Hi all,<br>
<br>
I measured network I/O performance of native Linux-2.4 and Fiasco-1.2+L4Linux-2.4 using the netperf benchmark.<br>
<br>
The TCP streaming test results are following:<br>
<br>
Recv Socket Size bytes: Linux = 87380, L4Linux = 87380<br>
Send Socket Size bytes: Linux = 16384, L4Linux = 16384<br>
Send Message Size bytes: Linux = 16384, L4Linux=16384<br>
Elapsed Time secs: Linux = 10.002, L4Linux = 10.002<br>
Throughput MBytes/s: Linux = 11.22, L4Linux = 11.206<br>
Utilization Send local % S: Linux = 11.359, L4Linux = 4.35<br>
...<br>
<br>
The last 2 lines indicates the network I/O throughput and CPU utilization for the network I/O.<br>
One weird thing is the L4Linux CPU utilization. It spent much less CPU cycles than the native Linux did.<br>
<br>
In my humble opinion, it is normal case because the L4 Fiasco schedules
the L4Linux application (netperf) as a L4 process and L4Linux kernel
doesn't have correct CPU usage information of the application.<br>
<br>
Is it correct?<br>
<br>
<br>