Re: [6lowapp] Charter and transports

[Lloyd Wood <[e-mail]>]

On 3 Nov 2009, at 17:09, Pascal Thubert (pthubert) wrote:

> Hi Lloyd,
>
> Glad you mention it.
>
> Is there a way to see Saratoga without the 'DTN' piece?

The DTNRG bundle support is specified in a separate DTNRG draft, which  
can
be ignored entirely, along with the single bit saying 'carrying  
bundles'.
That's unimportant.

http://tools.ietf.org/html/draft-wood-tsvwg-saratoga
specifies only Saratoga and how it does file delivery and streaming.

(As an aside, I'm not convinced by the utility of DTN bundling;
doing HTTP over a variety of transports including TCP to make
it useful for unusual environments seems promising, though.)


> Could you elaborate a bit on how flow control is done in Saratoga?

If you mean congestion sensitivity, it's entirely open.
There's sufficient information flowing between sender
and receiver for a variety of schemes to be implemented, as
http://tools.ietf.org/html/draft-wood-tsvwg-saratoga
mentions - LEDBAT, TCP friendliness, etc. Pick a scheme
that suits; in many private dedicated links where link
utilization is a concern, and losses are due to channel
errors and not congestion, one might choose to run at link rate
for fast transfers.

Otherwise - HOLESTOFILL packets are sent by the receiver
or requested by the sender, and include necessary selective
negative acknowledgement (SNACK) information for reliable
file transfer or reliable streaming.

L.

>
> Cheers,
>
> Pascal
>
>> -----Original Message-----
>> From: [e-mail] [mailto:[e-mail]] On
> Behalf Of Lloyd Wood
>> Sent: mardi 3 novembre 2009 17:57
>> To: [e-mail]
>> Subject: Re: [6lowapp] Charter and transports
>>
>>> Obviously UDP is something we will need to support as a transport  
>>> for
>>> CoAP, and I argue that we also need to allow the use of CoAP over  
>>> TCP
>>> as this is useful for some applications. When using UDP, some simple
>>> mechanism for reliability (e.g. stop-and-wait) would of course be
>>> needed along with a transaction ID (see e.g. 6lowapp-frank-chopan).
>>> When doing this we should of course cooperate with TSV people.
>>
>> http://tools.ietf.org/html/draft-wood-tsvwg-saratoga
>>
>> is such a protocol, providing reliability and IDs.
>>
>> L.
>>
>> DTN work: http://info.ee.surrey.ac.uk/Personal/L.Wood/saratoga/
>>
>> <http://info.ee.surrey.ac.uk/Personal/L.Wood/><[e-mail]>