# Transitioning from an L1 to an L2: A case study by Marek Olszewski | Devcon SEA

- Speakers: [Marek Olszewski](https://streameth.org/speakers/marek-olszewski)
- Channel: [Devcon](https://streameth.org/devcon)
- Date: 2025-10-09
- Duration: 08:36
- Topics: Science & Technology
- Watch: https://streameth.org/watch/yt-JerIZmTt-tE
- YouTube: https://www.youtube.com/watch?v=JerIZmTt-tE

## Description

This talk will cover the learnings from cLabs' experience rebuilding Celo from the ground up as an L2. We hope that it can be a useful case study for other L1s to follow.

Speaker(s): Marek Olszewski
Skill level: Intermediate
Track: Layer 2
Keywords: Layer 1, Layer 2s, Rollups, Scalability, Optimistic rollups, Use Cases, learnings, technical

Follow us: https://twitter.com/efdevcon, https://twitter.com/ethereum, https://warpcast.com/devcon
Learn more about devcon: https://www.devcon.org/
Learn more about ethereum: https://ethereum.org/ 

Visit the https://archive.devcon.org/ to gain access to the entire library of Devcon talks with the ease of filtering, playlists, personalized suggestions, decentralized access on Swarm, IPFS and more.

Devcon is the Ethereum conference for developers, researchers, thinkers, and makers. 
Devcon SEA was held in Bangkok, Thailand on Nov 12 - Nov 15, 2024.
Devcon is organized and presented by the Ethereum Foundation. To find out more, please visit https://ethereum.foundation/

## Transcript

[Music] hey everyone my name is mar oepi and once again we're going to be talking about um cell's transition from an L1 to an L2 uh you've probably heard that we're making this big transition and we really hope that we can become a case study for other l1s to follow suit and so today will be a talk really targeting those other l1s with our learnings I tried to make this fun for you all so uh I made this in the style of Zelda any any Zelda players in the audience cool we got a few hands up great well hopefully you'll enjoy the slides um and so what does cell's transition to an L2 look like it it looks like us taking our evm core which is shown here as kind of our little protagonist uh and all of the nice additional features that we've built into the sell blockchain like the ability to pay for gas with tokens like the ability to send value to phone numbers and of course Ultra green money this um mechanism that offsets all of the carbon that the network produces every time we transact and we've taken all of this and we've repackaged it into an OP stack base L2 that's sitting on top of ethereum and so once again we really hope that this transition that we're going through can be a really nice case study for other L1 ones to potentially follow suit as well and this talk is really um almost a map you could say um of you know five learnings from our journey that that these other l1s can learn from okay so learning number one um First Learning is if you're going to do this transition you should do it as a hard Fork you know when when many people think about doing this transition they um immediately think okay this is the way to do it right let's just launch a second L2 chain um and uh and then we can have our users figure out how to migrate from the L1 to the L2 and this is you know maybe appealing because it's not much work to do but it's it's a lot of work for the end users and for developers building on these chains and so this is not what we're doing uh instead we're actually hard forking to become an L2 Network at a certain block height where um before that block height the network will be progressing as an L1 and then after that block height the network will change to a new client uh that will then um run as an actual L2 and you know we've learned over this period where we've been building this out that this is in fact quite difficult um you know we like to say that this is is actually more than just a hard Fork it's actually a hard hard Fork two rocks and a fork and it's hard you know for for all of the obvious reasons right like you're um you know the metaphor of rebuilding a plane while it's flying at cruising altitude you know I think is very apt you know we're literally moving the engines um back and forth um rebuilding the plane while it's in the air and you know the advantage of this is that in the end only full nodes really need to do any work around this migration only they need to actually upgrade and um participate in in this transition your end users and developers building on top of the chain they don't have to do any work right all of this is transparent for them really only the full nodes have to do that work and again the people who are running these full nodes they're technical they have experience with hard Forks this is not something that's too complicated for them they can handle that so on to learning number two learning number two is to use an existing L2 stack you know we thought early on hey why don't we build our own L2 stack let's just upgrade our existing client into a new L2 stack but we didn't want to increase the number of L2 Stacks that exist in ethereum there's already a lot out there it's a lot of fragmentation that this is causing and we concluded that it's actually better to just pick an existing stack and so we ended up choosing op stack in igda for data availability of course we'd recommend that you uh do your own due diligence and figure out which stack is best for you um but certainly I think picking an existing stack not introducing more fragmentation I think is the way to go this of course means that you know we're effectively switching clients at the time of the uh upgrade so there will be an L1 client that's running today and an L2 client that will be running after the hard fork and you know this again means that there's a little bit more complexity around this hard Fork right full node operators will have to actually shut down one client and start a new client and this again adds to the hardness of this hard Fork but again we think it's we think it's worth it there's a bit of additional complexity for archive node operators because archive nodes of course should be able to respond to queries about all uh blocks and all data from the whole history and so archive node operators will actually have to run two clients that are connected to each other so that they can respond in unison to RPC requests across all time periods okay on to learning number three this is related to the last one um when you have a new client you run into the situation where the new client has to somehow be able to interpret the chain data from the first client and you can either modify the new client to be able to do that or you can migrate your chain data so that it it fits a format that the new client can understand and we think that this is the way to go and so that's exactly what we're doing at the point of the hard Fork we're going to be running a chain data migration script that repackages the data in a way that's compatible with the new client and this dramatically simplifies um a lot of work the new client doesn't have to um be Rewritten to understand the chain data of the old client okay learning number four Embrace ethereum so after we announced that we wanted to make this migration um there were a few other L1 chains that also made similar announcements but if you looked a little closer it was was clear that they were not really embracing this idea of settling on ethereum becoming a true L2 instead they were talking about you know canonical Bridges and things that looked a little bit more like halfway towards actually creating L2 and so our advice for for these folks is you know just take the leap of faith Embrace ethereum um you know I think your communities are going to be excited and and the ethereum community has been really welcoming of us and we don't think that you're going to regret it related to this I think you you might have some differences from ethereum in your protocol and and our advice here is to really distill those differences down to a patch Q on top of gu okay the final learning learning number five is to rethink where your token lives um this was a big one for us um you know as an L2 you have the choice to actually have a token that lives on ethereum in addition to the L2 and so that's what we're going to be doing we're going to be be deploying a new token on ethereum uh with the max Supply and then we're going to put the entire supply of that token into the native Bridge uh so that it's there already in escrow backing in effect all of the tokens that exist on the L2 and so this is a really interesting design we think that um it's one that others should should follow and we're really excited about it okay so final thoughts if you are thinking about uh transitioning to an L2 um then our recommendation for you is just do it you're not going to regret it uh take the leap of faith um you know we did and and we're really thrilled about it we highly recommend doing the same and if you want to follow our progress um you know we have one more test net that we're going to be upgrading next month and then at the beginning of next year mid January we'll be hopefully upgrading our main net if everything goes well with that test net upgrade and so yeah U mainnet is coming and we hope you'll uh be excited for it thank you so much thank you mik if you have questions for mik please find him uh beside the stage thank you a lot