The reason KUSANAGI achieves speed improvements of up to 260× lies in its vertically integrated architecture, consistently optimized from the OS level to the application layer. Rather than combining isolated technologies, the entire stack is integrated under a design philosophy specialized for CMS execution, minimizing context switching and I/O wait time. In addition, database, PHP, and Nginx components are integrated within a single VM, localizing inter-process communication and minimizing network latency and overhead typically introduced by distributed architectures.
1. Thorough Optimization at the OS and Middleware Level
OS Layer Optimization
Based on AlmaLinux/CentOS Stream, unnecessary services are disabled and resource allocation is optimized for web server workloads. Middleware is installed independently under /etc/opt/kusanagi/, avoiding conflicts with standard OS packages while enabling CMS-specific patches.
Maximizing PHP Execution Efficiency
- OPcache Enabled: Caches PHP scripts as bytecode in shared memory, eliminating parsing and compilation overhead for each request
- JIT: Further accelerates script execution through runtime compilation
- Parameter Tuning: Default settings optimized for WordPress execution
Nginx + PHP-FPM Optimization
- Data Compression: Reduces transfer size using gzip or Brotli
- Parameter Tuning: Worker process configuration and request handling optimized for high concurrency
- Native HTTP/2 and HTTP/3 Support: Multiplexing enables efficient transfer of multiple resources, improving handling of numerous concurrent connections
Database (MariaDB) Tuning
- Parameter Optimization including Query Cache: Settings tuned for WordPress’s complex query patterns
- Memory Allocation Optimization: Reduces bottlenecks during dynamic query execution
- Maximizes memory utilization to reduce disk I/O
WordPress-Specific Accelerators
- Translation Accelerator: Reduces or bypasses .mo file loading overhead during multilingual operation, significantly reducing execution time
- Theme Accelerator: Intervenes in rendering processes to optimize output generation paths
2. Layered Cache Strategy
A key factor in KUSANAGI’s performance is the layered combination of two caching mechanisms.
fcache
Operates at the Nginx layer, returning responses before communication with PHP-FPM, achieving near-maximum throughput.
- Operating Principle: Completely bypasses PHP and database access
- Smart Strategy Feature: Automatically switches cache usage between desktop and mobile clients
- Measured Performance: Maintains response times under 8 ms and processing capacity exceeding 80,000 requests per second under 1,000 concurrent connections
bcache
Operates at the WordPress layer, storing cache in the database (e.g., wp_site_cache table) to reduce query execution frequency.
- Scalability: The
CACHE_DB_HOSTsetting allows cache storage to be separated onto a database on another server, reducing memory and I/O load on the application server and supporting scale-out architectures - Page Caching: Reduces the load associated with dynamic page generation
Safety Design
Logged-in users, URLs with query parameters, and POST requests are excluded from caching by default. Additionally, as a DoS mitigation measure, the kusanagi ratelimit command limits requests from the same IP to a default of 100 r/s (burst=10), helping prevent server downtime.
3. Performance Proven by Measured Data
Measurements conducted in standard environments such as Microsoft Azure Standard D4as_v5 (4 vCPU, 16 GiB):
- With Cache: Up to 260× faster compared to a non-optimized baseline configuration
- Without Cache: Approximately 2× faster through optimization settings
- TCO Reduction Effect: Low-spec VMs with half the CPU and one-quarter the memory can achieve performance comparable to high-end servers, reducing infrastructure costs by up to approximately 75%
Summary: Technical Advantages
KUSANAGI’s speed is not simply the result of a fast server but of a vertically integrated platform grounded in coherent systems design. By optimizing consistently from deep OS-level configuration through middleware tuning to application behavior, and aligning all layers toward CMS acceleration, it delivers performance that cannot be achieved through isolated optimizations alone.
For engineers, KUSANAGI represents a practical approach that reduces dependence on individual expertise in infrastructure setup while achieving a balanced combination of performance, security, and cost efficiency.
