What Layer-2 Rollups Mean for CBDC Trading

Table of Links Abstract and 1. Introduction Abstract and 1. Introduction Background 2.1 Layer-2 Blockchains 2.2 Project Mariana Related Work and Contribution System Architecture Model 5.1 Automated Market Makers 5.2 Price Impact Simulation 6.1 Data and Parameters 6.2 Results Discussion Conclusions, Acknowledgements, and References Background 2.1 Layer-2 Blockchains 2.2 Project Mariana Background 2.1 Layer-2 Blockchains 2.1 Layer-2 Blockchains 2.2 Project Mariana 2.2 Project Mariana Related Work and Contribution Related Work and Contribution Related Work and Contribution System Architecture System Architecture System Architecture Model 5.1 Automated Market Makers 5.2 Price Impact Model Model 5.1 Automated Market Makers 5.1 Automated Market Makers 5.2 Price Impact 5.2 Price Impact Simulation 6.1 Data and Parameters 6.2 Results Simulation 6.1 Data and Parameters 6.1 Data and Parameters 6.2 Results 6.2 Results Discussion Discussion Discussion Conclusions, Acknowledgements, and References Conclusions, Acknowledgements, and References Conclusions, Acknowledgements, and References Graphs and Tables Graphs and Tables 2 Background This section introduces the concept of L2 blockchain scaling, presents the functionalities of Project Mariana, and summarizes this work’s contributions to the research on cross-border CBDCs exchange. 2.1 Layer-2 Blockchains There are two approaches to tackle the challenges of blockchain scalability and related high gas prices: Layer-1 (L1) and Layer-2 (L2) blockchains. L1 scaling involves creating entirely new blockchains, e.g., with unique consensus mechanisms or block sizes. These chains operate with their own validators and infrastructure, requiring decentralization to win users’ trust. L2 scaling takes a different approach and uses complex computations off-chain (outside of Ethereum), aiming to reduce on-chain data congestion. These off-chain activities include rollups, sidechains, plasma, and state channels. Layer-2 blockchains rely on the security of underlying Layer-1 chains, such as Ethereum, for final settlement. Rollups represent a form of L2 scaling that does not custody any data by themselves. Rollups offload complex calculations from the Ethereum mainnet and store the results (along with other transactions) in Ethereum after compressing them. Fig. 1 illustrates rollup architecture with key components—sequencers and verifiers. Sequencers roll up transactions to the Layer-1 chain. By bundling transactions, rollups manage to save on gas fees. Verifiers are smart contracts that operate on Ethereum and verify the transactions stored by the sequencer. They ensure the correctness of the transactions. 2.2 Project Mariana The specific focus of this work is on facilitating cross-border exchanges of CBDCs. One notable proposal for such a solution is Project Mariana on L1 blockchain (L1-Mariana) [4], initiated by BIS, the Swiss National Bank, the Bank of France, and the Monetary Authority of Singapore. Within L1-Mariana, a DEX based on AMM is employed to exchange wholesale CBDCs versions of Swiss Franc (CHF), Euro (EUR), and Singaporean Dollar (SGD). The selected AMM is the Stableswap Invariant, first introduced by Curve v2—the second largest DEX in terms of trading volumes in permissionless DeFi on Ethereum [1]. The setup involves bridges that facilitate the transition of wholesale CBDCs from domestic central bank blockchains to the chain where the Cryptoswap Invariant pool operates, as depicted in Fig. 2. Whereas the project proved the feasibility of AMM-DEX for cross-border CBDCs exchange, several questions arise—e.g., the choice of the blockchain (and its scalability and security), the AMM and the pool types (3-token pools vs 2- token pool)—which this research addresses. Authors: (1) Krzysztof Gogol, University of Zurich (gogol@ifi.uzh.ch); (2) Johnnatan Messias, Matter Labs; (3) Malte Schlosser, University of Zurich; (4) Benjamin Kraner, University of Zurich; (5) Claudio Tessone, University of Zurich. Authors: Authors: (1) Krzysztof Gogol, University of Zurich (gogol@ifi.uzh.ch); (2) Johnnatan Messias, Matter Labs; (3) Malte Schlosser, University of Zurich; (4) Benjamin Kraner, University of Zurich; (5) Claudio Tessone, University of Zurich. This paper is available on arxiv under CC BY 4.0 DEED license. This paper is available on arxiv under CC BY 4.0 DEED license. available on arxiv available on arxiv