Search Mailing List Archives


Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

[mininet-discuss] the maximum number of switches can be simulated

Alexander Turner alexanderturner at me.com
Tue Apr 15 20:05:33 PDT 2014


I believe the limit you’ll be encountering here is the performance of the computer you’re running mini net on.

There is a project called Maxinet that allows you to create multiple mininet instances on different machines and link them together using GRE tunnels. I believe maxinet also provides a central point of orchestration for the different mininet instances.

I could be wrong here and it could be a limitation with your controller as well. Try different controllers such as NOX.

AT


On 16 Apr 2014, at 11:46 am, Ke Liu <poppin.ke at gmail.com> wrote:

> Thanks for the reply
> 
> I create tree topo with depth=9 fanout=2 thus 511 switches but only the first 249 switches connecting to the controller (beacon controller)
> 
> 
> 
> Bob Lantz <rlantz at cs.stanford.edu>于2014年4月16日星期三写道:
> How many switches, hosts and links are you trying to create?
> 
> On Apr 14, 2014, at 11:42 PM, Ke Liu <poppin.ke at gmail.com> wrote:
> 
>> Hi, All, 
>> 
>> I create a tree topology that creates more than 500 switches, but our controller only receives the TCP connection requests  from only 247 switches, I saw the previous post that also raised the similar problem, they could not simulate 1000 switches using Mininet, I want to ask whether this problem is solved or not right now, or any direction to solve it 
>> 
>> Thanks
>> 
>> -- 
>> Liu, Ke, Ph.D
>> Assistant Professor
>> State Key Laboratory of Computer Architecture
>> Institute of Computing Technology,Chinese Academy of Sciences
>> Phone: +86-13810814278
>> Email: liuke at ict.ac.cn
>> Website: http://acs.ict.ac.cn/network-system/liuke/
>> _______________________________________________
>> mininet-discuss mailing list
>> mininet-discuss at lists.stanford.edu
>> https://mailman.stanford.edu/mailman/listinfo/mininet-discuss
> 
> _______________________________________________
> mininet-discuss mailing list
> mininet-discuss at lists.stanford.edu
> https://mailman.stanford.edu/mailman/listinfo/mininet-discuss

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.stanford.edu/pipermail/mininet-discuss/attachments/20140416/227d6df9/attachment.html>


More information about the mininet-discuss mailing list