Performance and Bytecode Size Benchmarks
The performance and bytecode size are critical factors for on-chain scripts in the CKB ecosystem. Understanding these constraints helps developers build efficient and deployable solutions:
-
Execution Cycles: Each transaction has a limited cycle budget (execution time). If a script exceeds this limit, the transaction cannot be processed by the network.
-
Bytecode Size: On-chain storage is expensive and limited. Each cell has a size limit of approximately 500KB, making code optimization essential.
Below are benchmark examples of on-chain scripts with their respective sizes and cycle consumption:
| Script | Size | Cycles |
|---|---|---|
| secp256k1/blake2b | 26KB | 14M cycles |
| simple_udt | 26KB | 12M cycles |
| silent berry AccountBook | 70KB | 20M-40M cycles |
These benchmarks can help you gauge the resource requirements when developing your own ckb-js-vm scripts.
Bytecode Size Impacts Performance
Unlike languages such as C and Rust where binary size has minimal impact on runtime performance, in ckb-js-vm the bytecode size directly affects execution efficiency. Our testing reveals that larger bytecode significantly increases ckb-js-vm boot time, potentially consuming 20-30M cycles.
This insight provides a valuable optimization strategy: reducing your overall bundle size delivers dual benefits of smaller bytecode footprint and improved performance. When developing on-chain scripts for ckb-js-vm, code size optimization should be considered a performance optimization as well.