I think it over and found it is my bad to have clouded this discuession with
many OS discuession. So allow me to simplify my question:
If there are vswitchs good enough and no special resource requirement for users
to install; and if there are virt-wires for users to build their own overlayed
network constructions. Users can help themselves with these sort of things. Why
CSPs are so deeply involved into this kind of affairs?
If any user can pay for whatever vswitchs they liked, instead of let CSPs
choose one together with a full-scale solution. If different small sized
virt-net devices can work thgether inside a big data-center why only a handful
of vswitches exist?
That is the real thing puzzled me.
On Sunday, June 15, 2014 12:27 PM, Shen Li <[email protected]> wrote:
Hi,
As far as I remembered DPDK have lots of difference from the thing I talked
about. But I am already expecting some take them as same ,so bingo! :)
Instead of talking about them specificly (because a vswitch's forum is not an
ideal place to do so), I would like to ask a that someone suggest me an example
of VM's usermode SMP openvswitch example works as fast as It does in host's
kernel mode, and they live as a nice neighbour as a common process when it
doing heavy realtime jobs. Which means a
switch proforms as a real box is as "chip" as a common process to be deployed
by any user.
And as the name suggested ,DPDK is a DP's DK, I am not talking about composing
another one of them. So actrually they are not even close.
You are right about Virt-wire can be implemented with encapsulation. It still
puzzled me that virt net have not yet formed as an isolated technique, instead
of being part of some Cloud data center network solutions. So by deploying it
,users can by itself solve many problems. I don't understand why things is not
being so.
On Friday, June 13, 2014 3:00 AM, Wes Felter <[email protected]> wrote:
On 6/12/14, 3:12 AM, Shen Li wrote:
> I worked on OS software, and know little about networking
> virtualization. I have a question. Assuming the following technologies
> come true, then we can let users install their own choice of
> vswitches(or such kind of "devices") in their own VM, without risk to
> influence CSP's hypervisor software; and we can allow users to build
> their own network construction overlaying upon the existing net, give
> them more flexibilities and control.
Most of this has already been done in the NFV and GENI worlds. Both
Snabb and DPDK are implementing low-overhead paths into VMs. Virtual
wires can be implemented with encapsulation.
--
Wes Felter
IBM Research - Austin
_______________________________________________
discuss mailing list
[email protected]
http://openvswitch.org/mailman/listinfo/discuss
_______________________________________________
discuss mailing list
[email protected]
http://openvswitch.org/mailman/listinfo/discuss
