A new open-source MySQL 8.0 optimization project has emerged, claiming significant performance improvements over the official release. The project addresses several long-standing issues including InnoDB scalability problems, redo log optimization, and performance degradation that occurred since MySQL 8.0.28. However, the project's unusual patch delivery method and anonymous nature have sparked community discussion.
Performance Claims and Technical Improvements
The enhanced MySQL version targets high-concurrency environments and claims to deliver substantial performance gains on high-end hardware. Key optimizations include improvements to hash join cost calculations, bulk insert performance enhancements, and replica replay acceleration. The project particularly emphasizes benefits in NUMA (Non-Uniform Memory Access) environments, where the official MySQL version reportedly struggles with scalability issues.
The optimizations also address Group Replication stability problems by simplifying the underlying Paxos protocol implementation. According to the project documentation, they removed conflict detection mechanisms to improve write performance, though this makes the enhanced version incompatible with official MySQL during runtime.
Key Technical Improvements:
- InnoDB storage engine scalability enhancements
- Redo log optimization for better performance
- Hash join cost calculation improvements (since MySQL 8.0.42)
- Bulk insert performance optimization via data structure replacement
- Replica replay acceleration for faster secondary synchronization
- NUMA compatibility fixes for better multi-processor performance
- Group Replication stability improvements with simplified Paxos protocol
Community Concerns About Delivery and Documentation
The project's approach to distributing improvements has raised eyebrows in the developer community. Rather than providing individual patches or maintaining a proper Git fork, the team releases a massive 14.7MB patch file containing over 2,300 individual changes. This delivery method makes it extremely difficult for developers to review specific modifications or understand the scope of changes.
I also just for-real don't understand how in the universe a ~15MB text file against an open source git hosted project is a sane way of delivering value.
The documentation style has also drawn criticism, with some community members suggesting it may have been generated with AI assistance due to its verbose nature and occasional inconsistencies. Despite these concerns, the project does reference a companion book that provides detailed explanations of the optimization principles.
Available Download Versions:
- MySQL 8.0.42 Optimization Patch
- Source Code of Improved Version Based on MySQL 8.0.48
- Binary releases for CentOS 8.0 and 7.0 on x86 architecture
- PGO (Profile-Guided Optimization) compiled versions recommended
Anonymous Origins and Adoption Challenges
One significant concern raised by the community is the anonymous nature of the project maintainers. Unlike other MySQL forks such as Percona Server, which have clear corporate backing and known development teams, this enhanced MySQL project provides no information about who is behind the optimizations. This anonymity may make enterprise adoption more challenging, as organizations typically prefer solutions with clear support channels and accountability.
The project claims to track official MySQL releases closely, promising new optimized versions within a week of official releases. They also offer long-term support for specific versions, addressing a common pain point where users must upgrade across major versions to resolve critical issues.
Performance Environment Recommendations:
- High-end machines: Substantial performance gains, especially with NUMA architecture
- Medium-end machines: Moderate improvements, PGO optimization effective
- Low-end/VM environments: Limited benefits, improvements not guaranteed
- Concurrency testing: Recommended limit of 1,000 concurrent connections
- Testing tools: BenchmarkSQL or TPC-C preferred over sysbench
Real-World Performance Validation
The enhanced MySQL project acknowledges that performance improvements are most noticeable on high-end hardware with significant concurrent workloads. For low-end environments or virtual machines, the benefits may be minimal or even negligible. The team recommends using tools like BenchmarkSQL or TPC-C for performance testing rather than simple benchmarks like sysbench.
Interestingly, the project addresses a common misconception about MySQL 8.0 performance decline. They explain that while 8.0 may appear slower in low-concurrency scenarios due to batch processing mechanisms, it significantly outperforms MySQL 5.7 in high-concurrency environments where its scalability improvements shine.
The emergence of this enhanced MySQL fork reflects ongoing frustration with the pace of official MySQL optimization efforts. While the technical improvements appear promising, the unusual delivery method and anonymous nature may limit widespread adoption until these concerns are addressed.
Reference: Enhancing MySQL: Performance, Stability, and High Availability