Re: [clouds] draft-yokota-cloud-service-mobility

[Vishwas Manral <[e-mail]>]

Hi Hidetoshi-san,

I would think TCP is a good option for this. PCE uses TCP too.

Thanks,
Vishwas

2011/1/13 Hidetoshi Yokota <[e-mail]>:
> Hi Vishwas,
>
> Thanks again for your good input.
>
> (2011/01/12 11:17), Vishwas Manral wrote:
>> Hi,
>>
>> Some other things I can think of are NACK's and retransmission of
>> messages when no message is received.
>>
>> Also I assume the protocol will probably work over TCP. A prot number
>> needs to be defined for the same.
>
> We haven't decided the transport protocol, yet. If it is TCP or SCTP,
> then the retransmission procedure can be left to them, which I think
> would be simpler than employing UDP. NACK-like message is definitely
> needed when, for example, the requested operation cannot be performed.
> Maybe, we should define the status code in the ACK message to convey
> more detailed information. Port number is also definitely needed and
> this I-D will request IANA to assign a specific one at a later stage.
>
> Regards,
> --
> Hidetoshi
>
>> Thanks,
>> Vishwas
>>
>> On Tue, Jan 11, 2011 at 5:26 PM, Vishwas Manral<[e-mail]>  wrote:
>>> Hi,
>>>
>>> I looked at the document and there are a few very basic things I
>>> wanted to state that need to be added:
>>>
>>> 1. There needs to be a capability exchange from the Execution node to
>>> the Manager node.
>>> 2. Scalability issues will occur if keepalives all go to the manager
>>> node. In my view there can be a heirarchy of keepalives.
>>> 3. There should be a heirarchy of manager nodes too, considering the
>>> number of Execution nodes that need to be managed. So there should be
>>> a messaging exchange allowed between Manager and Manager node.
>>> 4. All TLV and headers should have length of 16 bits atleast. 8 bits
>>> is not scalable at all with the amount of information that is there.
>>> 5. There needs to be authentication in the packets to provide some security.
>>> 6. There needs to be async messaging allowed both from server to
>>> client and client to server.
>>> 7. There is already a PCE framework that exists. We need to look at
>>> it. It is used for simialr purposes in a TE network.
>>>
>>> Thanks,
>>> Vishwas
>>>
>>
>>
>>
>
>