[ovs-discuss] in-kernel flow table in the datapath

Masoud Moshref Javadi masood.moshref.j at gmail.com
Thu Apr 5 19:16:53 PDT 2012


I have the following experiment and whenever I check ovs-dpctl 
dump-flows, it gives me less than 7k entries:
There are three machines A,B, and C. B runs openvswitch and connect the 
other two. A and B send 10k packets each second from different flows to 
the other node. I don't use TCP so there is no ack.  There should be 20k 
entries there after a second but I get at most 7k and the value changes. 
It seems that the facet cache cannot hold enough facets, so it replaces 
them so often. The high cpu usage for ovs-vswitchd process also confirms 
this claim that the cache cannot hold all entries.

suppose 100k new flows (however, because of the previously mentioned 
claim, it does not matter).

On 4/5/2012 5:41 PM, Jesse Gross wrote:
> On Thu, Apr 5, 2012 at 5:26 PM, Masoud Moshref Javadi
> <masood.moshref.j at gmail.com>  wrote:
>> What is the maximum size of the kernel table?
> There's no hard limit but the number of flows is regulated by
> userspace to avoid excessive overhead.  This is generally in the range
> of a few thousand flows.
>
>> What will happen if I have
>> 100k flows per second, will they be dropped? queued?
> 100k flow setups per second?  Or 100k sustained flows?



More information about the discuss mailing list