Retry budget

Guard 02
HTTP/2Users1,000 req/sAPI server4 vCPU · 4 GB
HTTP/2Users1,000 req/sAPI server4 vCPU · 4 GB

Tune the parts. Turn every goal green.

Traffic
1,000 req/s · reads only
Limit
API server answers about 13,000 req/s
Bill
  • API server16 USD/mo
16 of 20 USD/mo

Goals

Not tested yet

The share of requests that fail or get no answer.99 of 100 requests finish faster than this.How busy the busiest CPU is. Keep room for spikes.What the servers and services cost each month.

At 0:02, the request link fails for 0.4 seconds. Clients send 1,000 reads/s for 10 seconds. Tune retries and backoff. Finish at least 99.9 percent of requests, keep p99 below 1.5 seconds, and keep peak CPU below 80 percent.