Search Mailing List Archives
[mininet-discuss] DOT: A Distributed OpenFlow Testbed
hohoangphi at gmail.com
Thu May 14 01:39:25 PDT 2015
Thank you for the clarification.
On Thu, May 14, 2015 at 4:29 AM, Nicholas Bastin <nick.bastin at gmail.com>
> On Wed, May 13, 2015 at 3:08 PM, Phiho Hoang <hohoangphi at gmail.com> wrote:
>> I am not as much concerned with the through put as with how to handle the
>> massive number of hosts, sub-nets and datapaths in the emulation using
> If you use mininet hosts this gets more interesting - you are of course
> eating up a different set of system resources and have to consider that.
> Otherwise caring about things like subnets doesn't matter because datapath
> interfaces obviously don't have that kind of metadata.
>> (I plan to use dozens of cheap CHIP, the Raspberry Pi Killer ;-)
> We are getting some this summer to do some kernel hacking on the A13 for
> SDN-ish purposes, but I'm not sure how well they'll work for you here -
> clustering them would involve using the wifi interface (as there's no wired
> ethernet), and "dozens" in any close proximity are likely to be an RF
> nightmare. The raspberry pi 2 is already a pretty great small-scale
> solution for this problem (although still only 100Mbit ethernet).
>> Did you find any need for DHCP, DNS, routers... when you work with
>> Mininet at a massive scale across multiple servers so that all Mininet
>> hosts in the emulation can communicate with one another conveniently?
> We don't use mininet "hosts" in the emulation (we use other distributed
> hardware connected to the emulation environment to provide the end hosts,
> either as VMs, containers, or hardware generators, depending on the test).
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the mininet-discuss