Closed alexkroman closed 11 years ago
Not bad (comparable with @juliangiuca's Go implementation or @gnarg's similar Netty implementation:
Made 40000 requests in 18.04s.
Times in ms, min/mean/median/max (stddev)
Handled 40000 requests: 5.9/28.8/15.8/1167.9 (108.5)
IQR filtered to 39186 requests: 8.4/15.7/15.7/24.9 (3.0)
Throughput: 133.04krpm (IQR adjusted: 244.18krpm).
That said, @jthurman42 just stole the overall performance crown with his dirt-simple preforking implementation in C:
Made 40000 requests in 16.56s.
Times in ms, min/mean/median/max (stddev)
Handled 40000 requests: 0.7/26.5/11.0/2637.0 (157.6)
IQR filtered to 37545 requests: 7.1/11.5/10.9/16.1 (1.5)
Throughput: 144.93krpm (IQR adjusted: 334.23krpm).
Curious to see how your NIO implementation did...
This is a basic Java server using Netty.