Force10 Gear - Opinions

Jo Rhett jrhett at netconsonance.com
Thu Sep 11 07:40:01 UTC 2008


On Sep 5, 2008, at 12:37 PM, Paul Wall wrote:
> Jo Rhett wrote:
>> Note the "not random" comment.  People love to use the random  
>> feature of ixia/etc but it rarely displays
>> actual performance in a production network.
>
> Once upon a time, vendors released products which relied on CPU-based
> "flow" setup.  Certain vintages of Cisco, Extreme, Foundry,
> Riverstone, etc come to mind.  These could forward at "line rate"
> under normal conditions. Sufficient randomization on the sources
> and/or destinations (DDoS, Windows worm, portscans, ...) and they'd
> die a spectacular death.  Nowadays, this is less of a concern, as the
	...
> Either way, I think it's a good test metric.  I'd be interested in
> hearing of why you think that's not the case.  Back on topic, doing a

Yes.  And those problems were fixed in most gear.  What I found *also*  
was that the flow tables tended to fill up, and a lot of gear thrashes  
on the flow tables.  You need real bi-directional sessions to create  
the effect properly in many cases.  (ie Extreme, which handles random  
fine but bidirectional flows proved that too much of the work was  
being done in software)

>> I have a current spreadsheet here, and trust me your math went wrong
>> somewhere.  A completely full chassis is only a bit more than what  
>> you are
>> ...
>> But no, I'm not going to redo the math.  I'm not a F10 salesperson  
>> and I
>> have much more important things to do right now.
>
> I'd be interested in seeing where I went "wrong", in the interest of
> setting the record straight.  The original poster was interested in
> how Force 10 stacks up against the competition from a feature and
> price prospective.  He deserves some cold science, and I'm trying to
> help him out.

I meant what I said, and I wasn't trying to be rude.  There are F10  
people on this mailing list, it would serve you to engage them instead  
of me.  I'm quite happy with my Force10 units but I'm not making any  
commission selling them and I have too much to do to be doing someone  
else's job.

> To wit, you said F10 is cheaper than a comparable Cisco 6500 (in a
> basic gig-e configuration).  I demonstrated that's not the case.  You
> responded with ad-hominem attacks, followed by indifference, and
> later, claims of emotional distress; still you refuse to provide any
> hard numbers, claiming it's "not your job".  Where I come from, people
> like that are referred to as sore losers. :)


You're reading a lot more into it than I bothered to think about it.   
I've done the math repeatedly, and Force10 always comes out cheaper  
than Cisco in that scale of port density.  Your numbers looked off to  
me, but letting you know the previous sentence is about all the time I  
can spend on this topic.  Can we kill this now?  Thanks.

-- 
Jo Rhett
Net Consonance : consonant endings by net philanthropy, open source  
and other randomness






More information about the NANOG mailing list