Modular blockchains became broadly discussed last year when Celestia announced its ambitions. Since then, the modular thesis has taken off with countless data availability layers launching, encouraged by a flourishing L2 ecosystem.
However, the idea of modular blockchains isn’t entirely new or something that only Celestia has worked on. This post will cover the differences between monolithic and modular blockchains and how Astar Network fits in.
Most of the so-called alt-L1s are monolithic chains.
What are monolithic chains?
Monolithic chains combine all the core functions of a blockchain in one. While consensus was the most debated topic for alt L1 launches, the discussion has moved on to different layers.
Blockchains have four core responsibilities:
- Consensus: coming to an agreement in a decentralized network of nodes on the validity of state transitions
- Execution: actually processing transactions submitted by users and smart contracts
- Settlement: usually happens alongside execution in modular chains. Settled transactions are considered to be final because it’s improbable that the chain is rolled back.
- Data availability: an often lauded property of blockchain is transparency. Data availability is the function that enables it by providing all the data necessary to reconstruct the transaction history and verify the current state of the network.
In monolithic chains like Bitcoin and Ethereum, the same set of validators handles all the above functions. This usually puts an automatic cap on the amount of people who can run a validator as the computing requirements increase with the use of the network. These days, Ethereum Archive nodes store over 10 TB, not the type of storage you’d have on your phone. Similarly, Bitcoin miner's computing requirements are notoriously high, especially when competing for mining rewards.
Instead of exhausting the hardware and consequently reducing decentralization, newer scaling efforts are focusing on splitting tasks.
Modular blockchains
Put simply; modular blockchains are blockchains dedicated to one or two core functions of a blockchain while leaving the rest to developers to build on top of them. That means a modular blockchain like Celestia or Avail just focuses on Data availability, leaving the other functions aside.
Src: Celestia blog
Some background on Ethereum scaling and the need for cheap DA
While at first, it might be unclear how that is beneficial, it makes more sense when looking at how Ethereum scaling has evolved.
Ethereum’s initial vision for scaling was sharding, which means splitting the entire network into different smaller blockchains. However, this would require drastic changes to the blockchain, which is why the roadmap has shifted toward Layer-2 scaling.
Layer-2 scaling increases the throughput of the underlying chain by taking transactions off Layer-1, processing them, and submitting the result of batched transactions back. This year, Layer-2s have increased Ethereum’s overall throughput by more than 5x. However, as active L2 users might have noticed, that did not necessarily mean they could maintain low fees. During high traffic times, fees on the recently launched Linea network spiked to $0.87, which seems acceptable to a DeFi degen but is not in the context of mass adoption.
One reason for the fee spike is a natural adjustment of price during high demand, as well as an anti-spam mechanism. However, fees are often also driven by the costs these Layer-2s incur when publishing their data to Layer-1, Ethereum.
Up to 90% of the operating cost of Layer-2 are made up of fees from publishing transactions ata to the mainnet. And even with the eventual introduction of ProtoDanksharding and additional storage for L2 data on the mainnet in Blobs, the costs will remain fairly high.
Enter Modularity
Layer-2s need to publish their transaction data so that anyone can verify their state and ensure that their transactions are legitimate. However, they are not required to use the Ethereum mainnet for data availability. Instead, Rollup and L2 scaling solution operators can significantly reduce their costs by using solutions such as Celestia, Avail, or EigenlayerDA, which tend to have drastically lower storage costs.
Lower costs to publish data are passed on to users, creating an overall better experience for dApps. And even though Data availability might be the most obvious part in the stack that current rollups might want to optimize for, rollups themselves are nothing but modular blockchains for execution.
The overall idea of modular blockchains is that instead of forcing all different core functions on the same set of machines, it’s better to distribute tasks and optimize layers for their specific tasks. This also allows developers to create combinations of layers that work best for their use case. Just like building something from Lego blocks, they can combine the cheapest DA with an execution-focused layer like FuelVM and the security of Ethereum provided by Eigenlayer through re-staking. This approach reduces the cost of developing entire chains from scratch while offering ways to rely on other networks’ security instead of bootstrapping.
In short, the benefits of modularity include dropping cost and time to develop and launch new chains, a level of flexibility unmatched by monolithic chains, and scalability.
As mentioned before, though, modularity isn’t exactly new. It has been part of the narrative of appchain ecosystems like Polkadot and Cosmos since their inception. Both have built modular frameworks allowing chains to pick different modules and combine them to fit their needs.
That’s also how Astar Network started.
And what about Astar?
Astar Network started as a parachain on Polkadot, creating a smart contract hub that other parachains in the ecosystem could tap into. As such, one could assume it’s a monolithic chain, but because it’s part of the Polkadot environment, things are a little different. Essentially, building on Polkadot allows parachains to inherit the security guarantees of the underlying chain, making it akin to the current modular chains. The relay chain takes care of security and consensus, while developers on the parachains can focus on their unique strengths.
Having advanced to one of the most successful parachains, establishing dApp staking as a mechanism to power community-backed builders, Astar Network has recently partnered with Polgon to expand to the Ethereum ecosystem via a zkEVM L2.
With that, we’re also further embracing the modular narrative, having gained experience in building with modular frameworks in the past two years.
Outlook for modular chains
The future of blockchain looks modular. While there is space for monolithic chains that have already implemented sharding or found their own niche, it increasingly looks like we’ll be building modularly for the time to come.
While there is a lot of excitement around the idea of stacking layers like they are cake, it’s worth highlighting that a few challenges remain.
In the modular paradigm, security rests heavily on the underlying layer. Devs will have to choose carefully as not to run into situations where one layer failing halts operations on the rest of them. That’s the equivalent of forgetting the biscuit bottom in a black forest cake (it’ll all fall apart…)
Additionally, while optimizing each layer can result in an experience better than the sum of its parts, the cost is increased complexity. Dealing with fraud proofs or validity proofs to ensure that computation on other Layers has been correct is challenging.
Not to mention the further fragmentation that modular blockchains introduce in an already fragmented L2 landscape. For the modular paradigm to be successful, a lot of work will have to go into the user layer. At the end of the day, it’s unlikely users will care how much Ethereum-aligned you are, nor want to spend 50 minutes trying to understand what “inheriting security” even means.
Src: https://x.com/miyuki%5F%5Feto/status/1733708553712709899?s=20
Nevertheless, with technology improving rapidly and the overall focus shifting to catering to consumers via consumer apps, modular blockchains will certainly play a role in enabling adoption.
