In March this year, the scalable L1 blockchain network Artela launched EVM++, an upgrade targeting the next generation of EVM execution layer technology. The first “+” in EVM++ stands for “Extensibility,” achieved through Aspect technology to support developers in creating on-chain custom programs in a WebAssembly (WASM) environment. These programs can collaborate with EVM to provide high-performance, customized application-specific extensions for dApps. The second “+” represents “Scalability,” achieved through parallel execution techniques and the design of elastic block space, significantly enhancing network processing capacity and efficiency.

WebAssembly (WASM) is an efficient binary code format capable of achieving near-native execution speeds in web browsers, making it particularly suitable for handling compute-intensive tasks such as AI and big data processing.

Yesterday, Artela released a whitepaper detailing how it enhances blockchain scalability through the development of parallel execution stacks and the introduction of elastic block space based on elastic computing principles.

The Importance of Parallel Processing

In the traditional Ethereum Virtual Machine (EVM), all smart contract operations and state transitions must be globally consistent across the network. This requires all nodes to execute the same transactions in the same order, even if some transactions have no actual dependencies between them. This results in serial processing, causing unnecessary delays and inefficiencies.

Parallel processing allows multiple processors or computing cores to execute multiple computing tasks or process data simultaneously, significantly improving processing efficiency and reducing execution time, especially for complex or large-scale computing problems that can be divided into independent tasks. Parallel EVM extends or enhances the traditional EVM by enabling concurrent execution of multiple smart contracts or contract function calls, thereby boosting the overall network throughput and efficiency. Additionally, it optimizes efficiency compared to single-thread execution. The primary advantage of Parallel EVM is enabling decentralized applications to achieve performance comparable to that of the internet.

Artela Network and EVM++

Artela is an L1 that enhances the scalability and performance of the EVM by introducing EVM++. EVM++ upgrades the EVM execution layer, integrating the flexibility of EVM with the high performance of WASM. This enhanced virtual machine supports parallel processing and efficient storage, enabling more complex and performance-demanding applications to run on Artela. EVM++ not only supports traditional smart contracts but also allows for the dynamic addition and execution of high-performance modules on-chain, such as AI agents, which can run as on-chain co-processors independently or directly participate in on-chain games, creating truly programmable NPCs.

Artela ensures that network nodes’ computing power can be flexibly scaled according to demand through its parallel execution design. Additionally, validator nodes support horizontal scaling, allowing the network to automatically adjust the scale of computing nodes based on current load or demand. This scaling process is coordinated by an elastic protocol to ensure sufficient computing resources in the consensus network. By guaranteeing scalable computing power through elastic computing, Artela achieves elastic block space, allowing large dApps to apply for independent block space based on specific needs. This not only meets the need for expanding public block space but also ensures the performance and stability of large applications.

Detailed Explanation of Artela’s Parallel Execution Architecture

1. Predictive Optimistic Execution

Predictive optimistic execution is one of Artela’s core technologies and distinguishes it from other parallel EVMs like Sei and Monad. Optimistic execution refers to a parallel execution strategy that assumes initially there are no conflicts between transactions. In this mechanism, each transaction maintains a private version of the state, recording modifications without immediately finalizing them. After transaction execution, a validation phase checks for conflicts caused by global state changes from concurrent transactions during the same period. If conflicts are detected, transactions are re-executed. Predictiveness involves using specific AI models to analyze historical transaction data, predicting dependencies between transactions about to be executed—and identifying which transactions might access the same data. Based on this analysis, transactions are grouped and their execution order is arranged to reduce conflicts and redundant executions.

In contrast, Sei relies on developers defining transaction dependencies in advance via files, while Monad uses compiler-level static analysis to generate transaction dependency files. Neither Sei nor Monad achieve EVM equivalence and lack Artela’s adaptive capability based on AI-driven dynamic prediction models.

2. Async Preloading

Asynchronous preloading technology aims to address input-output (I/O) bottlenecks caused by state access, with the goal of enhancing data retrieval speed and reducing transaction execution waiting times. In Artela, before executing transactions, necessary state data is preloaded from slow storage (such as hard disks) into fast storage (such as memory) based on predictive models. This proactive loading of required data minimizes I/O wait times during execution. With data preloaded and cached, multiple processors or execution threads can simultaneously access this data, further increasing execution parallelism.

3. Parallel Storage

With the introduction of parallel execution technology, transaction processing can be parallelized, but if the speed of data read, write, and update cannot be synchronized, it becomes a critical factor limiting overall system performance. Consequently, the bottleneck gradually shifts to the storage layer. Solutions like MonadDB and SeiDB have begun focusing on optimizing the storage layer. Artela draws on and integrates various mature traditional data processing techniques to develop parallel storage, further enhancing the efficiency of parallel processing.

Parallel storage systems are designed primarily to address two major issues: achieving parallel processing of storage and improving the efficient recording of data states into databases. Common challenges in data storage include data inflation during write operations and increased pressure on database processing. To effectively tackle these issues, Artela adopts a separation strategy between State Commitment (SC) and State Storage (SS). This strategy divides storage tasks into two parts: one part handles operations that require fast processing without retaining complex data structures, thus saving space and reducing data redundancy; the other part is responsible for recording detailed data information comprehensively.

Moreover, to maintain performance while handling large volumes of data, Artela employs a method of aggregating small data blocks into larger ones, reducing the complexity of data storage operations.

4. Elastic Block Space (EBS)

Artela’s Elastic Block Space (EBS) is designed based on the concept of elastic computing, allowing automatic adjustment of the number of transactions a block can accommodate based on network congestion levels.

Elastic computing is a cloud computing service model that enables systems to automatically adjust the configuration of computing resources to meet varying workload demands. Its primary goal is to optimize resource utilization efficiency and ensure rapid provisioning of additional computing power when demand increases.

EBS dynamically adjusts block resources according to the specific needs of dApps, providing independent scaling block space for high-demand dApps. This aims to address significant differences in blockchain performance requirements across various applications. The core advantage of EBS lies in “predictable performance,” ensuring dApps receive predictable Transactions Per Second (TPS). Thus, regardless of congestion in public block space, dApps with independent block space enjoy stable TPS. Moreover, if dApps’ contracts support parallel processing, they can achieve even higher TPS. In essence, EBS provides a more stable environment compared to traditional blockchain platforms like Ethereum and Solana, which often experience performance degradation during network congestion, such as during NFT booms or DeFi peaks. Artela effectively resolves such issues through customized and optimized resource management.

In summary, Artela achieves high scalability and predictable network performance through its parallel execution stack and Elastic Block Space (EBS). This parallel execution architecture uses AI models to accurately predict transaction dependencies, thereby reducing conflicts and redundant executions. Moreover, large-scale applications can access dedicated processing power and resources as needed, ensuring stable performance even under high network loads. This capability enables the Artela network to support more complex use cases such as real-time big data processing and sophisticated financial transactions.

Statement:

1. This article is reproduced from [ChainFeeds Research], the copyright belongs to the original author [0XNATALIE], if you have any objections to the reprint, please contact the Gate Learn team, and the team will handle it as soon as possible according to relevant procedures.

2. Disclaimer: The views and opinions expressed in this article represent only the author’s personal views and do not constitute any investment advice.

3. Other language versions of the article are translated by the Gate Learn team and are not mentioned in Gate.io, the translated article may not be reproduced, distributed or plagiarized.