Somnia: Technological Innovations of the Next Generation Parallel EVM Public Chain

This article will explore two key issues:

1. Why might the Somnia "high-speed public chain" project be different from others?

2. Somnia claims to be the fastest and most cost-effective parallel EVM Layer 1. Is this statement credible?

Brief Overview

Summarize Somnia from three aspects: technology, background, and ecology:

Technical Highlights of Somnia

• Multi-flow consensus algorithm: Separating the data chain and the consensus chain helps to prevent MEV, reduce redundancy, and at the same time lower costs and improve efficiency.

• Innovative EVM Compiler: Achieves instruction-level parallel EVM, solving high-frequency interaction issues in extreme cases.

• Self-developed IceDB database engine: improves data read and write speed and network stability.

• Data compression technology: Improve data transmission efficiency.

The background advantages of Somnia

• The development team comes from Improbable, a multinational technology company founded in 2012, headquartered in London, UK, which has developed software, games, and Web3 metaverse products.

• Received a total investment of $270 million from well-known institutions such as MSquared, a16z, SoftBank, and Mirana.

The ecological progress of Somnia

• The testnet has onboarded 4 AI/social products, 7 games, 4 NFT projects, and 6 DeFi applications, with an additional 2 AI/social products, 11 games, and 1 DeFi application set to launch soon.

• As of June 26, 2025, the testnet has produced over 100 million blocks, with an average block time of 0.1 seconds. A total of 96,878,557 wallet addresses have participated, and the trading volume in the last day reached 26.43 million transactions.

The uniqueness of Somnia

• High-frequency interaction: Not only pursuing technical indicators, but also committed to serving applications in high-frequency interaction scenarios such as gaming and socializing.

• Integration of Web3 and Web2: The background of Somnia helps provide a seamless entry for Web2 users into the Web3 world, potentially leading to a user experience-centric application ecosystem.

Technical In-depth Analysis

Multi-Stream Consensus Algorithm

Somnia adopts the Multistream ( algorithm, which consists of two parts: the data chain and the consensus chain.

• Data Chain: Multiple data chains record transaction information, with each chain managed by one validator, operating independently.

• Consensus Chain: executes consensus, sorts transactions, records transaction references, and is jointly maintained by all validators.

Workflow:

1. After the user issues a request, the validator writes the transaction to the data chain.

2. The consensus chain periodically synchronizes the data shards at the top of the data chain.

3. Validators write the collection of all data shard sets at the top of the data chain into the consensus chain.

4. Validators sort transactions, update states, and synchronize writes to the IceDB database.

Advantages:

• Transaction ordering is beneficial to prevent MEV: using a deterministic pseudo-random function for ordering, making it difficult for attackers to predict and manipulate.

• Reduce redundancy and improve efficiency: Minimize data validation and storage redundancy among validators.

• Tamper-proof: Tampering with transaction information will lead to conflicts with the consensus chain records.

![The market has completely desensitized to "high-speed public chains". Why is it said that Somnia might be different?])https://img-cdn.gateio.im/webp-social/moments-110bee67b72246266e01aed97d7137f5.webp(

) Instruction-level parallel EVM

To solve the congestion problem of high-frequency related transactions, Somnia has innovatively developed an EVM compiler:

• Split the transaction into an instruction set, where non-conflicting and independent instruction sets can be executed in parallel.

• Compile EVM bytecode into x86 machine code and utilize CPU multi-threading to execute the instruction set in parallel.

• Regular trading still uses standard EVM interpretation execution, while high-frequency trading enables the compiler to achieve both cost and efficiency advantages.

![The market has completely desensitized to "high-speed public chains", why is Somnia said to be potentially different?]###https://img-cdn.gateio.im/webp-social/moments-15269b2b0e2c1e8b9839d87195577452.webp(

) IceDB database engine

Use LSM tree ### log-structured merge tree ( to replace Merkle tree:

• Data append writing, no need to calculate hash, improving writing speed.

• Create an optimized data cache for reading and writing, with an average read and write time of 15-100 nanoseconds.

• Return performance report to provide basis for calculating Gas, making the network fairer and more efficient.

) Data compression technology

• Achieve high compression rates using the theory of information quantity and frequency distribution power law.

• Stream compression enhances network transmission capabilities.

• Use BLS signatures to improve the speed of signature transmission and verification.

• The multi-stream consensus design makes network transmission more balanced and stable.

![The market has completely desensitized to "high-speed public chains", why is it said that Somnia might be different?]###https://img-cdn.gateio.im/webp-social/moments-6fb61b68f6ac9b456c191699498e87a1.webp(

Summary

Somnia integrates Web2 and Web3 technologies, providing performance guarantees for high-frequency interactive applications through innovations such as multi-stream consensus, instruction-level parallel EVM, and IceDB. Its background helps promote the integration of Web3 and Web2, potentially leading to a new ecosystem that combines the ownership of Web3 with the experience of Web2.