Dear authors,


Your work on cost calendar makes a lot of sense to me, and here are some 
comments I have.



Issues:

1. In the 4th paragraph of section 1 (Page 4), here is supposed to introduce 
the advantages of cost calendar. I would like to add another significant 
advantage which is the reduction of the frequent information exchange between 
ALTO client and ALTO server. And I think it may be the salient benefit that 
leads to saving the network resources for ALTO protocol.

2. In section 2.2.1, the word “time-sensitive” comes up several times, but what 
is the exact definition for “time-sensitive”? Maybe I’m wrong, but I reckon it 
is used to describe the data for real-time service. I suggest the authors can 
give a definition and examples here, or provide the references where the word 
is interpreted.

3. I think maybe the authors missed a point which many audiences are concerned 
about. When the network state is getting changed by e.g., link failures, node 
crashed, and if there is a conflict between the cost calendar that has already 
been conveyed and the current new cost map, then how the client can be aware of 
it? The ALTO server needs to notify the clients before they schedule data 
transfers according to the old cost calendar.



Typos:

1. In the first line of section 3.1, there is a redundant space before the 
comma.

2. At the end of the 3rd paragraph of section 3.1,  “ressources” ==> 
“resources”.



Hope these are helpful.



Best, Geng

Computer Science, Yale
_______________________________________________
alto mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/alto

Reply via email to