Celestia's 'Modular Blockchain' Attracts Investors Looking for Scaling Breakthrough
Warm Reception to Network Design Contrasts With Aptos Rollout
By: Owen Fernau •Dive
Two days later, a similar project called Celestia announced a $55M raise to noticeably less groaning and even a solid dose of enthusiasm.
The key difference: Celestia uses a newish design called “modular blockchain architecture” while Aptos is considered a “monolithic blockchain.”
Investors and devs appear to be far more excited about the former rather than the latter.
Under a monolithic approach, the four main functions of networks — executing and settling transactions, agreeing on their ordering, and making their data available — are executed by the same system.
This contrasts with the design philosophy employed by Celestia, which counts Bain Capital and Polychain Capital among its top investors. Modular blockchain architecture is based on the idea that the four functions don’t necessarily need to be handled by the same system. They can operate interdependently.
Comparing monolithic blockchain and modular blockchain architecture.
“Data availability is essentially making sure that the data behind some transactions has been published to the world,” John Adler, Celestia’s co-founder said at ETHDenver in February. “You need this because if you don’t have this data, you don’t actually know what’s happening in the blockchain.”
Celestia is aiming to offer both consensus and data availability, to projects which are serving as settlement, and especially execution layers.
First among potential users of Celestia are rollups, technologies which help Ethereum scale by executing transactions on a separate chain. Rollups then post their transactions to Ethereum’s main chain to be settled.
Right now, rollups use Ethereum for data availability, but buoyed by its $55M raise, Celestia is angling to take over that role.
‘The thesis of modular blockchains is that a single blockchain doesn’t need to handle all these components on its own.’
Celestia listed three blockchain projects as users of its data availability layer in its announcement of its most recent raise. The project is also giving 26 of what it called “Modular Fellows Cohort,” a $3,000 a month grant for three months. Recipients include Scott Sunarto, who’s worked for both the Ethereum Foundation and Uniswap Labs.
At a high level modular blockchain architecture may allow different blockchains to address what blockchains like Aptos try to do all at once. Adler believes that this will allow specialization at each of these layers which will lead to better performance.
It also may end what the crypto community may be fed up with — the constant launches of new blockchains with parameters tweaked to increase transaction capacity, yet without real innovation.
For example Binance Smart Chain was forked from Geth, an implementation of the Ethereum protocol. A few changes were made to increase transaction throughput, but this represented nothing beyond a linear improvement with the added downside of increased computational power required to be a node in the network.
This increase in requirements of nodes, lowers the possibility of decentralization because it requires more advanced hardware. This represents a potential race to the bottom which has new monolithic blockchains continually launching touting higher scalability and lower fees, but at the expense of decentralization.
Modularlization is meant to let projects address and specialize in different layers in the blockchain stack.
Al-Bassam posted this chart with his paper.
For those betting on the modular blockchain future, Celestia says it will deploy a token to secure its Proof of Stake network on its Frequently Asked Questions page. The token will also be used to pay transaction fees.
“The thesis of modular blockchains is that a single blockchain doesn’t need to handle all these components on its own,” Ahmed said. “Instead, by disaggregating these core components, individual blockchains can focus on specializing in a specific area, leading to significant optimizations.”
Mustafa Al-Bassam, co-founder and CEO of Celestia Labs, the primary company behind the Celesita protocol, saw Aptos as yet another Layer 1 which doesn’t push crypto forward.
“Crypto has a serious problem,” he tweeted. “We’re stuck in an endless cycle of new [Layer 1] smart contract platforms. Each purport to fix the problems of the Layer 1s in the previous cycles.”
The negative reaction to Aptos supports Al-Bassam’s point. Crypto can seem perpetually stuck in an infrastructure-building stage, with project after project selling picks and shovels to people who would rather turn around and sell those tools rather than use them to build something useful.
“Aptos was a tipping point,” Ahmed added.
Tom Dunleavy, a senior research analyst at the crypto information provider Messari, thinks success for blockchain networks lies more in the application layer. To that end, he doesn’t think a Solana or Aptos can be ruled out of becoming home to apps which attract hundreds of millions of long term users.
Eshita Nandini, another research analyst at Messari, also sees potential pitfalls of the modular model. “Modular blockchains might have to go through things like bootstrapping a new validator set, which sets something like an established [Layer 1] ahead,” she told The Defiant.
A blockchain’s security is generally dependent on how easily its validators can be co-opted, so starting with a small set of validators in a blockchain’s early stages can make it extremely vulnerable.
A Modular World
Still, Nandini sees definitive movement towards a more modular world. “Looking at Ethereum, which was initiated as a monolithic chain, it has now pivoted its entire roadmap to support rollup activity,” Nandini said. “Rollups are essentially modular execution layers.”
With consensus and data availability distinctly separated from settlement and execution a large number of different configurations between blockchains can emerge.
Maven 11, one of the investment firms which backed Celestia, posted a diagram of the many ways in which the different layers of blockchains can come together under the modular paradigm.
There are different ways to configure blockchain designs.
“With modular architectures like Celestia, rollups can use it for data availability and consensus, while choosing to self-settle — this would be considered a sovereign rollup,” Nandini said in an email. “It can also choose to settle on a monolithic L1.”
Looking forward, projects like Fuel, which bills itself as a modular execution layer, is building for a future where the components of blockchains are disaggregated.
“Modular execution layers are execution systems which are designed for modular blockchains,” Nick Dodson, CEO of Fuel Labs, the company developing Fuel, told The Defiant. “This means we are calibrating all our designs for a much higher bandwidth than what Ethereum L1 and L2’s or monolithic chains typically provide in their current instantiations.”
Fuel can take what Dodson called a “rollup configuration,” though its design means it can also take the form of a Layer 1, or other non-rollup scaling solutions like sidechains or state channels, the CEO said.
This diagram showing different configurations for Fuel.
Maven 11, Bain Capital Ventures, and others, have invested in both Celestia and Fuel, indicating a burgeoning group of investors betting on a modular future.
Dodson sees a modular blockchain world as one where developers and users will have more choice relative to using blockchains with more monolithic blockchain designs. “The modular paradigm is incredibly flexible which will allow the market to choose how they would like things settled or what kind of execution guarantees they are looking for,” he said.
With this week’s reaction to the launch of another monolithic Layer 1 in Aptos, the crypto community at large may have definitively signaled that it’s time for something new.