On Saturday, November 2, 2013, RANDRIAMASY, SABINE (SABINE) wrote:

>  Hi Richard and Qin,
>
> I agree that even a 10 minutes slot will be focused on the selection and
> specification of TE based ALTO Costs and on the next steps to this draft.
> The Cost schedule is a specific Cost Mode that is well suited to TE based
> metrics that usually change frequently and are a major use case. We may
> thus point this out as related discussion topic that may continue on the
> mailing list.
>
>
Personally, I feel that metric definitions and "schedule" will be better
to be defined in different drafts because they are different topics. More
generally, I see your "schedule" as a starting point for a powerful API
into potentially a large amount of metric data that may be available by a
network. The data can be based on both historically collected and/or
planned data (e.g., scheduled maintenance). The metrics may or may not
change rapidly---many diurnal patterns are slow enough for a
slow-path software controller to adapt to. In short, a sense that we are
defining a simple but general API in the statistics/prediction domain will
be very helpful to move forward.

Thanks!

Richard


Best regards,
>
> Sabine
>
>
>
>
>
> *De :* [email protected] <javascript:_e({}, 'cvml',
> '[email protected]');> [mailto:[email protected]<javascript:_e({}, 
> 'cvml', '[email protected]');>]
> *De la part de* Qin Wu
> *Envoyé :* dimanche 3 novembre 2013 02:14
> *À :* Y. Richard Yang
> *Cc :* IETF ALTO
> *Objet :* [alto] 答复: I-D Action: draft-wu-alto-te-metrics-00.txt
>
>
>
> I thought we got 10 min time slot.
>
> Sure, we can take offline if out of time.
>
> Sabine and I have already had some offline discussion and list discussion
> on the list when we prepare the slides for this draft.
>
> I did agree to add this feature but apparently as an option feature.
>
> I am also available before ALTO or after ALTO for discussing this.
>
> It will be great we can talk on Sunday.
>
>
>
> Regards!
>
> -Qin
>
> *发件人**:* [email protected] [mailto:[email protected]] *代表* Y.
> Richard Yang
> *发送时间**:* 2013年11月3日 8:39
> *收件人**:* Qin Wu
> *抄送**:* IETF ALTO
> *主**题**:* Re: [alto] I-D Action: draft-wu-alto-te-metrics-00.txt
>
>
>
> Hi Qin, Sabine,
>
> On Oct 31, 2013 5:48 AM, "Qin Wu" <[email protected]> wrote:
> >
> > Hi, Richard, Sabine:
> >
> > I think it is a very good idea to discuss schedule of TE metric as open
> issue to draft-wu-alto-te-metrics in ALTO session.
>
> Given potential extended discussions needed, I am not optimistic that your
> 5-min slot can accommodate a discussion on introducing schedule.
>
> I liked the idea of schedule as I see clear use cases. Google Map goes
> back and forth with its predicted traffic mode. See this thread
> http://0-productforums.google.com.library.ccbcmd.edu/forum/m/#!topic/maps/iVaKdxUhuMw
> <
>


-- 
-- 
 =====================================
| Y. Richard Yang <[email protected]>   |
| Professor of Computer Science       |
| http://www.cs.yale.edu/~yry/        |
 =====================================
_______________________________________________
alto mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/alto

Reply via email to