Create a gist now

Instantly share code, notes, and snippets.

What would you like to do?
Comparing keep-alive on/off for HTTP::Tiny - 11 bytes x 1,000 read/write requests x 10 runs

HTTP Tiny with keep-alive OFF

Write:
           Rate HTTPTiny      LWP  NetCurl     Hijk
HTTPTiny 1.12/s       --      -4%     -61%     -63%
LWP      1.17/s       4%       --     -59%     -61%
NetCurl  2.83/s     154%     143%       --      -7%
Hijk     3.03/s     171%     160%       7%       --
Read:
           Rate      LWP HTTPTiny  NetCurl     Hijk
LWP      1.28/s       --      -6%     -61%     -62%
HTTPTiny 1.36/s       6%       --     -59%     -60%
NetCurl  3.32/s     159%     145%       --      -2%
Hijk     3.38/s     164%     149%       2%       --

HTTP Tiny with keep-alive ON

Write:
           Rate      LWP HTTPTiny  NetCurl     Hijk
LWP      1.17/s       --     -39%     -58%     -64%
HTTPTiny 1.91/s      63%       --     -32%     -41%
NetCurl  2.82/s     141%      48%       --     -12%
Hijk     3.23/s     175%      69%      14%       --
Read:
           Rate      LWP HTTPTiny  NetCurl     Hijk
LWP      1.29/s       --     -48%     -61%     -64%
HTTPTiny 2.49/s      94%       --     -25%     -31%
NetCurl  3.31/s     158%      33%       --      -8%
Hijk     3.61/s     181%      45%       9%       --
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment