Clone wiki

meetings / 130712_webex

Minutes Webex 12 July 2013, 6TSCH group

Note: timestamps in PDT.

Taking notes (using Etherpad)

  1. Xavi Vilajosana
  2. Dominique Barthel
  3. Thomas Watteyne

Present (alphabetically)

  1. Alfredo Grieco
  2. Dominique Barthel
  3. Guillaume Gaillard
  4. Maria Rita Palattella
  5. Pascal Thubert
  6. Pouria Zand
  7. Qin Wang
  8. Thomas Watteyne
  9. Tom Phinney
  10. Xavi Vilajosana
  11. Yoshihiro Ohba

Recording

Slides

  • slides_130712_webex.ppt: slides shared during the call, edited "live" by Pascal during the call, and edited by Tom Phinney after the call.

Actions items

  • All to update drafts with new name: "6top" instead of "6tus".
  • Thomas to remove Orlando repository.
  • Thomas to reorganize meetings wiki in "last 2 meetings" and "all meetings".
  • Pascal to send slides to Tom. Tom to proof-read them and send back to Pascal and Thomas. Thomas to put final version online.
  • Thomas to create room in repo to store slides for presentations.
  • Each owner of an item to create a thread on the mailing list with the title of his/her presentation, discuss the content and get feedback from the group.

Agenda

  • Approval minutes & agenda bashing [2min]
  • Preparing for the BOF [50min]
  • Update draft-wang-6tsch-6tus [3min]
  • Update Architecture draft for PCE [2min]
  • Re-organization Bitbucket [2min]
  • AOB [1min]

Minutes

  • [08.00] pre-meeting discussion
    • Discussion about "Link" in the terminology draft
    • [Thomas] We have never really talked about a cell scheduled to multiple neighbors (i.e. multicast). Maybe remove that sentence from the terminology draft?
    • [Maria Rita] concept as discussed in the ML. Need to be differentiated to the IP link concept.
  • [08.05] Meeting starts
    • Recording starts at 8:07
    • [Pascal] Review agenda. Most of the call is preparing on the BOF.
  • [08.06] Approval Minutes

    No objections. Minutes approved.

  • [08.07] Changes to the agenda
    • [Pascal] Any changes to the agenda?
    • [Thomas] Have to leave 10min before end, can we do bitbucket section first?
    • [Pascal] Yes.
    • voting for a new name for 6tus now ended. Winner is "6TOP"

      Action item: drafts to use new name.

    • This means publish a new -00 version for 6TOP draft.
  • [08.09] Re-organization Bitbucket [Thomas]
    • Merged orlando repository into meetings repository
    • Added pages for Berlin meeting. Currently contains draft of agenda and people attending.
    • [Thomas] Can we remove the Orlando repository?

      No objections. Action item: Thomas to remove Orlando repository.

    • [Pascal] suggestion: leave the last one on the history page and have another page with the rest.
    • [Thomas] Any objection to that on the call?

      No objections. Action item: Thomas to reorganize meetings wiki in "last 2 meetings" and "all meetings".

  • [08.10] Preparing BOF
    • BoF is confirmed on Tuesday afternoon 1520-1650 (see https://datatracker.ietf.org/meeting/87/agenda.html)
    • Online voting for a date for a pre-BoF meeting now closed, winner is 5:30-7:00 on Monday afternoon.
    • Speakers: at the BoF, the chairs will not do the main talking.
    • We need to be convincing that the job is doable.
    • Provide Clear:
      • analysis of what is missing and what should be delivered.
      • when are we going to deliver the results.
      • why it has to be done at the IETF.
      • why it is not in an existing working group
    • We do not want technical discussions during the BOF
    • Address important points.
    • Speakers:
      • Show that we know the technology and have the skills needed.
      • Clearly point out that we are putting things together.
      • Highlight that we emulate the technology validated in the industry.
    • Show the activity of the group: ML figures, work in repository, drafts, etc.
    • There will be rehearsal time in the next 2 phone calls and pre-BOF.

      Pascal shares his desktop and edits the powerpoint live (see recording). The people on the call are asked to provide ideas on what to say in each of the sections which will be presented at the BoF.

      • 1.a) What is IEEE802.15.4e TSCH?
        • Show that we know what we’re talking about. Highlight the (good) differentiating features. Several contributions provided (Pascal, Thomas, Xavi, Dominique, Maria Rita, Qin, Alfredo...). Not get too much into MAC technical details, stay with the good properties.
        • what are the key points:
          • more controllable for power consumption, robustness, latency
          • several isolated flows
          • predictable power consumption
          • trade-off energy consumption, robustness and latency
          • more energy efficient than other as we know the traffic pattern.
          • compressed duty cycle.
      • 1.b) What is missing?
        • IETF-stamped architecture over TSCH. Adaptation to LLN of some existing protocols such as RSVP.
        • what is missing is a an architecture that puts everything together.
        • Adapt operation of reservation or PCE protocols to deterministic low power
        • 6top layer, missing time slotting for IP
        • Protocol between a centralized PCE and the nodes.
        • [Thomas] Missing protocol to PCE to express links and metrics.
        • [Xavi] how RPL fits with lower layer architecture, feeding RPL with underlying information (link metrics, etc)
      • 1.c) Why is this a problem?
        • ISA and WHART are not scalable to large networks.
        • [Thomas] This will raise lots of debate. Not sure we should emphasize it.
        • [Thomas] Other point is interoperability: in current situation, devices from different vendors operating 15.4e TSCH will not interoperate at link layer because of different ways to manage the MAC.
        • [Qin] what about mobility?
        • [Thomas] difficult selling point, little edgy for LLNs.
        • [Pascal] actually, RPL goes the right way with mobility.
        • [Alfredo] here we are focusing on deterministic service, which would exclude mobility altogether. * [Pascal] remove "will not scale" and replace with "will not allow mobility".
        • customer insatisfactions with compteing stds.
        • missing ietf arch. limits adoption
        • put everything as a big block.
        • Cannot do inter-operability with two existing solutions. Upper MAC is missing
        • lack of common network management.
      • 1.d) Status of 6TSCH group
        • 150+ people on the ML
        • week call
        • bitbucket team, git repositories
        • 6 drafts
        • Open source and vendor implementations
        • Thoughts:
          • add numbers (how many people per call, stats on facts)
          • formal bar bof in Orlando
      • 2.a) Background and introduction
        • [Thomas] mention existing TSCH products, we know it works. RPL works, too.
        • [Pascal] we have the people who know to make them work.
        • [Xavi] Different vendors with own scheduling algorithms, also academic research on distributed and centralized scheduling.
        • [Pascal] Careful about mentioning "research".
        • [Qin] open-source demonstrators.
        • [Alfredo] say "experimental" research.
        • [Maria Rita] mention IoT6 project.
        • Existing work demonstrates feasibility:
          • what do we want to show with the intro?
            • show that there are TSCH based deployments, implementation and commercial products
            • RPL is used in AMI/AMR, commercial products.
            • experimental research results and vendor implementations of scheduling shows the need for standardization on interfaces.
            • IoT6 project and other research directions
      • 2.b) Description of WG
        • Showing that we have the right communities and contributors in this group. Vendors and deployers (on the user side). Expertise in PCE (Maria Rita on academic side), RPL (JP, Pascal), PANA (Yoshi), ISA (Tom), WirelessHART (Pascal, Thomas, Jonathan), open source (Xavi).
        • Showing that we are the right community, vendors, reseachers, deployers, etc.
        • Experts in PCE, RPL, 6LoWPAN, ISA100.11a, WirelessHART, PANA
        • Multiple OSes and environments. Open-source efforts, etc..
        • technology item

          We're running out of time in the call, Pascal speeds up.

      • 2.c) Work items
        • identify work items. Protocol that we need to touch.
        • discuss in the ML
        • create 6TOP layer and modify or ammend other protocols.
      • 2.d) Non-milestone work items
        • co-existence of 6tsch /6top with other standards. WirelessHART, ISA100.11a, IEEE802.11 WiFi
      • 2.e) External work to other WG
        • PCE
        • 6MAN
        • 6LO? ROLL?
    • Draft Agenda assignments
      • 1) Problem statement
        • 1a) what is IEEE802.15.4e TSCH?
          • primary: Maria Rita
          • backup: Alfredo
        • 1b) what is missing?
          • primary: Xavi
          • backup: Thomas, Pascal
        • 1c) Why is this a problem?
          • primary: Robert
          • backup: Alfredo
        • 1d) Status of 6TSCH group
          • primary: Thomas
          • backup: Maria Rita
      • 2) Discussion of the charter
        • 2a) background and introduction
          • ??
        • 2b) description of WG
          • primary: Dominique
        • 2c) work items
          • primary: Maria Rita
          • backup: Pascal and Thomas
          • Show we know what we are doing, don’t get too much into details.
        • 2d) non-milestone work items
          • primay: Thomas
        • 2e) external work to other WG
          • primary: Pascal
      • 3) drafts:
        • 5min per draft
        • show what we are doing and what
        • 6top draft presented by Xavi. To work with Qin on the slides and contents.

          Action item: Pascal to send slides to Tom. Tom to proof-read them and send back. Thomas to put final version online. Action item: Thomas to create room in repo to store slides for presentations. Action item: Each owner of an item to create a thread on the mailing list with the title of his/her presentation, discuss the content and get feedback from the group.

  • [09.07] Update draft-wang-6tsch-6top [Qin]
    • added label switching commands
    • added trackID
    • changed the name to 6top
  • [09.07] Update Architecture draft for PCE
    • [Pascal] Waiting for input from JP.
  • [09.07] AOB
  • [09.08] Meeting ends

Updated