readme2.md

Concurrency Model (Phase 3.1)

In the current implementation, the system assumes a single-threaded execution model due to compiler and environment constraints.

Thread safety considerations:

• UrlRepository and LRUCache are shared resources.
• In a multi-threaded environment, these components would require synchronization using mutexes or read-write locks.
• This is intentionally deferred and documented as a design decision.

Phase 3.2 – Load Simulation

To simulate concurrent-like behavior without threads, multiple rapid redirect calls were executed.

Observations:

• Frequently accessed URLs stayed in cache.
• Rarely accessed URLs were evicted due to LRU policy.
• Cache significantly optimizes redirect performance.

Phase 3.3 – Scale Considerations

Current limitations at scale:

• Single-node deployment
• Centralized ID generation
• In-memory storage constraints

Planned scaling strategies:

• Distributed ID generation
• Sharded storage
• Multi-node cache deployment

Consistency trade-off:

• Availability is prioritized over strong consistency, as redirects must remain highly available.