[Improvement] Optimize API Performance: Log and Store API Latency, Refactor Long Operations

[iziang]

Background

The current API implementation can be further optimized for better performance and observability.

Suggestions

1. Log and Store API Latency

   • Print the latency (duration) of each API call in the logs.
   • Persist the latency data directly into the database for later analysis.

2. Refactor Long Operations

   • Move time-consuming operations out of the session context to avoid holding database connections for extended periods.
   • This will help reduce connection pool pressure and improve overall API responsiveness.

Expected Benefits

• Improved monitoring and analysis of API performance bottlenecks.
• Reduced risk of database connection exhaustion due to long-held sessions.
• Better scalability and reliability for the API service.