Cartesi: Layer 1 vs Layer 2 / Memory vs Hard Drive

6 min readJul 20, 2022


RAM memory and Hard Drives, strong competitors for storing solutions, or not.

You may have heard blockchain terms like Layer 0, Layer 1 or Layer 2, if not, we’ve written about those in one of our previous Cartesi’s articles. In this one we’re going to dig deeper into layer 1 and layer 2 as key layers of blockchains in the present and future.

When you hear about how to scale blockchain performance, usually there are two main aproaches for that: on the one hand is the path of creating more efficient Layer 1 blockchains that can handle more transactions per second (TPS) and shorter block finality times, and on the other hand is the path of creating off-chain Layer 2 solutions that settle to existing Layer 1 blockchains.

Let’s discuss about what will be the approach in the future, but first we’ll define a bit what these two type of blockchain layers are about.

Layer 1, what is it?

A layer 1 is what you traditionally think of a blockchain, that’s a network of computers distributed worldwide, that reach consensus on information in a decentralized and trustless form, so that that information is stored into the network of computers in a non cheatable way. Usually these networks are public, permissionless and transparent, so that any computer can take part of the network as a node and the users can inspect freely what information is inside the network blocks.

Most common examples of Layer 1 networks are: Bitcoin, Ethereum, Solana, Near, Cardano, Polkadot, etc. All these platforms have native tokens (BTC, ETH, SOL, ADA, DOT, etc.) that can be transferred between the accounts of its users.

Layer 1 networks have the big feature of decentralization, but this comes with the problem of not being secure or not being performant enough. This is also known as the blockchain trilemma, you can only achieve 2 of the following 3 features: decentralization, security and scalability. Here is where Layer 2’s come into play.

Layer 2, what is it?

The main path forward in computer science for solving some of these problems, is to create layers on top of the layer that have the issues. These new layers help solve the issues by using advanced techniques that help achieve the goal to the base layer.

Some Layer 2 approaches for scaling Ethereum network

For example, hard drives are devices for storing big amounts of data, but they are not as fast as other solutions such as chipset memories like RAM. Today computers use RAM memory as a storage layer on top of the hard drive, so that operations that need blazing fast data access(like applications, video playback, music production, video games, etc.) use mainly the RAM memory for instant data manipulation and use the hard drive to load or save information in/from the RAM memory, thus achieving a top performance close to the RAM memory but with the size capacity of a hard drive. That’s it you get the best of both worlds by combining them.

In the context of blockchain layers, Layer 1’s are the layers that are in charge of storing the data securely and in a highly decentralized manner, and Layer 2’s are the layers sitting on top that can handle the performance and scalability that the layer 1 is not able to achieve. Having in mind the previous example, the Layer 1 would be the hard drive, and the Layer 2 would be the RAM memory that enhances with performance to the hard drive.

In the case of Bitcoin, the most well known layer 2 is the Lighting Network. In the case of Ethereum there are several projects that help solve the scalability issue in different ways, but there’s one we like the most.

Cartesi, Ethereum Layer 2

We’ve dedicated some articles to Cartesi, a project that addresses this on Ethereum but has kind of a unique approach to it, and provides much more than of regular Layer 2 approach.

Cartesi uses optmistic rollups as their technical approach to solve the scalability issues of Ethereum. This same approach is used by other Layer 2 networks, but Cartesi has some big additional benefits that made it unique.

Main Cartesi feature is that implements a Linux Operating System Runtime that makes It kind of a Layer 2 that provides an operating system on top of the Layer 1. This is a very powerfull and flexible solution that brings a lot of benefits to Ethereum itself, for example, huge ones in dApp development (you can learn more about by reading the article “How dApp development benefits from Cartesi?”).

Another great feature is that it’s blockchain Layer 1 agnostic technology, so can be integrated not only into Ethereum but into other Layer 1 blockchains too (Avalanche, Elrond, Binance, Polygon, Iotex), giving great flexibility and portability for a future of interconnected blockchains.

Will Layer 1s still need Layer 2s in the future?

If you’re living in a cave you probably don’t know that Ethereum is migrating from its current consensus algorithm (Proof of Work) to a more performant one (Proof of Stake) in a near future. This will improve a lot, Ethereum scalability and transaction costs for sure. You may think that this type of Layer 1 evolutions will some day remove the necessity to use added technology layers on top of it, because they can perform well enough by themselves alone.

Our prediction is that, that won’t ever happen. You may increase the capacity of a Layer 1 network making it more performant and able to handle more transactions, but also the user base will grow putting much more transactions into the network, also applications will tend to grow in functionality and thus in transactions sent into the network, newer types of tokens will appear (utility, NFT, sovereign identity, you name it…) that will provide more utility to the Layer 1 and again more transactions to it, and so on and so forth.

Technology will always try to achieve the best possible things it can, using all the techniques and tools that humans create. Computer science solutions have never been a 1 layer solution, but more of an onion with several technology layers that in combination can become a much better solution than any of those layers standing alone by itself.

Have hard drives replaced RAM memory after several decades of hard drive development? They haven’t, and does not seem the case for the future. So long live to Layer 1s and Layer 2s in the blockchain world.

Hope you liked this new article on the Cartesi series and if you want to dive deeper check the links below.




We're a company of Blockchain passionate individuals that aim to help decentralize the world while having fun with technology.