to process tens of thousands of transactions per second.
Base is sitting at ~28 TPS, Solana at ~3,700, and MegaETH at ~42,000! 🤯
Their target is 100,000 transactions per second with sub-millisecond latency. To get there, the team rebuilt how blockchain nodes operate from the ground up. Below is my understanding of how it works.
It’s a bit technical, and I might have got something wrong (or more likely, might have painted an incomplete picture). I chose to make it as simple as possible to understand; my apologies if anything important is overlooked, please let me know in the comments and I will amend the article if that’s the case!
How are they able to achieve such a high TPS?
Traditional blockchains force every node to do everything: consensus, execution, validation, storage. This redundancy is what kills throughput. The slowest node in the network becomes the bottleneck for everyone.
MegaETH takes a different approach. It splits responsibilities across specialized node types, each optimized for a specific task.
The system runs on four node types:
Sequencer nodes handle the heavy lifting. They order and execute transactions, assemble blocks, and broadcast results. MegaETH runs a single active sequencer on high-end hardware: 100 cores, 1-4 TB of RAM, 10 Gbps network. The entire blockchain state lives in memory, eliminating I/O bottlenecks. This enables 10ms block times.
Full nodes receive transactions and apply state changes. They can re-execute for verification but do not need to. They run on standard hardware, similar to existing L2 requirements.
Replica nodes maintain an up-to-date view of the chain without re-executing transactions. They apply execution results directly from the sequencer. Hardware requirements stay low because verifying proofs costs less than running execution.
Prover nodes generate fraud proofs independently. They verify blocks and can operate out of order, reducing network load.
This separation means the sequencer can push maximum performance without waiting for slower nodes to catch up.
How it works in practice: a user submits a transaction through an RPC node. The sequencer receives it, executes it, and includes it in a block. The sequencer then broadcasts execution results (receipts, state changes) to replica and full nodes. It also submits blocks to EigenDA for data availability and to Ethereum mainnet for finality.
Because there is only one sequencer during normal operation, there is no consensus overhead. This means there’s no waiting for agreement between nodes and transactions can process immediately.
As with other L2s, MegaETH remains fully compatible with the Ethereum Virtual Machine. This is honestly one of the best things about L2s, as devs can deploy existing smart contracts without any real modification if they want, and they’ll just work.
Tradeoffs and Mitigations
Those familiar with the Blockchain Trilemma will know that there are always tradeoffs. To achieve such insane scale, you’re gonna sacrifice something else, at least to begin with.
Having a single sequencer will naturally introduce centralization risk. If you want to read more about this risk and some of the others, I recommend this X post and then this subsequent response which addresses the concerns.
The TLDR is that there are real centralization concerns at this point. These apply to all L2s, and it is on them to progress to a robust stage. It is on the community to hold them accountable.
One of the coolest things about MegaETH, to me, is that real-time performance unlocks applications and use cases impossible on other infrastructure: onchain games, extremely high-frequency trading, AI agents transacting autonomously and instantly. Of course these things are possible on other chains, but MegaETH takes it to the next level.
Whether it ultimately pays off and works will depend on what developers build, and whether users end up using the chain. I’m generally pretty bearish on new blockchains. I don’t think we really need more L2s or L1s. MegaETH is an exception in my mind, and I am (cautiously) optimistic that it will succeed where so many others have failed.
Thanks for reading, see you next week with another little learning!
Disclaimer: The content covered in this newsletter is not to be considered as investment advice. I’m not a financial adviser. These are only my own opinions and ideas. You should always consult with a professional/licensed financial adviser before trading or investing in any cryptocurrency related product. Some of the links shared may be referral links.
I enjoy your articles a lot and appreciate how clearly you explained MegaETH. But I find it a bit strange that someone who has been in crypto for so long does not mention the earlier pioneer of this kind of architecture. This multi node design was introduced years ago by another project and it's already much more established. MegaETH may be pushing performance further, but the main idea is not new and it feels worth acknowledging that history and broader context.