Hong Kong, June 2, 2023, ZEX PR WIRE, The ZK-PoW (Proof-of-Work) algorithm proposed by Opsides offers the following advantages:
- A market-oriented ZK computing power pricing mechanism, which can be used for scalability (ZK-Rollup) and future applications in AI (ZKML).
- It provides a huge computing power platform for the upcoming explosion of ZK-Rollup, especially zkEVM, and offers new mining scenarios for a large number of idle miners.
- The Two-Stage Commit Algorithm for Zero-Knowledge Proofs (ZKP) provides a standardized decentralized Prover mechanism for ZK-Rollup.
- The optimized ZKP calculation and sending mechanisms have improved the efficiency of ZKP generation by 80%.
Why do we need the ZK computing power PoW algorithm?
Currently, there are multiple ZK-Rollups running on the Ethereum mainnet, including Polygon zkEVM and zkSync era. However, most of these ZK-Rollup projects have not implemented a decentralized prover. For example, in the Polygon zkEVM mainnet beta, trusted aggregators are relied upon to send ZKPs, and zkSync era follows a similar approach.
While centralized provers are feasible when the number of ZK-Rollups is low, as ZK scaling technologies mature, especially the gradual implementation of zkEVM over the next year or two, the number of ZK-Rollups will experience significant growth. In the case of a huge number of ZK-Rollups, centralized provers will pose several problems:
- First, the cost of provers is high, and maintaining a centralized pool of provers requires specialized equipment and facilities. Not all ZK-Rollup operators have the ability to maintain such a centralized prover configuration. Therefore, we need professional miners to meet the computing power demand of the future massive ZK-Rollup ecosystem.
- Second, if there is only one prover, a single node failure could result in the inability to commit transactions for the entire ZK-Rollup. We need a decentralized trial mechanism to encourage multiple miners to simultaneously participate in calculating a ZKP and receive corresponding rewards.
- Finally, we need a standardized ZKP optimization algorithm to improve the overall hardware efficiency.
Opside ZK-PoW algorithm
As a highly decentralized public blockchain, Ethereum has become congested and gas tariffs have become extremely expensive. Many Web3 applications, especially financial derivatives, games, social networks and others, need to migrate to layer 2 or other public chains. In fact, providing a high performance, low gas consumption execution environment on your own is not difficult, as a few centralized solutions can easily achieve this goal. The challenge lies in maintaining a high level of decentralization while ensuring high performance and low gas tariffs.
In the Opsides design, each Web3 application can have its own dedicated ZK-Rollup and the freedom to choose a base chain. Currently, Opside supports four basic chains: Ethereum, Opside, BNB Chain and Polygon. This means that developers can choose to distribute their ZK-Rollup on any of these four public chains. To support the demand for massive hardware resources from large numbers of ZK-Rollups, Opside also provides a unified market for ZKP computing power, encouraging miners to generate ZKP for these ZK-Rollups.
Prize distribution mechanism for PoW
Opside uses a hybrid consensus of PoS and PoW. The PoS part is based on the consensus enhancement of ETH2.0. As a result, Opside will have over 100,000 validators to provide massive data availability while maintaining a high level of decentralization.
During the Pre-Alpha testnet phase, based on the PoW algorithm, each Rollup within an Opside block will send a sequence according to certain rules. PoW rewards for the current block are split across sequences based on the number of Rollup slots registered and the number of batches included. However, some rollups may not send a sequence in some blocks, resulting in lower-than-expected actual inflation.
Miners are free to choose to participate in the ZKP calculation of one or more rollups. Going forward, each sequence will be priced differently based on the corresponding ZK-Rollup type, number of Rollup transactions included, gas usage, and other factors to estimate workload.
To prevent malicious behavior by miners, they have to register and stake tokens in a special system contract. Miners are required to stake the corresponding tokens for a rollup in the system contract to be able to send ZKP for that rollup. The rewards that miners receive for sending ZKP will also be distributed according to the proportion of their stakes, thus avoiding harmful behavior from miners who send ZKP multiple times.
For more details, refer toOpside tokenomics.
The two-step commit algorithm for ZKP: standard decentralized prover mechanism
To encourage multiple miners to participate in calculating a ZKP at the same time, Opside proposes a two-step commit ZKP verification mechanism. The assignment of the PoW reward corresponding to a ZKP is distributed to the sender, i.e. the miner, of a valid ZKP on the basis of certain rules.
- Submit Proofhash: Within a specific time window for a given sequence, multiple miners can participate in calculating the zero-knowledge proof. Instead of sending the original proof directly, each miner calculates the proofhash of (proof/address) and submits it to the contract.
- Submit ZKP: After the time window, miners submit the original proof and verify it against the previously submitted proofhash. Miners whose evidence passes the verification process are eligible for PoW rewards, which are distributed proportionally based on the amount they have wagered.
For more details, refer toZKP’s two-step sending algorithm.
Optimized ZKP generation algorithm: 80% increase in miner efficiency
When a Rollups smart contract tests a ZKP, sending the original test data can potentially trigger on-chain attacks. To prevent malicious behavior, ZK rollups often require extra computational effort to obfuscate the original test data. One approach is to include an aggregation of the miners address in the submitted ZKP. Opsides Two-Step Submission Algorithm for ZKPs ingeniously adopts a submit-first-verify-later model, eliminating the need for unnecessary proof and address aggregation calculations.
Also, in some open source zkEVM implementations, ZKP calculations and dispatches are done sequentially. This means that when a ZK-Rollup sends a large number of sequences, miners cannot calculate multiple ZKPs at the same time. In Opside, the two-step submission algorithm enables parallel ZKP calculations and sequential submissions, allowing miners to perform multiple ZKP generation tasks simultaneously, significantly accelerating ZKP generation efficiency.
The Opside team also conducted a number of optimizations to the ZKP recursive aggregation algorithm, maximizing cluster resource utilization and further improving the speed of ZKP calculations.
In real-world stress test environments, miners have a cluster of 20 machines made up of 128-core CPUs and 1TB of RAM. The transaction rate tested stabilized at approximately 27.8 TPS for approximately 40 minutes. All other things being equal, Opside has reduced the average transaction confirmation time from around 5-6 minutes to around 3 minutes, increasing ZKP generation efficiency by around 80%. Going forward, as more ZK-Rollups and miners join the ZK computing power market, the efficiency improvements brought about by the Opsides PoW algorithm will become even more apparent.
Summary
Opside proposed a ZK-PoW algorithm that creatively defines a market-based pricing mechanism for ZK computing power. This computing power market provides a huge platform for the upcoming rise of ZK-Rollup, especially zkEVM, while also offering a new mining scenario for idle miners.
The two-step submission algorithm for ZKP provides a standardized decentralized Prover mechanism for ZK-Rollup, encouraging more miners to contribute stable and continuous ZKP computing power. In addition, the optimized ZKP calculation and sending mechanism has improved the efficiency of ZKP generation by 80%.
In the future, the Opsides PoW mechanism can be easily extended to other applications, not only for scalability (ZK-Rollup) but also for AI (ZKML).
#Indepth #analysis #Opsides #ZKPoW #algorithm