The CLOBs on Blobs Opportunity
The CLOBs on Blobs Opportunity
The blockchain scaling race isn’t about who can hype the loudest, it’s about who can capture the most value from real usage. Among the most demanding of these use cases are on-chain Central Limit Order Books (CLOBs) - a system where buyers and sellers place orders that are matched transparently in a shared order book, commonly used in both traditional and crypto exchanges for efficient, fair trading.
Bringing CLOBs fully on-chain has always been the “endgame” for DeFi, but until now, the infrastructure simply couldn’t handle the throughput and data requirements. That’s changing fast, largely due to advancements in scaling infrastructure like Celestia and other data availability (DA) layers.
Understanding Data Availability and Blobs
Core to any blockchain's operation is the need for all transaction data to be recorded and publicly accessible, a property known as data availability (DA). This fundamental transparency is what makes decentralized systems verifiable and trustworthy. For financial applications like CLOBs, which process continuous streams of orders, modifications, and cancellations, the demands on this underlying data infrastructure are immense. Each trade, each order update, needs to be rapidly and affordably documented on the blockchain to maintain market integrity and efficiency. These large, temporary data segments posted to DA layers are often referred to as "blobs."
For applications like CLOBs, where vast numbers of interactions occur rapidly, ensuring this data is efficiently managed and stored becomes critical.
Why CLOBs Need Obscene Amounts of DA
Image: Modular approach for CLOBs
Data availability (DA) means all transaction data for a blockchain is published and accessible to anyone who wants to verify the chain’s state. Without strong DA, users and nodes can’t be sure they have the full information needed to check the validity of blocks or detect fraud.
Each action in a CLOB (whether placing, modifying, or canceling an order) creates a transaction that needs to be recorded on-chain both rapidly and at low cost. For a CLOB to compete with a centralized exchange, it needs to process thousands of transactions per second, with minimal latency and cost. This is where the modular approach to data availability (DA) comes in.
Unlike Ethereum or Solana, which bundle all core functions together, using specialized infra means developers can optimize for the specific needs of high-performance CLOBs - especially when it comes to scalable, affordable data availability.
How Much Data Is That?
The data throughput required for a single, high-performance on-chain CLOB can easily saturate a general-purpose L1. To compete with centralized exchanges, these platforms must process a constant stream of orders, modifications, and cancellations, each one a transaction that demands onchain data space.
This is not an outlier but a baseline for what future high-performance DEXs will demand. Looking ahead, A recently launched Celestia testnet demonstrated 21.33 MB/s of data throughput, equivalent to over 200,000 transactions per second, or more than 30 times the throughput of Solana.
Monolithic vs. Modular: The Core Architectural Debate
Image: monolithic vs modular stacks for development
The pursuit of high-performance on-chain Central Limit Order Books has intensified a fundamental architectural debate within the blockchain space: is it better to have a single, highly-optimized L1 handle all activity (the monolithic approach), or to use a specialized, multi-layered system (the modular approach)?
The monolithic view is that a single blockchain should be fast enough for any application, including a demanding CLOB. Proponents believe this model is better for value accrual, as it keeps all transaction fees and economic activity contained within the L1, benefiting the network's validators and stakers.
The modular perspective offers a contrasting argument. It suggests that the immense transactional load of a high-performance CLOB (potentially millions of orders daily) would disproportionately burden a monolithic chain, thereby degrading the user experience for all other applications. With this approach, rollups do not simply redistribute existing activity; instead, they unlock entirely new use cases and economic opportunities that are infeasible on the L1 due to its structural constraints. As a result, the data availability fees paid by a successful rollup to a modular layer represent additional revenue generated.
The data underscores the technical trade-offs. An L1 like Ethereum has a sustainable DA capacity of just 0.0625 MiB/s and a slow 12-minute finality.
In contrast, Celestia offers over 21 times more throughput at 1.33 MiB/s with a rapid 6-second finality. This combination of massive scale and low latency is why a specialized approach is gaining traction, offering a solution that even the fastest monolithic L1s like Solana, Monad, or Fogo struggle to match in both raw capacity and architectural flexibility.
This level of throughput is not just a technical milestone, it’s a necessity.
A single high-performance CLOB can generate more data than the entire daily capacity of Ethereum’s blobspace. Even after EIP-4844, storing a MiB of data on Ethereum is significantly expensive, and the total daily capacity is a fraction of what a single CLOB might need. Solana, Monad, Fogo and other L1s offer higher throughput than Ethereum, but are still constrained compared to what a specialized chain using altDA can provide, and can’t offer the same flexibility in execution environments.
Furthermore, a key advantage for rollups, especially for CLOBs, is the ability to implement custom sequencing rules, such as "cancel-priorities," which can prevent front-running and ensure fairer trade execution. This level of granular control is generally not available on a general-purpose L1.
Rollups can also offer ultra-low latency "soft confirmations" in as little as 3-5 milliseconds. This speed is critical for trading, where every millisecond affects outcomes.
This combination of massive scale and low latency is why DA layers are purpose-built for the demands of on-chain CLOBs, offering a solution that even the fastest L1s cannot match in both capacity and flexibility.
Hence the meme, CLOBs on blobs.
The Modular Stack: The Right Architecture For CLOBs
Why a flexible environment is necessary for CLOBs
This distinction is key. When developers build on an L1 like Solana, they are building on a pre-set foundation, meaning they must use its native platform, in this case, the Solana Virtual Machine (SVM), or the Ethereum Virtual Machine (EVM) for Ethereum. While powerful, this locks them into a specific set of rules and limitations.
A modular stack offers a different architectural approach: since a modular DA layer only handles data availability and consensus, it doesn't impose an execution environment. This gives developers the freedom to build with it as a component. A team can create a rollup using the familiar EVM, a high-performance SVM like Eclipse did, or even a completely custom virtual machine designed solely for their CLOB’s needs.
Image: Modular architecture for CLOBs
This flexibility is important for high-performance CLOBs, which may require specialized features like custom transaction sequencing to prevent front-running, something a general-purpose L1 might not offer out of the box. Using a single sequencer, often colocated across multiple regions, offers far faster throughput and lower latency than a network which requires consensus like an L1.
To understand how these pieces fit together, let's use a simple analogy of running a world-class restaurant.
An analogy: The Professional Restaurant Kitchen
Image: Description of analogy
1. What is a DA layer? (The restaurant's utility grid & food code)
A modular data availability (DA) layer provides the foundational security and data integrity for other blockchains to build upon. In our analogy, the DA layer is the city's reliable utility grid and its food safety code. It provides two fundamental things:
Consensus: A shared agreement on the rules. This is the health and safety code. Every restaurant must follow it to be considered legitimate.
Data Availability (DA): The reliable, essential services. This is the highly reliable power grid and water supply. It provides the essential services that every kitchen needs to function, and everyone has access to it, but it doesn't cook the food for you.
Crucially, the utility grid doesn't decide what's on the menu. It only provides the secure, reliable foundation for a kitchen to operate. Celestia is an example of this specialized infrastructure.
2. What is a rollup? (The professional kitchen)
A rollup is a specialized blockchain (Layer 2) built on top of the foundational DA layer. In our analogy, the rollup is the professional, high-performance kitchen. A professional kitchen is designed for one thing: executing a specific menu with extreme speed and efficiency. All the complex prep work and cooking happens inside this dedicated space.
Instead of running out to the public market for every single ingredient (which would be incredibly slow), the kitchen manages its own inventory and workflow internally. At the end of the night, it logs its inventory and expenses with the restaurant's main office (the DA layer).
This is what a rollup does: it processes its own transactions in an optimized environment, then posts the data to a DA layer for security.
3. What is a CLOB? (The michelin-star sushi counter)
A CLOB is a specific, highly demanding application. In our analogy, the CLOB is the Michelin-star sushi counter operating within the professional kitchen. This isn't just any food service; it's an art form that requires absolute precision, speed, and order. Every cut must be well-done, every order filled instantly.
Putting it all together
When we say "CLOBs are being built on rollups," this is the structure:
The application (the restaurant concept): A team wants to build a high-performance sushi counter (CLOB).
The execution layer (the kitchen): They know they can't do this in a slow, shared kitchen. So, they first build their own professional-grade sushi kitchen (the Rollup), optimized for speed and precision.
The DA layer (the utilities): They build this kitchen in a building connected to the city's reliable power grid and food code (a DA layer like Celestia) for security and verification.
This architecture is what allows for the incredible speed and efficiency needed to bring high-frequency trading on-chain.
Why CLOBs Are Being Built on Rollups
This is why rollups on a modular DA layer are emerging as the standard for high-performance DeFi; unlike monolithic chains, a modular stack allows developers to optimize for the specific needs of their application.
This approach is being validated by the number of teams building modular CLOBs. Projects like Hibachi and RISE are building directly on Celestia, while others like Bullet, Hyperliquid, MegaETH, Monad, and Kuru are exploring different approaches.
They are all chasing the same goal: CEX-level performance through specialized infrastructure.
For CLOBs, this means:
Scalable and affordable DA: Using a modular layer for data availability provides the necessary bandwidth at a fraction of the cost of an L1.
Faster finality: Rollups posting to a DA layer with fast finality (like Celestia's 6 seconds) benefit directly from that speed.
Ultra-low latency: Sequencers can offer "soft confirmations" in as little as 3-5ms, which is critical for trading where every millisecond counts.
Why it matters:
For DEXs and rollups: A modular DA model enables the kind of high-frequency, low-latency trading that CLOBs require, at a fraction of the cost of Ethereum.
For appchains: Teams can launch custom, application-specific chains with their own CLOB implementations, without having to bootstrap a validator set or worry about DA bottlenecks.
For scaling: High-throughput chains (e.g., gaming/social apps) can’t rely on Ethereum’s limited blobspace.
For DeFi as a whole: The ability to build performant, composable CLOBs on modular infrastructure could enable decentralized exchanges to reach new levels of capital efficiency and innovation - potentially rivaling the capabilities of their centralized counterparts.
Image: Modular DA benefits
The Fee Thesis
For a modular data availability layer to achieve long-term sustainability, its fee model must align with its value proposition. Taking Celestia as a prominent example, Celestia's current fee model is a strategic choice for a growing network; fees are intentionally set near zero to encourage experimentation and initial adoption. However, once the network matures and a fee structure is enabled, the economics get interesting. Here’s how value accrues:
1. The mechanics of value accrual
Celestia's fee model is expected to be driven primarily by a base fee, rather than relying solely on volatile, short-term congestion fees. However, this base fee can still increase with overall demand. As more chains post data and compete for blobspace, the underlying cost to use the network can rise due to simple supply and demand. This approach contrasts with some monolithic chains, where blockspace may be underutilized but still subsidized by users or block rewards.
More chains = more competition for blobspace = higher fees. Simple supply and demand.
Instead of relying solely on per-transaction fees, future models could be more dynamic. One possibility is a revenue-sharing model where rollups contribute a small percentage of their revenue back to the DA layer that secures them. Even with a simple fee increase, Celestia has significant headroom; it could potentially increase fees tenfold and still remain far cheaper than posting data to L1s like Ethereum or even Solana. This creates a clear path to sustainable revenue that grows alongside the success of the ecosystems building on top of it.
2. Switching costs
Once a chain deploys on a data availability layer, switching costs are high. Reconfiguring DA layers isn’t trivial, making it a long-term infrastructure choice rather than a temporary experiment.
3. The Ethereum complement (not competitor)
Any modular DA layer doesn’t need to “kill” Ethereum to win. It just needs to be the best DA layer for rollups that don’t want to pay Ethereum’s premium or need more bandwidth for high throughput applications.
Think of it like AWS vs. on-prem servers: most projects won’t run their own DA. This of course comes with it's own set of tradeoffs.
If a modular layer such as Celestia were to capture even a modest share of the rollup DA market, the resulting fee revenue could quickly reach into the hundreds of millions annually, or even billions if the network achieves high scale and fees increase significantly (e.g., tenfold). As more chains (especially those in gaming, high frequency trading, and social sectors) launch and scale, Celestia’s blobspace will increasingly become a sought-after service.
Potential Growth Trajectory
Imagine this adoption curve:
Today: 40+ teams, minimal fees.
Post-monetization: Fees start low but scale with usage.
2025+: If Celestia becomes the default DA for non-Ethereum chains, fees compound as new apps onboard.
The DA layers’ TVS are rapidly growing as more chains choose them for rollups, reflecting increasing trust and adoption.
Looking Ahead: The CLOB Opportunity
The next chapter of DeFi innovation is likely to be driven by CLOBs and other apps that need huge amounts of data availability. The modular design of DA layers are uniquely suited to meet this demand.
Risks? Sure.
Ethereum’s edge: If EIP-4844 fees drop enough, some rollups might stick with Ethereum.
Competition: Avail and EigenDA are also attracting adoption.
Execution risk: Monetization must be implemented smoothly. A botched fee model could stall momentum.
Conclusion
When a modular DA layer enables DA fees, the network’s value will be increasingly tied to real usage - chains paying for a service that underpins their operations. With CLOBs emerging as one of the most demanding and promising use cases, DA layers stand out as a foundation for building the next generation of decentralized markets.
The DA wars aren’t about ideology. They’re about cost, scalability, and developer traction. And right now, the modular approach is coming up as a strong player.