On Tue, Aug 2, 2016 at 8:31 PM, Xavier Vilajosana <
[email protected]> wrote:

> Hi Jiliang,
>
> inline again :)
>
> 2016-08-02 13:58 GMT+02:00 Jiliang Wang <[email protected]>:
>
>> Thanks. please see inline my comments.
>>
>> On Tue, Aug 2, 2016 at 7:29 PM, Xavier Vilajosana <
>> [email protected]> wrote:
>>
>>> Hi Jiliang,
>>>
>>> see inline please,
>>>
>>> 2016-07-29 15:33 GMT+02:00 Jiliang Wang <[email protected]>:
>>>
>>>> Thanks very much for your response. My comments are as follows.
>>>>
>>>> On Tue, Jul 26, 2016 at 9:14 PM, Xavi Vilajosana Guillén <
>>>> [email protected]> wrote:
>>>>
>>>>> Dear Jiliang,
>>>>>
>>>>> see inline please,
>>>>>
>>>>> 2016-07-26 10:05 GMT+02:00 Jiliang Wang <[email protected]>:
>>>>>
>>>>>> Thanks for the great work. This is Jiliang Wang from Tsinghua
>>>>>> University. I am very interested in this version and I just have several
>>>>>> questions.
>>>>>>
>>>>>> 1.       In 4.3.10, it says “If node A requests more cells than can
>>>>>> fit in the response, node B MUST return code RC_ERR_NORES and an empty 
>>>>>> cell
>>>>>> list.”
>>>>>>
>>>>>> 1.- Is there any fragmentation method to transmit all the cells if
>>>>>> the requested cells cannot be fit in a response?
>>>>>>
>>>>> No, we keep it simple. The LIST command supports pagination. You can
>>>>> retry with a smaller page size if it does not fit in a packet.
>>>>>
>>>> Is the pagination decision made by node A? This seems to be not simple
>>>> since the actual number is not know and thus it is difficult to determine
>>>> whether to use pagination.
>>>>
>>>
>>> A node knows the bytes it can fit in a packet. This gives an indication
>>> to the node of what is the number of items that can be requested. When a
>>> node issues a LIST it starts by offset 0, this returns the first N
>>> scheduled cells. Then it seems reasonable that the node will indicate
>>> offset N and request N more cells ...
>>> If the number of requested cells does not fit in the packet an error is
>>> returned and the node retries again with a smaller value of N.
>>> I do not see a big complexity on that. What issues do you see?
>>>
>> Then how to deal with the case that a node has cells more than that can
>> be fitted in a packet. Is there any method to return all those cells?
>>
>
> that is why we use pagination. Assume the command LIST_AB issued by node A
> to node B. Assume B has 10 cells in its schedule being
> (1,3,5,7,9,11,13,15,17,19), assume also that a packet can only transport 4
> cells.
>
> Node A can issue STATUS to Node B. B will respond ti has 10 cells.
> then
> Node A will issue
> LIST_AB offset=0 num cells=4
> Node B will respond
> 1,3,5,7
>
> Node A then will issue
> LIST_AB offset=4 num cells=4
> Node B will respond
> 9,11,13,15
>
> Node A then will issue
> LIST_AB offset=8 num cells=2
> Node B will respond
> 17,19
>

Then why node A should request 3 times? Should node B respond with all
those cells in 3 packets after receiving one request? This makes the
operation easier and more clear.
Semantically,  this means that A should not care about how many times
should request to return all cells.

>
>
> 2.       Will the clear command result in inconsistency in current
>>>>>> design? For example, a clear command is sent from A to B and the response
>>>>>> from B to A is lost.
>>>>>>
>>>>> The CLEAR is not executed until the response is received. B side
>>>>> executes the CLEAR after receiving the L2 ACK of the response message.
>>>>>
>>>> It there any problem when the L2 ACK is lost in which A clears while B
>>>> does not?
>>>>
>>>
>>> In a TSCH network the probability of losing the ACK is very low. As the
>>> packet has already been sent the ACK has extremely high chances to pass as
>>> well (it is shorter in length which still favours the transmission
>>> success). Of course this can remotely happen and if this happens the system
>>> should recover from that inconsistency. To do so the generation counters
>>> will enable the detection of the inconsistency and both nodes schedule will
>>> be cleared (only cells between them).
>>> The alternative is to use a 3 step transaction, which requires 33% more
>>> packets over the air at every transaction. We decided to avoid that. Maybe
>>> a research question is to evaluate in a real deployment the energy overhead
>>> of using 3 way transactions vs eventually recovering from a lost ACK during
>>> a CLEAR. Given the chances of losing an ACK I would bet for the later.
>>>
>> Yes, agree. I also agree that the probability should be low and 3-way
>> handshaking may be overwhelming in this case. As long as clear can be used
>> to deal with the inconsistency (with low probability), the protocol should
>> be ok.
>>
>>>
> :)
>
>>
>>>
>> 3.       Currently, the add command supports best-effort cell adding,
>>>>>> i.e., when the cells is not enough, partial of the cells can be added – 
>>>>>> if
>>>>>> I understand correctly.
>>>>>>
>>>>>>  –what is the purpose of best-effort adding. Why not just return
>>>>>> success and fail?
>>>>>>
>>>>> Best effort minimizes the amount of data to be sent.
>>>>>
>>>> Is it possible that a node needs to issue an all-or-no adding request?
>>>> For example, a node prefers to transmit all data at once.
>>>>
>>> I would separate the data plane from the control plane. For me ADD is an
>>> operation that deals with the reservation of cells (control plane). If ADD
>>> does not meet the requirements in a particular moment in time, the SF will
>>> take care of requesting other cells. If this introduces a delay between
>>> when the cells are requested and when all of the requested cells can be
>>> used is something we need to evaluate.  There is always the option that the
>>> application requesting the bw blocks until all cells have been allocated.
>>> This is out of the scope in 6top.
>>>
>> My question is that A node may want to allocate, say 6 cells, B node
>> should return 0 or 6 instead of any other number in between. Otherwise, the
>> upper layer scheduler SF cannot schedule the communication in a fine
>>  granularity. For example, SF may realized 6 cannot be satisfied and it may
>> decide to wait instead of using partial of the cells.
>>
>
> I see. I wonder if this case is something we should worry. Usually the
> schedules are sparse in these networks. If this eventually happen maybe we
> can tolerate the SF to delete the cells and wait. What alternative would
> you propose? and what would be the benefit vs overhead/drawback vs a
> best-effort approach in the most common cases?
>
This may depends on the normal upper layer  operations. Is satisfying
partial requests the common case?

>
>
>>> kind regards,
>>> Xavi
>>>
>>>>
>>>>>
>>>>>
>>>>>> best,
>>>>>>
>>>>>>
>>>>> Kind regards,
>>>>> Xavi
>>>>>
>>>>>> -------------
>>>>>> Jiliang Wang
>>>>>> Assistant Professor
>>>>>> http://tns.thss.tsinghua.edu.cn/~jiliang/
>>>>>>
>>>>>> On Tue, Jul 26, 2016 at 2:32 PM, Xavier Vilajosana <
>>>>>> [email protected]> wrote:
>>>>>>
>>>>>>> Dear all,
>>>>>>>
>>>>>>> we have submitted a revision of the 6top draft. It includes most of
>>>>>>> the changes proposed in the WG meeting in Berlin.
>>>>>>>
>>>>>>> please send us your comments.
>>>>>>>
>>>>>>> kind regards,
>>>>>>> Xavi
>>>>>>>
>>>>>>> ---------- Forwarded message ----------
>>>>>>> From: <[email protected]>
>>>>>>> Date: 2016-07-26 8:30 GMT+02:00
>>>>>>> Subject: [6tisch] I-D Action: draft-ietf-6tisch-6top-protocol-02.txt
>>>>>>> To: [email protected]
>>>>>>> Cc: [email protected]
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> A New Internet-Draft is available from the on-line Internet-Drafts
>>>>>>> directories.
>>>>>>> This draft is a work item of the IPv6 over the TSCH mode of IEEE
>>>>>>> 802.15.4e of the IETF.
>>>>>>>
>>>>>>>         Title           : 6top Protocol (6P)
>>>>>>>         Authors         : Qin Wang
>>>>>>>                           Xavier Vilajosana
>>>>>>>         Filename        : draft-ietf-6tisch-6top-protocol-02.txt
>>>>>>>         Pages           : 28
>>>>>>>         Date            : 2016-07-25
>>>>>>>
>>>>>>> Abstract:
>>>>>>>    This document defines the 6top Protocol (6P), which enables
>>>>>>>    distributed scheduling in 6TiSCH networks.  6P allows neighbor
>>>>>>> nodes
>>>>>>>    in a 6TiSCH network to add/delete TSCH cells to one another.  6P
>>>>>>> is
>>>>>>>    part of the 6TiSCH Operation Sublayer (6top), the next higher
>>>>>>> layer
>>>>>>>    of the IEEE802.15.4 TSCH medium access control layer.  The 6top
>>>>>>>    Scheduling Function (SF) decides when to add/delete cells, and
>>>>>>>    triggers 6P Transactions.  Several SFs can be defined, each
>>>>>>>    identified by a different 6top Scheduling Function Identifier
>>>>>>> (SFID).
>>>>>>>    This document lists the requirements for an SF, but leaves the
>>>>>>>    definition of the SF out of scope.  Different SFs are expected to
>>>>>>> be
>>>>>>>    defined in future companion specifications.
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> The IETF datatracker status page for this draft is:
>>>>>>> https://datatracker.ietf.org/doc/draft-ietf-6tisch-6top-protocol/
>>>>>>>
>>>>>>> There's also a htmlized version available at:
>>>>>>> https://tools.ietf.org/html/draft-ietf-6tisch-6top-protocol-02
>>>>>>>
>>>>>>> A diff from the previous version is available at:
>>>>>>> https://www.ietf.org/rfcdiff?url2=draft-ietf-6tisch-6top-protocol-02
>>>>>>>
>>>>>>>
>>>>>>> Please note that it may take a couple of minutes from the time of
>>>>>>> submission
>>>>>>> until the htmlized version and diff are available at tools.ietf.org.
>>>>>>>
>>>>>>> Internet-Drafts are also available by anonymous FTP at:
>>>>>>> ftp://ftp.ietf.org/internet-drafts/
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> 6tisch mailing list
>>>>>>> [email protected]
>>>>>>> https://www.ietf.org/mailman/listinfo/6tisch
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> 6tisch mailing list
>>>>>>> [email protected]
>>>>>>> https://www.ietf.org/mailman/listinfo/6tisch
>>>>>>>
>>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> 6tisch mailing list
>>>>>> [email protected]
>>>>>> https://www.ietf.org/mailman/listinfo/6tisch
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Dr. Xavier Vilajosana Guillén­
>>>>> Research Professor
>>>>> Wireless Networks Research Group
>>>>> Internet Interdisciplinary Institute (IN3)
>>>>> Universitat Oberta de Catalunya­
>>>>>
>>>>> +34 646 633 681| [email protected]­ | Skype­: xvilajosana
>>>>> http://xvilajosana.org
>>>>> http://wine.rdi.uoc.edu/
>>>>>
>>>>> Parc Mediterrani de la Tecnologia
>>>>> Av. Carl Friedrich Gauss, 5. Edifici B3
>>>>> 08860 Castelldefels (Barcelona)
>>>>>
>>>>>
>>>>>
>>>>> ­
>>>>>
>>>>
>>>>
>>>
>>
>
_______________________________________________
6tisch mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/6tisch

Reply via email to