Introduction to monolith
Latest Updates
- [2026.03.30]: Adapted and optimized the ByteDance's open-source monolith project for the Kunpeng Arm platform.
Project Introduction
Based on the open-source monolith large-scale real-time recommendation system framework, this project focuses on its core component, the Embedding Lookup module. This module is responsible for efficiently retrieving embedding vectors from large-scale sparse hash tables, which serves as a key performance bottleneck during the online inference phase. To accelerate the migration and deployment of monolith on the Kunpeng Arm architecture platform, this project performs deep adaptation and performance optimization on the native code while ensuring strict functional correctness. By fully leveraging the hardware capabilities of the Kunpeng Arm platform, the project introduces multiple optimizations including compiler option tuning, spinlock optimization, memory alignment adjustment, and Arm SIMD vectorization. These enhancements significantly reduce the table lookup latency of the Embedding Lookup module, delivering high-performance inference capabilities on the Arm platform. They can provide a solid foundation for migrating the monolith recommendation framework to the Arm architecture.
Directory Structure
The directory structure of the open-source monolith repository is as follows:
monolith
├─ docs
| ├─ en
| | ├─ *.md // Documentation
| └─ LICENSE // License file
├─ 0001-boostsra-monolith.patch // Optimization patch
└─ README_en.md // Code repository introductionDocuments
| Introduction to | ||
|---|---|---|
Disclaimer
This code repository contributes to the monolith open-source component, and is used exclusively for monolith Arm platform adaptation. It strictly adheres to the coding style and methods, as well as security design of the native open-source software. Any vulnerability and security issues of the software shall be resolved by the corresponding upstream communities according to their response mechanisms. Please pay attention to the notifications and version updates released by the upstream communities. The Kunpeng computing community does not assume any responsibility for software vulnerabilities and security issues.
Contributions
We welcome your contributions to the community. If you have any questions/suggestions or want to provide feedback on feature requirements and bug reports, you can submit issues. For details, see the contribution guideline. You are also welcome to share insights in the Discussions. Thank you for your support.
License
This project is released under the Apache License 2.0. For details, see LICENSE.
The documents of this project are licensed under CC-BY 4.0. For details, see License.