Summarizing nine parallel EVM research papers: what are the key discussion points?
TechFlow Selected TechFlow Selected
Summarizing nine parallel EVM research papers: what are the key discussion points?
Conduct an in-depth analysis of its ecosystem and characteristics.
Navigating Web3 tides with focused insightsBy 0xNative
TL;DR
-
Parallel EVM has attracted attention from top-tier VCs, with numerous projects exploring this direction, including Monad, MegaETH, Artela, BNB, Sei Labs, and Polygon.
-
Parallel EVM is not merely about enabling parallel processing—it also involves comprehensive performance optimization of various EVM components. These enhancements support more complex and efficient blockchain applications.
-
Parallel EVM must differentiate itself within the open-source ecosystem while balancing decentralization and high performance, addressing potential security risks and market adoption challenges. The introduction of multithreaded programming complexity creates difficulties in coordinating simultaneous transactions, requiring effective solutions to ensure system stability and security.
-
In the future, Parallel EVM will enable on-chain Central Limit Order Books (CLOBs) and programmable CLOBs (pCLOBs), significantly enhancing DeFi efficiency and driving substantial growth in the DeFi ecosystem.
-
Integrating other high-performance virtual machines (AltVMs) into the Ethereum ecosystem will significantly improve performance and security. This integration leverages the strengths of different VMs, further advancing Ethereum's development.
This year, Parallel EVM has drawn investments from leading venture capital firms such as Paradigm and Dragonfly, attracting widespread market attention. Parallel EVM represents an optimized upgrade over the traditional EVM. The conventional EVM processes transactions sequentially—one after another—leading to congestion and delays during peak transaction periods, degrading user experience. In contrast, Parallel EVM introduces parallel processing technology, allowing multiple independent transactions to execute simultaneously, greatly increasing transaction throughput. With the emergence of complex applications like full-chain games and account abstraction wallets, demand for blockchain performance continues to rise. To support broader user bases, blockchain networks must efficiently process large volumes of transactions. Thus, Parallel EVM serves as a critical solution for the evolving Web3 landscape.
However, implementing Parallel EVM faces several common challenges that require precise technical solutions to ensure stable system operation:
-
Data Consistency: In a parallel EVM, multiple transactions occur simultaneously, potentially reading or modifying account data concurrently. Effective locking or transaction management mechanisms are needed to ensure all operations correctly reflect in the final state and maintain data consistency during state updates.
-
State Access Efficiency: Parallel EVM requires fast state access and updates, necessitating efficient state storage and retrieval mechanisms. Optimizing storage structures and access paths—such as advanced indexing techniques and caching strategies—can significantly reduce data access latency and improve overall system performance.
-
Transaction Conflict Detection: During parallel execution, multiple transactions may depend on the same data states, making transaction ordering and dependency management complex. Sophisticated scheduling algorithms must be designed to identify and manage dependencies among parallel transactions, detect potential conflicts, and determine resolution strategies to ensure results match those of serial execution.
For example, MegaETH decouples transaction execution from full nodes, assigning tasks to specialized nodes to optimize system performance; Artela uses predictive optimistic execution and asynchronous preloading, leveraging AI to analyze transaction dependencies and preload required states into memory to enhance state access efficiency; BNB Chain has developed a dedicated conflict detector and re-execution mechanism to strengthen transaction dependency management and minimize unnecessary re-executions.
To gain deeper insights into the development trajectory of Parallel EVM, below are nine curated high-quality articles on the topic, offering comprehensive perspectives on specific implementations across chains, ecosystem research, and future outlooks.
MegaETH: Unveiling the First Real-Time Blockchain
Author: MegaETH, Date: 2024.6.27
MegaETH is an EVM-compatible Layer 2 aiming to achieve near Web2 server-level real-time performance. Its goal is to push Ethereum L2 performance to hardware limits. MegaETH delivers high transaction throughput, ample computational capacity, and millisecond response times, enabling developers to build and compose complex applications without performance constraints.
MegaETH improves performance by decoupling transaction execution from full nodes and introducing parallel processing. Its architecture consists of three main roles: Sequencer, Verifier, and Full Node.
-
Sequencer: Responsible for ordering and executing user-submitted transactions. After execution, the sequencer sends state changes (state diffs) via a peer-to-peer (p2p) network to full nodes.
-
Full Node: Upon receiving state diffs, full nodes directly apply them to update their local blockchain state without re-executing transactions—significantly reducing computational overhead and improving system efficiency.
-
Verifier: Uses a stateless verification scheme to validate blocks. Instead of sequential validation, verifiers can validate multiple blocks simultaneously, further boosting verification speed and efficiency.
This node specialization allows different node types to set independent hardware requirements based on functional needs. For instance, sequencers require high-performance servers, while full nodes and verifiers can operate on relatively low-spec hardware.
Presenting Artela Scalability Whitepaper - Parallel Execution Stack and Elastic Block Space
Author: Artela, Date: 2024.6.20
Artela has significantly enhanced blockchain parallel execution efficiency and overall performance through several key technologies:
-
Parallel Execution: Predicts transaction dependencies, groups transactions, and leverages multiple CPU cores for parallel processing to improve computational efficiency.
-
Parallel Storage: Optimizes the storage layer to support parallel data processing, avoiding storage bottlenecks and enhancing system performance.
-
Elastic Computing: Enables multiple computers to collaborate, providing elastic computing nodes and block space for dApps, delivering higher transaction throughput and predictable performance.
Specifically, Artela’s predictive optimistic execution uses AI to intelligently analyze dependencies between transactions and contracts, predicting and grouping potentially conflicting transactions to reduce conflicts and redundant executions. The system dynamically accumulates and stores historical transaction state access data for use by prediction algorithms. Through asynchronous preloading, required transaction states are loaded into memory ahead of time, eliminating I/O bottlenecks during execution. Meanwhile, parallel storage separates state commitments from storage operations, optimizing Merkleization and I/O performance. This separation enables independent management of parallel and non-parallel operations, further boosting system parallelization efficiency.
Additionally, Artela builds Elastic Block Space (EBS) on elastic computing. In traditional blockchains, all dApps share the same block space, causing resource competition among high-traffic dApps, resulting in volatile gas fees and unpredictable performance. EBS provides dApps with dedicated, dynamically scalable block space, ensuring predictable performance. dApps can request dedicated block space based on demand, and validators can scale processing power by adding elastic execution nodes as block space grows, ensuring efficient resource utilization across varying transaction volumes.
Road to High Performance: Parallel EVM for BNB Chain
Author: BNB Chain, Date: 2024.2.16
On BNB Chain, the team has taken multiple steps to implement Parallel EVM to enhance transaction processing capability and scalability. Below are the major efforts made by BNB Chain in developing Parallel EVM:
Parallel EVM v1.0
In early 2022, the BNB Chain community advanced Parallel EVM execution, with core components including:
-
Scheduler: Assigns transactions to different threads for parallel execution, optimizing throughput.
-
Parallel Execution Engine: Leverages parallel processing to independently execute transactions on dedicated threads, drastically reducing processing time.
-
Local State Database: Each thread (the basic unit of independent instruction execution) maintains its own "thread-local" state database to efficiently record state access information during execution.
-
Conflict Detection and Re-execution: Ensures data integrity by detecting and managing transaction dependencies, re-executing conflicting transactions to guarantee accurate outcomes.
-
State Commitment Mechanism: Once transactions are executed, results are seamlessly committed to the global state database, updating the blockchain’s overall state.
Parallel EVM v2.0
Building on Parallel EVM 1.0, the BNB Chain community introduced several performance-enhancing innovations:
-
Streaming Pipeline: Improves execution efficiency by enabling smooth transaction processing within the parallel engine.
-
Universal Unconfirmed State Access: Optimizes state access by allowing other transactions to temporarily use results from unconfirmed but completed transactions, reducing inter-transaction wait times.
-
Conflict Detector 2.0: An enhanced conflict detection mechanism that improves performance and accuracy, ensures data integrity, and reduces unnecessary re-executions.
-
Scheduler Enhancement: The scheduler now employs both static and dynamic scheduling strategies to allocate workloads more efficiently and optimize resource utilization.
-
Memory Optimization: Significantly reduces memory usage through shared memory pools and lightweight replication techniques, further boosting system performance.
Parallel EVM v3.0
Following performance gains from Parallel EVM 2.0, the BNB Chain community is actively developing Parallel EVM 3.0, targeting the following goals:
-
Reduce or Eliminate Re-execution: Introduce a Hints-based scheduler that leverages external hint providers to analyze transactions and predict potential state access conflicts. These hints enable better transaction scheduling, minimizing conflicts and reducing the need for re-execution.
-
Modularization: Break code into independent modules to improve maintainability and adaptability across environments.
-
Codebase Refactoring: Align with the latest BSC/opBNB codebase to ensure compatibility and simplify integration.
-
Comprehensive Testing and Validation: Conduct extensive testing under various scenarios and workloads to ensure solution stability and reliability.
Sei’s Parallel Stack
Author: Sei, Date: 2024.3.13
Sei Labs has created an open-source framework called Parallel Stack, designed to build Layer 2 solutions supporting parallel processing technology. The core advantage of Parallel Stack lies in its parallel processing capabilities, leveraging modern hardware advancements to lower transaction costs. The framework features a modular design, allowing developers to add or modify functional modules based on specific needs, adapting to diverse application scenarios and performance requirements. Parallel Stack seamlessly integrates with the existing Ethereum ecosystem. Applications and developers using Parallel Stack can directly leverage Ethereum’s infrastructure and tools without extensive modifications.
To ensure secure execution of transactions and smart contracts, Parallel Stack incorporates multiple security protocols and verification mechanisms, including transaction signature verification, smart contract auditing, and anomaly transaction detection systems. To facilitate application development and deployment on Parallel Stack, Sei Labs provides a complete suite of developer tools and APIs, empowering developers to fully utilize Parallel Stack’s high performance and scalability, driving further growth of the Ethereum ecosystem.
Innovating the Main Chain: a Polygon PoS Study in Parallelization
Author: Polygon Labs, Date: 2022.12.1
The Polygon PoS chain doubled its transaction processing speed through the introduction of a Parallel EVM upgrade, enabled by the minimal metadata approach.
Polygon adapted principles from Aptos Labs’ Block-STM engine, developing the minimal metadata method to meet Polygon’s specific needs. Block-STM is an innovative parallel execution mechanism that assumes no conflicts exist among transactions. During execution, Block-STM monitors each transaction’s memory operations, identifies and marks dependencies, and reorders conflicting transactions for revalidation to ensure correctness.
The minimal metadata method records transaction dependencies within the block and stores them in a Directed Acyclic Graph (DAG). Block proposers and validators first execute transactions, record dependencies, and attach them as metadata. When the block propagates across the network, this dependency information is already included, reducing computational and I/O overhead for revalidation and improving verification efficiency. By recording dependencies, the method also optimizes transaction execution paths, minimizing conflicts.
What Does EVM Parallelization Mean? Or the Endgame Under EVM Hegemony?
Author: Zhixiong Pan, Founder of ChainFeeds, Date: 2024.3.28
Parallel EVM technology has gained recognition and investment from top-tier venture capitalists including Paradigm, Jump, and Dragonfly. These investors believe Parallel EVM can突破 existing blockchain performance limitations, enabling more efficient transaction processing and broader application possibilities.
Although “Parallel EVM” literally refers only to “parallelization,” the technological improvements it encompasses go far beyond that. It does not simply allow multiple transactions or tasks to be processed simultaneously, but also includes deep performance optimizations across various components of the Ethereum EVM—such as faster data access, improved computation efficiency, and optimized state management. As such, it may represent the performance ceiling achievable under the EVM standard.
Beyond technical hurdles, Parallel EVM also faces challenges in ecosystem development and market adoption. It must differentiate itself within the open-source ecosystem while finding an appropriate balance between decentralization and high performance. From a market adoption perspective, it must demonstrate tangible performance improvements and cost benefits—especially given that many existing Ethereum applications and smart contracts are already stable. The incentive to migrate to a new platform must be compelling. Moreover, promoting Parallel EVM requires addressing potential security issues and emerging technical flaws to ensure system stability and asset safety—key factors in driving broad technology adoption.
Death, Taxes, and EVM Parallelization
Author: Reforge Research, Date: 2024.4.1
With the advent of Parallel EVM, the feasibility of on-chain Central Limit Order Books (CLOBs) has increased, and DeFi activity is expected to grow significantly.
In a CLOB, orders are prioritized by price and time, ensuring market fairness and transparency. However, implementing CLOBs on blockchain platforms like Ethereum has been limited by processing capacity and speed, often resulting in high latency and transaction costs. With Parallel EVM, network processing capability and efficiency are greatly enhanced, enabling DeFi platforms to perform faster and more efficient order matching and execution—making CLOBs viable.
Building on this, programmable CLOBs (pCLOBs) extend CLOB functionality further. pCLOBs not only provide basic buy-sell order matching but also allow developers to embed custom smart contract logic during order submission and execution. These custom logics can enable additional validations, conditional executions, and dynamic fee adjustments. By embedding smart contracts into the order book, pCLOBs offer greater flexibility and security, supporting more sophisticated trading strategies and financial products. Leveraging the high performance and parallel processing capabilities of Parallel EVM, pCLOBs can deliver complex and efficient trading functions in decentralized environments comparable to traditional financial exchanges.
Nonetheless, despite significant performance gains from Parallel EVM, existing Ethereum Virtual Machine (EVM) and smart contract security remain insufficient and vulnerable to hacker attacks. To address these issues, the author recommends adopting a dual VM architecture. In this setup, alongside the EVM, an independent virtual machine (e.g., CosmWasm) monitors EVM smart contract execution in real time. This secondary VM acts like antivirus software in an operating system, performing advanced threat detection and protection, thereby reducing hacking risks. Emerging solutions like Arbitrum Stylus and Artela are believed to have strong potential in realizing such dual VM architectures. These new systems could embed real-time protection and other critical security features from inception.
What Will Be the Next Step Towards Enhanced Scalability While Maintaining EVM Compatibility?
Author: Grace Deng, Researcher at SevenX Ventures, Date: 2024.4.5
New Layer 1 blockchains like Solana and Sui achieve higher performance than traditional L1s and L2s by using novel virtual machines (VMs) and programming languages, along with parallel execution, new consensus mechanisms, and database designs. However, these systems lack EVM compatibility, leading to fragmented liquidity and higher barriers for users and developers. In contrast, EVM-compatible L1s like BNB and AVAX have made improvements at the consensus layer but implemented fewer changes to the execution engine, limiting their performance gains.
Parallel EVM enhances performance without sacrificing EVM compatibility. For example, Sei V2 improves read-write efficiency using Optimistic Concurrency Control (OCC) and a new state tree (IAVL trie); Canto Cyclone optimizes its state management system using the latest Cosmos SDK, ABCI 2.0, and in-memory IAVL state trees; Monad proposes a novel L1 combining high throughput, decentralization, and EVM compatibility, incorporating OCC, a new parallel-access database, and the HotStuff-based MonadBFT consensus mechanism.
Beyond this, integrating other high-performance VMs (AltVMs) into the Ethereum ecosystem—especially those supporting Rust development, such as Solana’s Sealevel or Near’s WASM-based VM—could resolve EVM incompatibility issues, bring Rust developers into the Ethereum ecosystem, improve overall performance and security, and unlock new technical possibilities.
A Deep Dive into Parallel EVM: How to Break Through Blockchain Performance Bottlenecks?
Author: Gryphsis Academy, Date: 2024.4.5
Parallel EVM primarily focuses on performance optimization at the execution layer, falling into two categories: Layer 1 and Layer 2 solutions. Layer 1 solutions introduce parallel transaction execution, enabling transactions to run as concurrently as possible within the VM. Layer 2 solutions essentially leverage already-parallelized Layer 1 VMs to achieve off-chain execution with on-chain settlement. In the future, the Layer 1 space may split into two camps—parallel EVM and parallel non-EVM—while the Layer 2 space evolves toward blockchain VM simulators or modular blockchains.
Parallel execution mechanisms fall into three main categories:
-
Message Passing Model: Each actor can only access its private data, requiring message passing to access others’ data.
-
Shared Memory Model: Controls access to shared resources using memory locks, divided into lock-based and optimistic parallelization models.
-
Strict State Access List: Based on the UTXO model, pre-calculates account addresses each transaction will access, forming an access list.
Different projects adopt distinct strategies in implementing parallel execution:
-
Sei v2: Shifted from a memory lock model to an optimistic parallelization model to optimize potential data contention.
-
Monad: Introduced superscalar pipelining and improved optimistic parallelization to achieve up to 10,000 TPS.
-
Canto: Adopted Cyclone EVM with optimistic parallelization and innovated in decentralized finance infrastructure.
-
Fuel: As a modular Ethereum rollup OS, uses the UTXO model and optimistic parallelization to increase transaction throughput.
-
Neon, Eclipse, and Lumio: Enhance cross-ecosystem performance by integrating capabilities from multiple L1 chains, employing dual-VM support strategies.
While Parallel EVM offers an effective solution, it also introduces new security challenges. Parallel execution brings multithreaded programming complexities, increasing system intricacy. Multithreading is prone to race conditions, deadlocks, livelocks, and starvation, affecting system stability and security. It also introduces new vulnerabilities—malicious transactions might exploit the system’s parallel execution mechanism to cause data inconsistency or launch race attacks.
Join TechFlow official community to stay tuned
Telegram:https://t.me/TechFlowDaily
X (Twitter):https://x.com/TechFlowPost
X (Twitter) EN:https://x.com/BlockFlow_News
@ChainFeedsxyz
