Cloud Cup is a modern, high-performance reverse proxy built for Linux in Zig, designed to handle today's web traffic demands with speed, scalability, and resilience.

At its core, Cloud Cup implements a master-worker architecture that ensures uninterrupted service, dynamic traffic management, and robust request handling. Whether you’re a developer building microservices or an operations engineer managing web infrastructure, Cloud Cup offers the tools to keep your applications running smoothly and efficiently.

---

Master-Worker Architecture

Cloud Cup’s architecture revolves around its master and worker processes, each with distinct responsibilities:

Master Process

• Validates and applies configurations at startup, ensuring they are correct before spawning workers.
• Hot reloads configurations via the CLI (cupctl), where the CLI process validates the new configuration and distributes it to workers dynamically.
• Spawns multiple worker processes and monitors them to ensure high availability. If a worker process exits unexpectedly, the master immediately spawns a replacement.
• Manages the CLI command listener, enabling seamless interaction for updates and diagnostics.

Worker Process

• All workers listen on the same port and address, leveraging the kernel to distribute incoming requests efficiently among them.
• Each worker operates with its own epoll instance for high-performance request handling.
• Utilizes an independent thread pool to handle requests concurrently, ensuring scalability and responsiveness.
• Dynamically receives and applies configuration updates from the master process, maintaining consistency across all workers.

This architecture ensures Cloud Cup remains operational even under heavy load or in the event of individual worker failures, delivering unparalleled reliability and performance.

---

Why Choose Cloud Cup?

Cloud Cup addresses modern challenges in web traffic management with cutting-edge features:

• Zero Downtime: The master-worker model ensures high availability, with automatic worker respawning and dynamic configuration updates.
• Kernel-Level Efficiency: Workers listening on the same port let the Linux kernel manage request distribution, optimizing resource utilization.
• Performance at Scale: Each worker uses its own epoll and thread pool, maximizing throughput and responsiveness.
• Dynamic Management: Use the CLI to validate, reload, and apply configurations without interrupting traffic.
• High Customizability: Define granular routes and backends using a simple JSON file.

---