​
Introduction

​
Background Information

With the rapid development of blockchain technology, the demand for scaled applications is increasing. To enhance performance and throughput, blockchain projects need to handle large amounts of data. This need is not only driven by technological development, but also to create user-friendly, value-driven applications, as proven by data-driven internet applications. Furthermore, the core belief of blockchain technology—decentralization—requires us to maintain system decentralization and security while improving performance. Therefore, designing an efficient parallel computation execution layer is especially important.

​
Features of the Execution Layer

​
Parallelization

Parallelization is one of the core features of the execution layer, including data parallelism and task parallelism:
  • Data Parallelism: Data parallelism involves dividing a large dataset into smaller data chunks and processing them in parallel on multiple threads, significantly speeding up data processing.
  • Task Parallelism: Task parallelism involves breaking down computational tasks into independent subtasks that are executed in parallel. This approach is suitable for complex computational tasks that can be processed in parallel, improving the system’s computational capabilities and efficiency.
Through data parallelism and task parallelism, the execution layer can handle large-scale data, enhancing overall performance and throughput.

​
On-chain Database

The next-generation on-chain database, Chainbase DB (CDB), optimizes data management and storage efficiency through state storage separation:
  • State Storage Separation: State storage separation is a method of separating the latest state and historical data storage. This design significantly improves data access performance and reduces state bloat problems.
  • Efficient State Management: Through state storage separation, the system can manage and access state data more efficiently, reducing bottlenecks in data access during computation and enhancing overall performance.
The optimization of the on-chain database enables the execution layer to efficiently process and store large amounts of data, providing a solid foundation for parallel computing.

​
Decentralized Environment Based on Eigenlayer AVS

The introduction of Eigenlayer is aimed at enhancing the system’s decentralization and security:
  • Decentralized Verification: Eigenlayer provides a decentralized verification mechanism, introducing Ethereum economic incentives through Restake to ensure system security and reliability.
  • Economic Security: By utilizing Ethereum economic security, Eigenlayer ensures high security in a decentralized environment, preventing malicious attacks and tampering.
The introduction of Eigenlayer not only enhances the system’s decentralization feature but also provides additional economic security, balancing high performance and high security in the execution layer.

​
Programmable Runtime Environment

The Manuscripts runtime environment is designed specifically for the execution of data processing computational logic, providing efficient, flexible execution support:
  • Virtual Machine Environment: The execution layer provides an efficient virtual machine environment through the Chainbase Virtual Machine (CVM) specifically designed for executing Manuscripts. CVM supports multithreaded parallel processing, enhancing computational performance and throughput.
  • Computational Logic Execution: The CVM environment allows developers to write and execute complex data processing logic, supporting various data transformation and processing tasks, and enhancing data processing efficiency.
  • Flexibility and Scalability: CVM provides a flexible development environment and good scalability, supporting various programming languages and frameworks, enabling developers to easily create and deploy data processing logic.
By providing a powerful computational logic execution environment, the execution layer provides developers with an efficient, flexible platform that supports complex data processing tasks and application scenarios.