Foresight Ventures: What is RaaS? What kind of RaaS will win the market?

Author: Yihan Xu, Foresight Ventures

Overview

• Through this article, you can learn about:
• What RaaS is;
• Analysis of RaaS technology paths;
• Analysis of RaaS projects on the market;
• The value of RaaS;
• The ultimate form of RaaS as I understand it…

1. What is RaaS

1.1 Rollup: The Most Promising Scaling Solution

The original intention of Layer 2 was to alleviate the congestion problem of the main network, providing services for Dapps at a lower cost and higher TPS while ensuring security. Rollup moves the high-cost transaction execution to L2 and packages transactions for verification on L1, while ensuring that the complete transaction content can be verified. Under the premise of inheriting Ethereum's security, it possesses stronger overall performance. Therefore, Rollup has carved out a path among various layer 2 solutions and is undoubtedly the most promising off-chain scaling solution at present.

1.2 Rollup-as-a-Service: One Form of App-specific Chain

As some DApps gradually grow and various new applications expand, the general-purpose scaling of Rollup clearly cannot perfectly meet these projects' pursuit of user experience and cost structure. The massive traffic and extremely high performance requirements (such as AAA games that focus on player interaction) necessitate more customized scaling solutions.

App-specific chains are currently one of the best solutions for these DApps.

The concept of app chains is not unfamiliar; different projects can customize blockchain design based on their application scenarios and needs, allowing Dapps to enjoy exclusive resources on a chain while obtaining lower operational costs and higher performance without severing ties with other ecosystems, thus providing users with a better experience.

For example, the Cosmos based on Tendermint consensus provides Dapps with a low-cost environment to build a sovereign L1 public chain. At the same time, based on the IBC communication protocol, different app chains can also more easily achieve cross-chain asset/information transfer. You can refer to the IBC packet lifecycle provided by Cosmos?

Discussing scaling without considering the ecosystem is meaningless.

The feasibility of solutions like app chains must be based on strong interoperability and ecosystem support. For instance, Cosmos gradually improves its ecosystem through the cross-chain advantages brought by L1 sovereign public chains and IBC.

Based on this understanding, another approach to app-specific chains is to achieve the pursuit of customized features, high performance, and low costs through customized Rollups. Some RaaS based on layer 2 networks can also make project interactions faster, positively impacting ecosystem layout.

2. The Value of RaaS

The multi-chain and multi-Rollup landscape in the crypto world seems to be an inevitable trend, and the emergence of RaaS projects has created a foundation for exploring new DApp forms. However, under this consensus, I still want to ask a realistic question in the opposite direction:

Allowing anyone to quickly launch a Rollup is indeed attractive, but aside from being broadly correct and cool, does it really create enough value for those in need? This question can be further broken down into two points:

• Are there enough projects in the market that have sufficient motivation to use RaaS?
• Does RaaS create considerable value for project parties?

This question essentially discusses the demand and the value brought by RaaS. There must be enough project parties with demand, or RaaS must provide attractive enhancements.

From the perspective of demand, as some Dapps continue to grow, project parties indeed urgently need to seek:

• Lower costs
• Higher performance
• Special features

Cost

According to data provided by L2fees, L2 Rollup has achieved extreme optimization in cost, showing a significant improvement compared to the Ethereum main network. At the same time, looking at the RaaS test data from Caldera Chains, it can be found that there is no qualitative change in cost, more like a 99-100 optimization. Additionally, the implementation of EIP4844 and danksharding will further reduce the costs of L2 Rollup, and the differences brought by RaaS in terms of cost and efficiency will also shrink.

A solution that can significantly reduce transaction fees is attractive, but most RaaS cannot achieve this. Considering factors such as migration, overall ecosystem, interoperability, and security, do project parties really have enough motivation to use RaaS? For most conventional Dapps or users who are not very sensitive to performance and cost, perhaps general-purpose scaling is sufficient.

Performance

L2 Rollup already has the capability to provide ultra-high TPS. According to data provided by Caldera, the RaaS based on Op has almost no advantages in block time. Although ZK RaaS can provide more customized data storage and compression, there is not much demand for such services. RaaS based on layer 2 networks can indeed bring faster speeds and lower costs by settling transactions on L2, thereby enhancing user experience.

As mentioned above, given the imperfect ecosystem and other migration/development costs, do project parties still have enough motivation to use RaaS?

Customized Features

From the perspective of value creation, some RaaS can indeed provide functionalities that are currently difficult to achieve or designs that are relatively inefficient in general-purpose scaling. For example:

• The primary element of L2's ZK circuit design is compatibility, which sacrifices efficiency to serve all Dapps and is not optimized for specific Dapps. The value of RaaS can be clearly demonstrated: customizing ZK circuits for specific Dapps or providing more efficient storage structures and data compression services to achieve higher performance;
• Implementation of privacy features. Although ZKrollup is friendly to privacy, due to decentralization and security considerations, users' transaction data still needs to be published to L1 as a history log after compression, allowing all users to verify. Therefore, the current general-purpose scaling rollup cannot achieve privacy. RaaS can customize privacy features on the basis of rollup or even rollup of rollup, creating value for projects with strong privacy demands.

Thus, the current value of RaaS lies in customization > mere cost and efficiency. (This does not exclude cost and efficiency improvements brought by customization.)

To answer the initial question: Does RaaS really create enough value for those in need?

I believe that the current demand for RaaS is limited, and general-purpose scaling can meet over 90% of the needs. Although customized rollups have begun to play an irreplaceable role in some niche areas, they are not mainstream. The value created by RaaS is limited and needs further exploration based on comprehensive factors such as ecosystem and interoperability.

3. Exploring the Ultimate Form of RaaS

Since the emergence of L2 Rollup, the exploration of RaaS has never stopped, and various Rollup-as-a-Service implementations have appeared on the market. Referring to the ecosystem layout on Messari, we can roughly see the different implementation paths of RaaS. The most critical questions are:

• What solutions make sense?
• What kind of RaaS will ultimately win the market?

3.1 OP or ZK?

The discussion about optimistic versus zero-knowledge has never ceased. Although ZKrollup theoretically has stronger performance, much faster finality time than optimistic rollup, and higher security, optimistic rollup has better compatibility and lower thresholds.

Among the existing RaaS projects, most are based on optimistic rollup, and I believe the main reasons are as follows:

• Ecosystem always comes first. RaaS based on optimistic rollup has better compatibility, greatly reducing the migration/development threshold for project parties, allowing more project parties to deploy quickly and rapidly build a more prosperous ecosystem, gaining first-mover advantage;
• Lower thresholds, not relying on computational power support. RaaS based on optimistic also verifies the validity of transactions through fraud proofs, thus requiring lower machine performance and reserves. This is also one of the limiting factors for many RaaS to not adopt ZK from the start.
• Easier to scale. RaaS based on optimistic has a lower development threshold, unlike ZK RaaS, which pursues performance and deeper customization, requiring providers to deeply participate in development. At the same time, due to the computational power required to generate ZKP, ZK RaaS is also difficult to deploy on a large scale like optimistic RaaS.

Although optimistic rollup has obvious advantages in ecosystem layout, ZK-based RaaS also has significant strengths.

• True customization brings better performance and lower costs. In the customized design of rollup, ZK-based RaaS can bring greater value in functionality and performance, which is difficult for general-purpose scaling to achieve, and can be seen as a change from 0 to 1. In contrast, optimistic RaaS mainly makes changes from 90 to 99 in terms of cost and efficiency.
• Higher security. ZK RaaS can be trustless, while op-based services require trust in the challenger to function properly and prevent sequencer malfeasance.
• Better interoperability and finality time. OP-based RaaS requires a 7-day fraud-proof verification, while ZK's trustless characteristics allow for faster finality time, and the 7-day verification period poses challenges for OP-RaaS in cross-rollup construction.

Conclusion

In the short term, the advantages of optimistic RaaS in the ecosystem cannot be shaken, but from the perspective of long-term demand and value creation, I believe that ZK-based RaaS will certainly gain a larger market in the future.

3.2 Layer 2 or Layer 3?

Choose a more suitable implementation solution based on different RaaS use cases and implementation goals. The biggest difference regarding this question, in my opinion, lies in cost and user experience (interoperability).

If L2 is regarded as the settlement layer and RaaS is laid out as L3, it can achieve lower transaction costs and faster cross-rollup interactions, thus providing a better experience. Although Ethereum's L2 RaaS inherits the security of the main network well, it falls short in cross-chain costs and speeds compared to multi-layer network designs.

Therefore, L3 > L2

• For an introduction to Layer 3, you can refer to an article I wrote earlier???

Foresight Ventures: In-depth Interpretation of Layer3

3.3 RaaS or L1 App Chain: The Trade-off Between Ecosystem and Cost

Cosmos and Polkadot were the first to propose the app-specific chain solution. Which is more suitable for providing customized services for DApps, RaaS or L1 app chains?

Interoperability

• For L1 app chains, besides the Cosmos ecosystem based on the IBC communication protocol mentioned in the first section, applications can establish parachains on Polkadot and perform cross-chain information interaction based on XCM. However, due to security and cost considerations, in practical applications, we can see that most projects still develop customized L1 app chains based on Tendermint or Substrate consensus engines, with very few utilizing cross-chain communication functions. This leads to relative independence between these cross-chain ecosystems, which does not align with my vision of the ultimate form of app chains, where different app chains should together form a prosperous ecosystem with strong interoperability.
• For RaaS structures like Starknet that further extend based on layer 2 networks, they have stronger advantages in interoperability, allowing low-cost cross-chain interactions between different Dapps' maintained rollups, and since transactions can be settled on layer 2, the speed and user experience will also be better. However, the premise of all this interoperability is that RaaS can build a sufficiently strong ecosystem.

Security

It depends on the design of RaaS. DA based on Ethereum's RaaS mostly inherits the same security as Ethereum L1, which is higher than that of L1 app chains in terms of security and decentralization. RaaS based on DA layers or side chains has its security guaranteed by these layer 2 networks.

Cost

For L1 app chains, transaction costs converge to the native token of the Dapp project itself, allowing for extremely low operational costs;

For RaaS, L2 RaaS, due to the need to directly interact with the Ethereum main network, has relatively higher costs, while L3 RaaS based on Polygon or Starknet can settle on L2, resulting in relatively lower costs.

4. RaaS Project Analysis: Who Will Win the Market

There are currently many RaaS projects being developed or already deployed, including but not limited to StarkNet L3, Opside, Caldera, Celestia, Dymension, Sovereign, Stackr, Eclipse, Altlayer, Saga…

Below are some representative projects for analysis.

4.1 ZK Series

Including but not limited to Sovereign Labs, Fractal, StarkNet, Opside, ZKsync

4.1.1 StarkWare: Customized L3 Based on ZKRollup

Referring to this old image, the StarkWare team first proposed the design of Ethereum's multi-layer network in the article "Fractal Scaling: From L2 to L3." However, the proposal of multi-layer networks is not merely for further scaling; it allows project parties to control more on-chain resources through the overlay of customized Rollups based on L2 general-purpose scaling, providing user experiences that L2 rollups cannot achieve.

Although from a computational perspective, one can generate a ZKP to prove validity for a pile of ZKPs, data cannot be compressed further after being compressed. Since it is necessary to ensure data availability and allow anyone to verify the validity of the proof, rollups need to send the complete or compressed transaction content to L1.

Therefore, the application scenarios of StarkWare's app-specific chains must pursue high performance or specific features.

• High performance: Games with high performance requirements can exclusively utilize ZK circuit resources to provide a better user experience;
• Privacy: For some projects with privacy needs, implementing privacy features on a customized basis in rollups or even rollups of rollups;
• Compatibility scaling: Providing an EVM-compatible environment, even compatible with more programming languages, providing positive value for the ecosystem itself;
• Low cost: By sacrificing a certain degree of decentralization and security, operational costs can be significantly reduced through Validium.

The L3 solution based on Validium from StarkNet can theoretically reduce costs intuitively, while interoperability is also guaranteed.

However, from the perspective of customization, it can be further inferred that this ZKrollup-based app-specific chain, while providing considerable performance improvements, also raises the development costs and participation thresholds for project parties. Therefore, RaaS providers need to deeply engage in development, and the scalability and speed of commercialization have limitations.

4.1.2 Opside: Another Three-layer Network Structure Designed for App-specific Chains

Referring to the diagram below, compared to StarkWare, ZKsync's app-specific L3 based on L2 rollup, Opside proposes a three-layer network designed specifically for high TPS applications. A side chain is designed as L2 based on PoS+PoW consensus, and the app-specific chain connects to the side chain as L3.

Opside conducts data interaction through the developed ZK-bridge, and unlike traditional side chains, the proof of legitimacy is completed through ZKP rather than multi-signature, thus ensuring higher security. At the same time, Opside integrates app-specific rollups into the consensus of the L2 sidechain through native rollup, incentivizing third parties to maintain the rollups on the L2 sidechain from a consensus perspective.

Interoperability is crucial for RaaS. The native rollups in Opside share a world state tree and global message queue. Therefore, asset and information interactions between app-specific rollups will be very efficient and cost-effective, and cross-chain asset interactions only require directly invoking the target rollup's contract method in a single L3 rollup contract. However, compatibility and ecosystem development remain challenges for ZK-based rollups.

The tradeoff brought by ZK's trustless and faster finality time limits the commercialization scale of RaaS due to the need for hardware support to generate ZKP, which is also one of the reasons why most RaaS do not adopt ZK. Additionally, designing side chains as L2 poses a challenge for the security of RaaS providers.

4.2 Optimistic Series

Including but not limited to Caldera, Eclipse

4.2.1 Caldera: Maximizing User Experience Based on Op Stack

Caldera is a RaaS based on Op stack, providing project parties with high throughput, low latency, and customized features of L2 rollup. Currently, the testnet allows anyone to create an L2 rollup in a very short time, and the user experience is very smooth; you can try it out: https://dashboard.caldera.xyz/

The design based on Op stack gives Caldera a significant advantage in compatibility, with fully EVM compatibility and the team's optimization of user experience greatly lowering the migration/development threshold. At the same time, Caldera's RaaS is not limited by the support of underlying hardware in terms of computational power, allowing more project parties to deploy quickly and build a more prosperous ecosystem.

Referring to the structural diagram in Caldera's official documentation, Caldera Chains can not only launch L2 rollup-as-a-service on Ethereum but can also provide services on any EVM-compatible L1. By sending fraud proofs to L1, it ensures transaction validity. In terms of data availability layers, Caldera has also innovated by decoupling the Data Availability Layer from the Settlement Layer, allowing customized rollups to send transaction content to Ethereum or dedicated DA layers, such as Eigenlayer or Celestia. This design optimizes Caldera's scalability and transaction costs to a greater extent.

The interoperability of the Caldera ecosystem is achieved through internal cross-chain bridges, enabling asset and data cross-chain interactions by deploying contracts on the corresponding L1 and app-specific rollups. At the same time, Caldera also provides a high-level JavaScript SDK to help developers more efficiently integrate cross-chain functionality into customized rollups.

Although Caldera has done a lot in terms of interoperability and cross-chain bridges, optimistic rollups require a 7-day fraud-proof time, making the construction of interoperability between rollups a challenge. Additionally, RaaS based on optimistic cannot achieve trustlessness, as it requires trusting at least one challenger to prevent sequencer malfeasance.

Moreover, in terms of customization, Caldera and similar optimistic RaaS focus more on low costs and high TPS, making it difficult to provide greater value in functionality and performance compared to ZK-based RaaS. It can be seen that the current general-purpose scaling rollups have already achieved considerable block time, TPS, and transaction costs, and the data shows that the differences with RaaS are not significant, suggesting that they have completed a 0-1 improvement. Therefore, whether the cost and throughput improvements brought by Op-based RaaS are what the current market needs remains a question.

4.3 Modular Blockchain

Including but not limited to Celestia, Dymension

4.3.1 Celestia: Building Modular Blockchains Based on DA Layer

Celestia is essentially a data availability layer. The DA layer based on Tendermint consensus constructs a highly scalable blockchain architecture. Through rollmint (an application blockchain interface implementation), Dapps can build their own rollups and deploy them to Celestia, with data stored in the DA layer, while the state root and proof are uploaded to L1 for verification. Celestia optimizes the DA layer through data availability sampling (DAS), where each light node in the network only needs to sample and download a small portion of block data. Therefore, the more nodes there are, the more transactions can be included in each block, achieving the purpose of scaling the DA layer.

This easily brings to mind familiar Validiums: a scaling solution that verifies computational results through ZK algorithms, where data is not uploaded to L1 but relies on validators for hosting. Since data exists off-chain rather than being directly published to Layer 1, Validium reduces gas costs. However, from the perspective of decentralization and security, data availability relies on a third-party committee, so the use of Validiums is not widespread.

From the implementation perspective, the Dapps in the entire ecosystem are essentially building their own Validium and maintaining sequencers and provers, with Celestia providing unified data storage space. Similar to Validiums, this implementation reduces the operational costs of Dapps but sacrifices a certain degree of decentralization and security. Compared to other solutions that inherit Ethereum's security, the security of Dapp chains on Celestia relies on nodes and the DA layer.

Additionally, Celestia currently does not support fraud proofs, so nodes need to re-execute all transactions based on pessimistic assumptions to ensure validity. At the same time, rollmint only supports a single sequencer, so there is significant room for improvement in efficiency and decentralization.

However, as a DA layer, the potential of Celestia is not limited to this; for example, the Eclipse solution based on optimistic RaaS uses Celestia as the consensus and DA layer.

5. Summary and Outlook

RaaS can intuitively bring improvements in cost and performance, but from the perspective of performance, these optimizations do not possess strong attractiveness; greater value still needs to be tied to customized features. Currently, market demand is limited, but with the future development of crypto, greater traffic will lead to a linear increase in Dapps' pursuit of low costs and high performance, and customized rollup services are clearly a desirable solution.

To answer the initial questions, what do I understand as the ultimate form of RaaS? What kind of RaaS will capture the market?

From the product itself

The advantage of OP-based RaaS lies in quickly building ecosystems and forming barriers, but the slight improvements brought purely by cost and efficiency are not enough to attract project parties, thus lacking long-term value. In contrast, ZK-based RaaS can address pain points in customized features, but demand is still not mainstream.

The design of multi-layer network structures allows L3 RaaS to achieve lower costs and stronger interoperability. Strong interoperability is the foundation for building a prosperous RaaS ecosystem, so the multi-layer network design based on ZK can combine the advantages of customization and low costs, indicating longer-term value.

I believe that in the long term, ZK-based multi-layer network RaaS will become the ultimate choice in the market.

Market and Demand

RaaS with sufficient scalability can meet the customized rollup needs of all project parties while ensuring performance. Moreover, the true rise of RaaS heavily relies on ecosystem construction. Therefore, a scenario where multiple RaaS coexist clearly does not make sense.

I believe that the endgame will definitely be one or a few RaaS occupying the entire market.

6. Reference

https://ethresear.ch/t/rollup-as-a-service-opportunities-and-challenges/13051

https://ibcprotocol.org/

https://messari.io/report/the-rollups-as-a-service-ecosystem