First of all great article and great idea! I made a very quick test on a Macbook Pro 2017 with i7 2,9 Ghz and MacOS High Sierra beta and initially results on aspnetcore were a disaster due to the logging in console at info level. I change level to error to align it to Iris and results are very near:
Aspnetcore:
Bombarding http://localhost:5000/api/values/5 with 5000000 requests using 125 connections 5000000 / 5000000 [=] 100.00% 1m11s
Statistics Avg Stdev Max
Reqs/sec 69977.53 6359.43 87478
Latency 1.79ms 0.86ms 82.17ms
HTTP codes:
1xx — 0, 2xx — 5000000, 3xx — 0, 4xx — 0, 5xx — 0
others — 0
Throughput: 12.68MB/s
Go:
Bombarding http://localhost:5000/api/values/5 with 5000000 requests using 125 connections 5000000 / 5000000 [=] 100.00% 1m8s
Statistics Avg Stdev Max
Reqs/sec 73112.62 6154.42 94812
Latency 1.71ms 403.20us 26.51ms
HTTP codes:
1xx — 0, 2xx — 5000000, 3xx — 0, 4xx — 0, 5xx — 0
others — 0
Throughput: 13.59MB/s
I will make more changes and tests soon.
