# How to Raise the Gas Limit: Use Ultra High Resolution Data by Storm Slivkoff | Devcon SEA

- Speakers: [Storm Slivkoff](https://streameth.org/speakers/storm-slivkoff)
- Channel: [Devcon](https://streameth.org/devcon)
- Date: 2025-10-09
- Duration: 08:05
- Topics: Science & Technology
- Watch: https://streameth.org/watch/yt-v8Po5KOutow
- YouTube: https://www.youtube.com/watch?v=v8Po5KOutow

## Description

Recent advances in EVM data processing enable a more rigorous approach for understanding and enacting Ethereum’s scaling roadmap. In the past, discussions around whether to raise Ethereum’s gas limit have been held back by imprecise terminology and a lack of detailed quantitative evidence. The debate is often “vibes-based”. Leveraging ultra high resolution datasets enables a more scientific understanding of the gas limit, including issues like state growth, hardware bottlenecks, and gas pricing.

Speaker(s): Storm Slivkoff
Skill level: Intermediate
Track: Core Protocol
Keywords: Layer 1, Gas, Scalability, bandwidth

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] so let me just maximize the [Music] window great hi everyone um today I want to talk about the gas limit and how we can get a a better understanding of it so this question comes up all the time should we raise the gas limit lots of people have strong opinions on this but the truth is nobody really knows the answer and this is because nobody has detailed quantitative evidence on what the optimal gas limit is so ethereum is a complex system and downstream of the gas limit there are hundreds of different parameters and bottlenecks and metrics that need to be considered so to tweak the gas limit um we need detailed quantitative evidence to justify what we're doing rather than relying just on intuition but this is a really hard problem so over the past year um I've been trying to understand the gas limit in a more data driven way and the most interesting thing has been learning what the true bottlenecks of ethereum actually are so for example there's a common uh belief or common view that the main problem with raising the gas limit is State growth the state grows too big and this is somewhat true but it turns out that other bottlenecks are more significant like history growth and bandwidth so this is a uh this is a graph showing that history growth is a much larger storage burden for node operators um compared to State growth so I think that the only way for us to gain Clarity on SC scaling ethereum is doing this type of datadriven analysis sort of times 100 and leaving no stone unturned so how can we use data to supercharge gas limit research I want to describe a general approach that can help us answer three of the most important questions related to the gas limit so the first problem is how do we price long-term storage storage is a lot different than other Hardware resources like uh CPU or bandwidth because storage creates a long-term commitment a long-term burden uh whereas other resources are used instantaneously and right now I think storage is mispriced especially on l2s and we don't really know the true cost that storage imposes on the network a way to approach this problem is we come up with a new pricing scheme for storage such as multi-dimensional gas or state expiry or some ZK thing and then we do a datadriven comparison between the new scheme and the current scheme and this can be extremely detailed we can look at every single transaction and block and protocol in ethereum's history and we can rigorously examine um all the downstream effects of this new pricing scheme another gas limit problem is how do we make the worst case closer to the average case um so that we can do more in each block without risking a denial of service attack and so EIP 7623 um is a great example of this it brings the worst case block size down closer to the average case uh block size and you might ask yourself if we can do this for Block size can we do it for a bunch of other block metrics like the number of times that a certain op code gets used a simple solution is let's just impose new rules on block building we can evaluate this solution sort of using the same approach um that I just described where we look at the entire history of ethereum blocks we can pull in um old snapshots of the me me poool and we can evaluate um a detailed breakdown of how these new constraints might affect um all the different um things that we care about and then the final boss of the gas limit problems is what is the optimal gas limit and how do we safely safely balance Network Health with network performance and again this is a hard problem because ethereum is a complex system with hundreds of different parameters and bottlenecks and metrics but it's not an impossible problem we can model all of these different bottlenecks and metrics as a function of the gas limit and use data to inform those models and I think ultimately this is the endgame approach for understanding the gas limit and guiding our decision on what it should be if you want to learn more about how to actually dig into the crypto data um come to my other longer talk in an hour on stage three and thanks for listening thank you so much are there any questions from the audience and maybe people are still typing but i'm interested to know what what was the motivation behind deep diving into this topic more like what what what like inspired you yeah so right now ethereum um it's sort of notorious for being more expensive than other blockchains and this is due to the gas limit sort of of uh clamping down on the amount of transactions per second that ethereum can handle and there are really good reasons for doing this um if you allow too many transactions that can compromise Network Health decentralization other things we care about uh but we at the same time we want to make sure that we're in the right sort of Sweet Spot um in that trade-off space okay thank you this so really one question what kind of tools you use to collect the date are they mature or we need better tools um at this point the tools are they're getting really good um and I actually go into detail um like I said in this other talk um in an hour if you can make it but the short answer is I use cryo which is a tool I built to extract a bunch of blockchain data sets and then I use polers to analyze that data um and slice it up in a bunch of different ways and it's really efficient and really robust okay thank you um the questions are popping in now in um which bottleneck do you see as lower hanging fruit to fix so uh probably history growth is the most obvious example there history takes up a huge amount of space on every full node and we actually have a solution for this that's uh people have been working on for a long time called EIP 4444 where um instead of every node um keeping a copy of the entire history we move um sort of toward a different distribution scheme and then all of the nodes will have a lot more runway for either the state to grow big or just make it less burdensome to run a node thank you so much S I think time is up those who have um additional questions as he said there will be long format of his talk maybe you can follow there and then place your questions there otherwise he's there for answer questions after um after this session thank you so much storm thanks thank you [Music]