Investing in Radius
Radius, Expanded
When we first met Radius in early 2022, there were no great schemes like PVDE or buzzwords in the market like Shared Sequencing.
However, Radius’ adamant insistence that a single sequencer has limitations that must be addressed, and that the solution must be a way to go beyond mitigation to “completely” eliminate them, gave us a glimpse of where the market was headed.
In March 2022, we became the first investor of Radius, alongside DSRV. As investors who have followed Radius closely and for the longest time, we write about why we think Radius can do even better in the future.
Limitations of Single Sequencer
Blockchain is a cool technology because no one can change the order and content of data already recorded on the blockchain, anyone can verify information, and it’s transparent. But if these values are compromised by a centralized component, blockchain is no longer cool. The centralization of sequencers, which were introduced to make blockchain scalable, is threatening the beauty of blockchain.
Sequencers perform very important functions in L2 architectures, such as selecting transactions to be contained in blocks in the rollup chain, determining the order of processing, computing state in the determined order, and sending state values to Layer 1. Currently, most rollups have two common characteristics: (1) they have their own system each, and (2) they adopt a single sequencer system. This leads to several limitations.
Limitation #1. Siloed
In most L2 architectures today, rollups work in their own silos. As each rollup has its own sequencer, prover (for ZK rollups), and execution environment, the liquidity and user pool is fragmented. This forces users active in more than one rollup to incur the cost and complexity of cross-sequencer bridges for interoperability. Dapps face a fragmentation problem.
Limitation #2. Exacerbated MEV
If the sequencing layer is centralized, PBS (Proposer-Builder Separation. A concept designed to prevent monopolization and centralization of MEV revenue by introducing the role of ‘Builder’ into the block creation and ordering process) is not effective. A sequencer controlled by a single node becomes a de facto monopolist, which can leverage information about users’ transactions to maximize its profits.
Limitation #3. SPOF (Single Point of Failure)
In most current rollup architectures, the sequencer is the single point of failure. Since the sequencer is the only point at which transactions are selected, ordered, computed, and sent to the L1 chain, rollups cannot function properly when the sequencer fails. The current architecture, where the single point of failure is even a “single node,” is even more vulnerable.
Of course, rollups already recognize these limitations and are working to address them in various ways. For example, Arbitrum, Optimism, zkSync have announced plans to mitigate risks by decentralizing their sequencers in the future. However, Radius argued that decentralizing sequencers alone does not completely eliminate potential risks and that a more robust solution is needed.
Enter Radius
Radius is a shared sequencing layer designed to eliminate harmful MEVs and censorship while creating economic value for rollups. Radius implements an encrypted mempool via PVDE, a ZK-based encryption scheme. PVDE prevents a centralized sequencer from front-running, sandwiching, and censoring user transactions.
Radius provides a solution to each of the limitations of rollups with a single sequencer.
Solution #1. Siloed → Shared Sequencing Scheme
The sequencing layer is a modular component of the blockchain that focuses on ordering transactions without executing them. With the introduction of the sequencing layer position, rollups focus on “execution” and the sequencing layer is responsible for ordering and executing transactions.
There can be multiple rollups that receive blocks from the sequencing layer. If the rollups use the same sequencing layer, interoperability is greatly improved by making it easier to synchronize data between rollups.
There can be multiple rollups that receive blocks from the sequencing layer. If the rollups use the same sequencing layer, interoperability is greatly improved by making it easier to synchronize data between rollups.
However, Radius argues that decentralizing the sequencer can mitigate the risk, but not eliminate it. As a more complete solution, they came up with the concept of hiding the contents of a transaction until they decide on the order in which to process it, and then automatically revealing them, which they implemented in a scheme called Practical Verifiable Delay Encryption (PVDE) and published on Ethresear.ch.
Solution #2. Exacerbated MEV → PVDE
Rollups face harmful Maximum Extractable Value (MEV) and censorship issues when they rely on centralized sequencers. This not only directly and indirectly results in financial losses for users, but also undermines trust in the ecosystem.
Decentralizing sequencers can mitigate some of these risks, but at the cost of reduced scalability and increased operational costs. Also, even if decentralized, the contents of transactions are still visible to the sequencer, which means that the sequencer can reorder transactions to extract profits.
Radius’ goal is to eliminate the possibility of these potential risks ever occurring in the first place.
Radius does not completely eliminate this potential risk by decentralizing the sequencer, but by introducing a cryptographic solution that makes it impossible for the sequencer to verify the contents of transactions in advance when putting them into blocks and determining their order.
One scheme Radius has developed to accomplish this is Practical Verifiable Delay Encryption (PVDE), which implements an encrypted mempool. Radius has also been recognized for its use of PVDE with an Ethereum grant
Solution #3. SPOF → Distributed Sequencer Network
While cryptographic processing via PVDE ensures the unreliability of the sequencer, the sequencer still exists as a single point of failure (SPOF). To address this issue, Radius proposes a distributed sequencer network model in which multiple sequencers operate simultaneously. This allows the blockchain network to function normally even if a particular sequencer fails because the remaining sequencers can take over its role. Radius is investigating various methodologies for a rogue distributed sequencer network, including private single leader elections, block building with multiple sequencers, and sequential ordering.
Team
I remember the first time I met with co-founders AJ, Jayden, Jongbeen, and Hyun at Seoul WeWork in March 2022. Over the course of several meetings, I was able to hear about the problems and solutions that Radius was trying to solve, and there was a phrase that I heard in every meeting:
“It’s not good enough.”
I think this sentence best describes the team Radius. Radius developed the PVDE scheme because they felt that decentralizing a single sequencer wasn’t enough, and they weren’t satisfied with what they had accomplished, so they pushed harder, and in less than two years of building their team, they have accomplished a lot.
AJ and Hyun are expanding the zero-knowledge proof research community in Korea by founding ZK-SEL, the largest zero-knowledge proof technology research society in Asia, which has grown to nearly 800 members as of July 2023.
Co-founders AJ, Jayden, Hyun, and Jongbeen have a deep understanding of the industry and market opportunities. AJ, the CEO, has 12 years of experience as a software engineer at Samsung Research, five of which were spent developing security technologies, gaining expertise in cryptography. Jayden, COO, worked as a security engineer at Samsung Research for 7 years and led the development of Samsung’s security management system. Jongbeen, CTO, has a PhD in distributed systems from Seoul National University, three patents and four papers on blockchain, and has developed a new consensus engine and MPC-based multi-sig wallet for Tendermint. Hyun, CRO, has been conducting professional-level blockchain research for more than 7 years and leads the implementation of cryptographic properties related to zero-knowledge proofs into the product.
About nonce Classic
nonce Classic is a VC/accelerator borne out of Seoul’s largest Web3 hacker house. Backed by a strong alliance of builders and angels, nonce Classic lays out a bedrock for founders by providing a strong community of like-minded people
Website: https://nonceclassic.org/main
Twitter: https://twitter.com/nonceclassic
Disclaimer: This article is for general information purposes only and is not investment advice or solicitation. You should consult a professional for investment decisions or advice. The information contained herein has been obtained from other sources, but nonce Classic has not independently verified it and makes no warranty as to its accuracy or suitability. This article reflects the opinions of the author and does not necessarily reflect the opinions of others associated with nonce Classic. Opinions are subject to change at any time without notice.
Copyright 2023. nonce Classic. All Rights Reserved
