Overloaded database
Difficulty 1 of 5
Users send 2,000 requests/s. The database answers about 1,000 reads/s, so requests wait and fail. Most requests read the same data. Put a cache between the API and the database.
1,600 keys · hit 81% · mean 3.2 ms
Goals and limits
- p99 latency
- under 200 ms
- Error rate
- under 0.1%
- Monthly cost
- under 20 USD/mo
- Parts you can add
- Redis cache
- Part limit
- 1
- Hosts
- 2 vCPU · 8 GB
- Traffic
- 2,000 req/s