paint-brush
"Defi & Traditional Finance Aren't Much Different, But Defi Requires Lesser Trust" - Alex Leeby@ishanpandey
346 reads
346 reads

"Defi & Traditional Finance Aren't Much Different, But Defi Requires Lesser Trust" - Alex Lee

by Ishan PandeyMarch 28th, 2021
Read on Terminal Reader
Read this story w/o Javascript
tldt arrow

Too Long; Didn't Read

Decentralised finance- DeFi-continues to grow at an exponential rate. Total value locked of Defi protocols has exceeded $10 billion as of today, according to DeFi Pulse data, the total value locked marked $43.53 billion. ZKSpeed is a ZK-Rollup-based layer2 scalability protocol for ZKSwap. It combines practicality and engineering optimisation. ZK speed is a practical and scalable and scalable protocol. It significantly reduces the average cost of one transaction by aggregating multiple gas fees.

Companies Mentioned

Mention Thumbnail
Mention Thumbnail

Coin Mentioned

Mention Thumbnail
featured image - "Defi & Traditional Finance Aren't Much Different, But Defi Requires Lesser Trust" - Alex Lee
Ishan Pandey HackerNoon profile picture

Ishan Pandey: Hi, Welcome to our series “Behind the Startup”. Please tell us about your journey and the story behind ZKSwap?

Alex Lee: Since 2019, the blockchain industry has undergone breathtaking changes. Decentralised finance- DeFi-continues to grow at an exponential rate. The total value locked of Defi protocols has exceeded $10 billion. As of today, according to DeFi Pulse data, the total value locked marked $43.53 billion, and we believe the number is very likely to exceed $100 billion.

Of course, these on-chain assets would require fast, frictionless, trust-free and real-time exchange services, which has led to a rise of new decentralised exchange protocols such as Uniswap.

Although the new DEX model spearheaded by Uniswap has achieved significant development, it still has obvious drawbacks. First of all, the high gas fee could potentially repel new users from entering. Secondly, every transaction and operation requires block confirmation, which leads to transactional delays. Thirdly, Uniswap has a clear bottleneck in the number of transactions and transaction capacity per second. Those drawbacks are not unique to Uniswaps. They are common issues that exist among all DEXes.

There is a need in the market for a new product that can effectively reduce gas fees with high-speed processing and automated market-making functionality. We believe that the emergence of such a product will effectively fill the market gap.

Ishan Pandey: Please explain how ZK-rollup technology works? Further, I have read that transactions on the automated market maker (AMM) based token swap protocol does not require Gas fees. How is this possible? (Please give a technical explanation)

Alex Lee: ZKSpeed is a ZK-Roll Ups-based layer2 scalability protocol for ZKSwap. It combines practicality and engineering optimisation. It is a practical and scalable ZK-Rollups protocol. ZKSpeed achieves high layer2 TPS and significantly reduces gas fees by aggregating multiple zero-knowledge proofs, parallel processing of the PLONK algorithm, and categorising off-chain data. All data for transactions interacting with Layer1 are stored on-chain (ZK-Rollup), and data for pure Layer2 transactions are stored off-chain (Validium), and transaction hash data are stored on-chain.

Ishan Pandey: What is a Layer-2 server? Further, please explain ZKSwap’s zero-knowledge proof system and how the PLONK algorithm works?

Alex Lee: 1. ZKSpeed adopts the PLONK zero-knowledge algorithm, where all transaction types share a common trusted setup, thus eliminating the need for packaging blocks by transaction type.

2. ZKSpeed uses a GPU-implemented PLONK algorithm which runs three times faster than the ordinary CPU-implemented version. Besides, with top-notch hardware equipment, the solution significantly shortens proof generation time and improves system throughput.

3. After thorough research, the ZKSwap team adopted the Aggregation proof for ZKSpeed, which was the first to implement it in an AMM model DEX. (The Aggregation proof was first proposed by Matter Labs and was used in simpler applications). Aggregation proof gathers proofs of multiple blocks and aggregates them into a single proof, making it possible to validate multiple blocks on-chain at once. Thus, it significantly reduces the average cost of one transaction.

Ishan Pandey: Please explain how ZK proof is calculated and why it uses computational power similar to mining?

Alex Lee: Zero-Knowledge Proof guarantees the security of the network. We apply 100 AMD 7542 servers and 200 2080Ti cards to calculate the zk proof, so we can always keep Layer2 and Layer1 aligned.

Ishan Pandey: Please explain what proof-of-liquidity is and how does it work?

Alex Lee: Liquidity mining, also known as yield farming, is the act of providing liquidity via cryptocurrencies to decentralised exchanges (DEXs). Since an exchange’s primary goal is to be liquid, DEXs seek to reward users willing to bring capital to their platform.

Ishan Pandey: How did the ZKSwap team develop the fastest GPU version of the zero-knowledge proof system?

Alex Lee: We use over 100 high-performance servers and 200 pieces 2080Ti GPU Cards for the zk proving system, and as far as we know, it’s the fastest zk proving system in the market.

Ishan Pandey: What advice and tips will you give a smart contract developer on how to code a complex smart contract protocol?

Alex Lee: It is recommended to refer more to the audited contracts of the launched mainnet. Try to avoid re-creating your own theory. Try to reduce the risk of complex contracts by modularising the attack area (e.g. Uniswap split the function into core and router. core code is dedicated to the core function to facilitate auditing, router implements business logic that does not involve security)

Ishan Pandey: What is the future of DeFi and what next major innovation are we going to see in the Layer2 ecosystem?

Alex Lee: Defi and traditional finance are not fundamentally different, but Defi lowers the threshold of trust and improves resource allocation efficiency in a sense. I am looking forward to a trust-free and scalable programmability of layer2 for future innovation, such as integrating optimistic rollup and ZK Rollup is a direction worth exploring.