As of now, the number of validatorsNodes on the Aleo Mainnet has increased to 25, and the TPS has increased from the original 30 to 50. So, what level is TPS at 50? Before that, let's first clarify what TPS is and what it includes in traditional public blockchains. TPS is the abbreviation for Transactions Per Second, which means the number of transactions processed per second. It is an important metric for measuring the performance of the Blockchain network, indicating how many transactions the network can handle in a unit of time. In traditional public blockchains, a transaction refers to an operation through which users can initiate Token transfers or execute Smart Contracts. There are two main state models used in traditional public blockchains—UTXO (Unspent Transaction Output) and the account model (introduced by the Ethereum network). Aleo uses a variant of the UTXO model called the Record model. While the account model is more intuitive for developers, it indexes the global state using account Addresses. This means that although the private account model can achieve privacy in inputs and outputs, the lack of encryption for account Addresses still compromises user privacy. Another issue with the private account model is the lack of concurrency, as the entire program state must be used to update it (only one user can perform this operation at a time). The Record model in Aleo uses Program IDs as clustering points (instead of account Addresses) to enhance privacy and provide programs with internal states, which is more efficient and solves the concurrency problem. Records are fundamental data structures that can contain any valid payload used to encode user assets or application states. Records represent some state of a program in the global state, such as account balances or identification documents. The main difference from Zcash's note model is that records also include information about which program they are associated with (Program ID). This opens the door to more complex logic and allows for encryption of any payload (assets, application states, etc.), not just simple value transfers. Each record contains information about its owner, the stored value, and its application state: account Public Key #最近关注的新币 APK(: specifying the owner of the record and who has the right to use it; small points: Aleo balance stored in the program record; data: data payload containing any information dependent on the application, such as account balance; random number: a unique serial number for each record similar to the Nullifier in Zcash. By checking if the nonce is already included in the spent record set, it prevents the reuse of the same record, an important feature to prevent double spending; visibility: by default, private in Aleo, but can also be set to public if users wish to share information or engage in public transactions.

Records in Aleo are used and created through transition functions. A transaction can contain multiple transitions, each responsible for the use and creation of its own records. This allows for multiple state updates in a single transaction. Each transaction can contain up to 32 transitions, with one used for transaction fees. Transition records can be seen in transaction records on the browser: since each transaction can support up to 32 transitions, with one used for transaction fees, Aleo's TPS should be multiplied by 32. With the current 25 validatorsNode, each Node averages 2-5 TPS. In fact, the TPS can reach a level of 1500+. The number of transitions supported per transaction depends on the settings of the Aleo Mainnet. Currently, the focus is on increasing the number of validatorsNodes and optimizing network performance, with plans to increase to 40. By traditional blockchain TPS calculations, increasing TPS to around 100 would essentially reach the current level of ETH L1. Adjusting the number of transitions supported per transaction to 32 transitions, the TPS can actually reach 3000+. What does 1500+ or 3000+ TPS look like? Let's compare with other public chains: ETH 2.0: After the PoS upgrade, it is expected to support 100-200 TPS, but actual throughput depends on network scale and other technical improvements (such as Sharding). Solana: Using PoS + Proof of History (PoH) consensus mechanism, it supports 5000-6000 TPS, making it one of the strongest in terms of throughput among PoS public chains. Avalanche: C-Chain can handle 4500 TPS. Sui: According to Sui Foundation data, Sui's highest TPS can reach 297,000, with the current highest TPS around 800. Aptos: According to Chainspect data, Aptos' theoretical highest TPS can reach 160,000, with the highest recorded TPS in operation at 10,734, and daily TPS maintained at 500-1000. In comparison, if Aleo adjusts the number of transitions supported per transaction, the TPS is already sufficient to support most ecosystem applications, such as large-scale DeFi, high-frequency DEX, large NFT trading markets, Blockchain games, and more. Even without adjustments, when the number of validatorsNodes increases to 40, according to traditional calculations, TPS can already reach a level similar to ETH L1, with the official intention being to see the development of L2. Originally, we thought Aleo did not need L2, as L2 is after all a scaling solution, and Aleo can achieve infinite scalability with its zkVM feature. However, for Aleo, L2 is more like an application layer, with the role of increasing TPS, through methods similar to ETH's Rollup. According to traditional blockchain TPS calculations, the maximum number of transactions that can be supported per day with 50 TPS and 100 TPS are: 50 TPS: 4,320,000 transactions/day 100 TPS: 8,640,000 transactions/day
In fact, with transaction data of 50 TPS, it is already sufficient to develop some medium or large-scale ecological applications. If Aleo increases to 100 TPS, reaching a similar TPS to the current ETH mainnet L1, then following the development of its L2 like Ethereum, while adjusting the number of transitions supported by transactions, and relying on its block time of around 2-5 seconds, which is much faster compared to Ethereum's 12 seconds, with the development of Aleo's ecosystem, if Aleo can continue to maintain or further optimize this block time, then the actual transaction volume and transaction experience will far exceed Ethereum! In addition, for the improvement of TPS, besides increasing validatorsNode, snarkVM (zkVM), and snarkOS, performance optimization is also crucial. As introduced in our previous articles: The Significance of CUDA for snarkVM (zkVM) and snarkOS Support! CUDA support is crucial for improving TPS, enhancing the efficiency of zk proof generation and verification, which can further increase the throughput and response speed of the entire system. In summary, Aleo's potential is still very significant. From the progress of the official and community, the ecosystem is advancing rapidly, and more ecological projects will emerge. It is hoped that the number of validatorsNode will be increased to 40 soon. If Aleo can continue to optimize performance, introduce more efficient scaling technologies, it will be able to support a larger ecosystem, including ultra-high-frequency transactions and more complex decentralized applications, while maintaining its advantage in privacy protection. The future of Aleo is very promising!