Preliminary results from testing..

[Randall Stewart]

Here are the graphs..

Drew tells me they got stripped off :-)



http://people.freebsd.org/~rrs/c35.png
http://people.freebsd.org/~rrs/c36.png


R
On Dec 9, 2015, at 8:13 AM, Randall Stewart <rrs at netflix.com> wrote:

> All:
> 
> Ok so I have had a first cut of an A/B test with a machine using fixed calls to tcp_do_segment/output etc and
> one using the new pluggable pointer based stack. As expected I can’t really tell any real difference between
> the two machines.
> 
> They are both Intel Ivy Bridge E5-2650L Intel CPU’s. Running at 1.7Ghz, 64G of memory and 10 physical cores.
> 
> Here are png’s of the per second cpu use and the overall serving bandwidth (if anyone wants the raw numbers
> let me know).
> 
> The two machines are basically the same, serving the same content.
> 
> I have a few other pairs I will be trying this on.. though I don’t expect any real difference.
> 
> Robert, how are you coming on investigating the overhead of function pointers on non-intel devices ?
> 
> I would like to commit the tcp-stack changes soon if possible :-)
> 
> R
> 
> 
> <c35.png>
> <c36.png>
> --------
> Randall Stewart
> rrs at netflix.com
> 803-317-4952
> 
> 
> 
> 
> 

--------
Randall Stewart
rrs at netflix.com
803-317-4952