kernel vs. Debian/Red Hat/SuSE e.t.c.?

Robert Wilson bwilson4web at hotmail.com
Thu Apr 26 13:32:46 CDT 2007


As I pointed out, I'm not out for any sort of release wars.

. . .

>>My current kernel is 2.6.20-7 and I see 2.6.21 has just been released. 
>>I'll
>>have 2.6.21 built and tested in probably a couple of days. But I get the
>>impression that Red Hat is a further back and have no idea of the other
>>common Linux packages.

>Correct observation, . . .. Is there something in particular which you need 
>from a current released Debian kernel that's not present on a currently 
>released RedHat kernel?

A fair question. My application is network testing, primarily using iPerf to 
stress test routers, firewalls and circuits. This is very I/O intensive and 
the performance limitations are a function of: (1) network interface card, 
(2) device driver, (3) CPU/IO architecture, (4) kernel dispatching and (5) 
task dispatching.

For performance that begins to match 'wire speed,' I have to baseline and 
test everything built from source code to get every compiler and processor 
and operating system optimization available. My task often includes  use of 
code profile and other real-time debugging techniques to performance tune my 
systems. I have no choice if we're going to get usable metrics.

In the recent kernel development area, there has been a lot of good work in 
SMP, hyperthreading and preemptable kernel as well as more recent device 
drivers that improve the performance of small packet processing. The primary 
application, iPerf, also runs better if when the generated instructions use 
code optimized for the processor. So I tend to hang out 'at the bleeding 
edge.' I haven't been rebuilding the common libraries but none of my profile 
testing has pointed to them as making a significant performance impact.

Thanks for pointing out that the if-up/down scripts are independent of the 
kernel. I was afraid they might have been more closely tied to the kernel 
version. I had noticed a number of paths showing up in some of the kernel 
build scripts and was concerned.

I can sympathize with those who have a large application base they must 
support. For them, stability is paramount. But when trying to get to 'wire 
speed,' I need to be able to use the "latest hotness."

Bob Wilson

_________________________________________________________________
Need a break? Find your escape route with Live Search Maps. 
http://maps.live.com/default.aspx?ss=Restaurants~Hotels~Amusement%20Park&cp=33.832922~-117.915659&style=r&lvl=13&tilt=-90&dir=0&alt=-1000&scene=1118863&encType=1&FORM=MGAC01



More information about the Linux-PowerEdge mailing list