Re: Hardware benchmark list (was: Recommended hardware for iptables based firewall/router)

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Sun, Nov 9, 2014 at 4:08 PM, Michael Schwartzkopff <ms@xxxxxxx> wrote:
> Am Sonntag, 9. November 2014, 15:39:35 schrieb shawn wilson:
>> (I include the original email subjec to show what spawned this)
>>
>> What would be cool is to have a hardware benchmark list for network
>> cards and processors (similar to JTR's list for hashes per second
>> [1]).
>>
>> What I've been thinking for a while is how to benchmark network
>> hardware. Obviously there's issues with doing this and in order for
>> this to work everyone would need to be on the same page (or there'd
>> need to be a link showing settings used that someone could revise if
>> they found something that was off).
>>
>> I think the biggest issue with this idea is figuring out affordable
>> hardware to test cards with. I'm unsure if this is possible so just
>> figured I'd throw this out there. Though since most machines have two
>> stacked interfaces, a decent benchmark might just be to plug one
>> interface into the other as that should shoe hardware noise issues,
>> PCI bus limitations (on that board), and driver issue.
>>
>> 1. http://openwall.info/wiki/john/benchmarks
>
> Cool idea. Anyway some work to do:
>
> 1) You (we) would have to define some standard tests (paket size, size
> distribution, ...), some defined testbed, ...
>

Yes, though this wouldn't need to be static - it could be changed in
sysctl on the fly (I think - without taking down an
interface/reloading the driver?)

> 4) Add tests for IPsec encryption. I did some tests for VIA Eden processor:
>

Hadn't even thought about IPsec - though I'm sure there are other
things that can be offloaded into hardware or kernel space drivers
that we'd want to test. (Nice short posts you have too - thanks)


I am however mainly concerned with the hardware aspect of doing this -
if plugging one stacked if into another works most of the time, the
rest should just be simple scripting/util choices (and a home for the
stats).
--
To unsubscribe from this list: send the line "unsubscribe netfilter" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html




[Index of Archives]     [Linux Netfilter Development]     [Linux Kernel Networking Development]     [Netem]     [Berkeley Packet Filter]     [Linux Kernel Development]     [Advanced Routing & Traffice Control]     [Bugtraq]

  Powered by Linux