[lemonade] Asymmetry of limited capability access networks

[Eric Burger <eburger at bea.com>]

Just a general comment:

Many access networks have asymmetric bandwidth constraints.  Examples are
ADSL and WiMax.  Thus, if all things are equal, bytes on the down-link
(S->C) being traded for bytes on the up-link (C->S) is generally a good
thing.


For the specific case here, 30 bytes at 50kb/s is 6ms (including overheads).
Do we really care?


On 2/12/07 9:34 AM, "Lingyan Wu" <wulingyan_48795 at huawei.com> wrote:

> 
> ----- Original Message -----
> From: "Arnt Gulbrandsen" <arnt at oryx.com>
> To: <lemonade at ietf.org>
> Cc: "Lingyan Wu" <wulingyan_48795 at huawei.com>; "Randall Gellens"
> <randy at qualcomm.com>
> Sent: Monday, February 12, 2007 3:03 PM
> Subject: Re: [lemonade] draft-gulbrandsen-imap-notify-01 use cases
> 
> 
>> Lingyan Wu writes, answering me:
>>>> The STATUS commands are necessary (in this particular case at least) to
>>>> be notified of changes that happened while the client was offline.
>>> 
>>> Two small methods to save bandwidth as follows:
>>> 
>>> 1, If STATUS commands (which are sent after starting NOTIFY) contain
>>> UIDVALIDITY and HIGHESTMODSEQ (catched in client) of related mailboxes ,
>>> after receiving the STATUS commands the server can determine which
>>> mailboxes are out of sync and send STATUS(HIGHESTMODSEQ) responses of the
>>> mailboxes to client.
>>> E.g. The client sends 1000 STATUS commands containing UIDVALIDITY and
>>> HIGHESTMODSEQ immediately after starting NOTIFY, the server determines
>>> and sends only a few STATUS(HIGHESTMODSEQ) to client for those mailboxes
>>> out of sync.
>> 
>> I don't see real savings. As far as I can tell, you're adding 30 bytes to
>> the status command in order to save 30 bytes on the response.
> 
> Yes. You are right.
> 
>> 
>>> 2, After receiving Notify command, the server sends unsolicited  STATUS
>>> responses for all of the mailboxes to the client.
>> 
>> For all of the mailboxes where NewMessage events would be sent, you mean?
>> Maybe that's a good idea. Not 100% sure.
> 
> For all of the mailboxes.
> I meant the server sends STATUS responses for all of the mailboxes without
> receiving large numbers of STATUS commands first. After receiving the
> unsolicited STATUS responses, the client can determine which mailboxes are
> out of sync.
> 
>> 
>> Arnt
>> 
> 
> 
> 
> _______________________________________________
> lemonade mailing list
> lemonade at ietf.org
> https://www1.ietf.org/mailman/listinfo/lemonade

_______________________________________________________________________
Notice:  This email message, together with any attachments, may contain
information  of  BEA Systems,  Inc.,  its subsidiaries  and  affiliated
entities,  that may be confidential,  proprietary,  copyrighted  and/or
legally privileged, and is intended solely for the use of the individual
or entity named in this message. If you are not the intended recipient,
and have received this message in error, please immediately return this
by email and then delete it.

_______________________________________________
lemonade mailing list
lemonade at ietf.org
https://www1.ietf.org/mailman/listinfo/lemonade
Supplemental Web Site:
http://www.standardstrack.com/ietf/lemonade