Safe load limit

Level 6 of 41
HTTP/2Users6,000 req/sRate limiter1 vCPU · 0.25 GBAPI server1 vCPU · 1 GB
HTTP/2HTTP/2Users6,000 req/sRate limiter1 vCPU · 0.25 GBAPI server1 vCPU · 1 GB

Tune the parts. Turn every goal green.

Traffic
6,000 req/s · reads only
Limit
API server answers about 3,300 req/s
Bill
  • Rate limiter4 USD/mo
  • API server4 USD/mo
8 of 20 USD/mo

Goals

Not tested yet

How busy the busiest CPU is. Keep room for spikes.Requests served each second.99 of 100 requests finish faster than this.What the servers and services cost each month.

Users send 6,000 requests/s to one API. Keep API CPU below 70 percent and p99 below 100 ms. Serve at least 1,800 requests/s. You may reject extra requests.