Re: [mif] about draft-hui-ip-multiple-connections-ps-02

[Min Hui <[e-mail]>]

Hi,

pls see inline reply. thx.

2009/3/19 marcelo bagnulo braun <[e-mail]>:
> Min Hui escribió:
>>
>>>
>>>>
>>>>    2.2.3. TOS consideration
>>>>
>>>>  TOS can be a parameter of routing item to indicate which kind of IP
>>>>  data is suitable to deliver by this routing. The multiple interfaces
>>>>  will connect to multiple access networks, so that the preference of
>>>>  TOS need to be merged to have a better performance of data delivery.
>>>>  For example, the WiFi access could indicate itself has broader
>>>>  bandwidth comparing with 2G access, and set the TOS as broad
>>>>  bandwidth. When another interface connects the Ethernet, the TOS is
>>>>  also set as broad bandwidth preferred. In this situation, it needs
>>>>  some mechanism to merge and reorder the TOS getting form multiple
>>>>  interfaces.
>>>>
>>>> I am not sure how widely used is this...
>>>>
>>>> A: As depicted in RFC1122, TOS is a necessary parameter to choose a
>>>> suitable routing. And different applications have different
>>>> requirements of access network, which can be reflected as TOS. So we
>>>> think TOS will be useful in multiple interface situations.
>>>>
>>>>
>>>>
>>>
>>> right, my question was how widely used is this. I mean how much of the
>>> internet traffic actually sets the TOS?
>>>
>>
>> The 3G network use TOS as one of the QoS parameters, and we suppose to
>> use it in the multiple interfaces situation.
>>
>>
>
> ok, but does any other network uses this?
> I mean, if only 3g networks use this, this is only a problem when
> simultaneously connecting to multiple 3G networks, so i am not sure how
> common this is
There are 3GPP and 3GPP2 mobile network which have been deployed, both
of them support it.
Some network use it for Wifi access network as well.
There are more and more motivation to do different service quality
categorization,
such as put the p2p more low level.

>
> ....
>>>>
>>>>  5 Network side should be capable of distributing the IP flow
>>>>  according to some parameters, such as IP address prefix, network type
>>>>  and so on.
>>>>
>>>> i don't understnad what you mean by this one, could you expand?
>>>>
>>>> A: That is the policy in the network side. Corresponding to the policy
>>>> of sending data mentioned in the fourth bullet, the policy of
>>>> receiving data is also needed, which can be apply in the network side.
>>>> The network can determine forward a specific IP flow to which
>>>> interface of the destination host according to the policy.
>>>>
>>>>
>>>>
>>>
>>> right, i see what you mean. I am not sure which element of the network
>>> would
>>> do that... i mean, are you assuming that all interfaces are being
>>> connected
>>> to the same ISP? If not, i am not sure how would you do this...
>>>
>>
>> In technical perspective,  the element would be the gateway which in
>> charge of all the interfaces, e.g. PDN GW in 3GPP architecture, LMA in
>> PMIP domain; In deployment perspective, it's better when all
>> interfaces connect to the same ISP, the negotiation is needed when
>> multiple ISP are involved.
>>
>
> but it is not the network who can decide the IP address to use to forward
> packets....
>
> I mean, suppose that you have a node A in the internet that wants to send a
> flow to a mif device.
> Let's suppose that the MIF device has different IP addresses per interface
> Now, the selection of the interface is perfromed by node A when it selects
> the destiantion address it will send the flow to. I  fail to see how the
> netwokr would be able to send packet to a different address (wihtout using
> MIP, of course)

I agree with you, if we don't consider MIP and PMIP, the requirement
of network side policy is not essential.

>
> What is your view on this?
>
> Regards, marcelo
>>>>
>>>> Regards, marcelo
>>>>
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> mif mailing list
>>>> [e-mail]
>>>> https://www.ietf.org/mailman/listinfo/mif
>>>>
>>>>
>>>>
>>>
>>>
>>
>>
>
>