Retry storm
Difficulty 3 of 5
One thousand callers per second use a one-core service. Its link fails from 0:02 to 0:04. Recover at least 99.9 percent of requests without a second overload wave.
0:04. The link is back, but the service overloads again. Graph: two error peaks, the second one longer.
3 retries per call · no jitter · peak 6.0× normal load
Goals and limits
- Error rate
- under 0.1%
- p99 latency
- under 5 s
- Monthly cost
- under 100 USD/mo
- Edit limit
- 3
- Traffic
- 1,000 req/s
Concepts
How players did
Player results load from the server.