[Sis-mia] Response to Green Book CESG Review comments

[Grubbs, Rodney P. (MSFC-HP27)]

All, attached is our Green Book draft with comments/revisions since the CESG review.

Below are all the comments.  The attached document includes the CESG review comments as MS Word comments to the document so you can see where they apply in context, and see how they???ve been dealt with.  The attached document also has text highlighted in different colors to keep up with who originally authored that text.

Scott, I believe you and I still need to discuss Peter Shames??? comments on section 5.2.4.

Also I believe Scott, Keith and I need to decide how to address Gian Paolo???s comment on Annex A since his comment really addresses another book, not this Green Book.

So, to the entire MIA WG, I ask that you look at the comments below and how they are dealt with in the attached document, and let me and Osvaldo know by COB this Friday, March 2 if you have any issues with the changes.

Scott and Keith, we should try to schedule some time this week or next to discuss the remaining issues to be resolved. Please let me know your schedule options.  Osvaldo, if your schedule allows we will include you in that conversation.

CESG Polling Comments:

CESG E-Poll Identifier: CESG-P-2017-12-006 Approval to publish CCSDS 730.2-G-1, Requirements for Streaming Services over Bundle Protocol (Green Book, Issue 1)
Results of CESG poll beginning 29 December 2017 and ending 19 January 2018:

                 Abstain:  0 (0%)
 Approve Unconditionally:  1 (16.67%) (Burleigh)
 Approve with Conditions:  5 (83.33%) (Barkley, Merri, Shames, Calzolari, Wilmot)
 Disapprove with Comment:  0 (0%)

CONDITIONS/COMMENTS:

Erik Barkley (Approve with Conditions): the recommendation to "Requirements for Video Streaming Services over Bundle Protocol". The report only deals with video streaming. There are in fact other forms of streaming data (e.g, streaming monitor data, streaming analytics, etc) which I don't think this report addresses.

Mario Merri (Approve with Conditions): Sections 3.1.1: It should be made clear that the described data stream consolidation strategy is just one possible example. Clearly, it makes sense, but - depending on the requirements of the video reproducing application - other strategies are also possible. Similarly, for Figure 3.1, it should be made very clear that what is provided there is just one example of a possible MMI. Many other solutions are possible.

Section 3.3 (pag 3-6): "Return (ground-to-space) video". Please note that in CSS "Return" is used for the opposite direction (space-to-ground) and "forward" for ground-to-space. I suggest to harmonise the use of these terms and update the document accordingly.

Peter Shames (Approve with Conditions): There were a number of items that were troubling, and that should be updated in this doc before it is published.

These are the "biggies" for me:

BSS and particularly BSSP appear to be significant protocol artifacts, but there are no specs for them. As I read it they are only documented in a paper (Sotirios-Angelos Lenas, Scott C. Burleigh, and Vassilis Tsaoussidis. ???Bundle Streaming Service: Design, Implementation and Performance Evaluation.??? Transactions on Emerging Telecommunications Technologies, 26, no. 5 (May 2015): 905??917. ) This strikes me as being not at all the norm for either CCSDS nor IETF documents. If BSSP is an important CLA spec I think it really ought to be properly documented.
The references to ION and D-VADER, which are implementations as opposed to specifications also seems a little "off-color" for a CCSDS document.
A " Picture Quality Analysis (PQA) system" is referenced, but it is really not defined. Is there some specification or system that is actually being referenced.
" IBR-DTN" is mentioned, but there does not appear to be a reference to it either. In the figure where it appears, fig 5-3, it is impossible to tell just what it does. It is a "black box" with no obvious functional assignment nor architecture.
I think that one or more end-to-end protocol diagrams would aid understanding of how all of these items fit together. I know that DTN has many different underlying CLA configurations as well as the optional use of LTP. Showing these in even a simplified EE diagram would be a useful addition.

Assuming you accept these inputs, it may be that this doc ought to be titled something like "Concepts, Rationale, and Requirements for Streaming Services over BP". That will give you the scope to cover all these topics adequately.

Gian Paolo Calzolari (Approve with Conditions): Section 5.1.3 states "The other component of the Bundle Streaming Service is Bundle Streaming Service Protocol (BSSP), a BP ???convergence layer??? protocol." and "Figure 5-1: Overview of Transparent Gateway" includes a box marked .
The figure is the only place where BP CLA is used, so it could be wise to replace it with BSSP.
Alternatively the reference to BP CLA can be added in bracket at the end on the above mentioned sentence in Section 5.1.3.

Moreover I think that some improvement to the description of BSSP can be worthwhile.
From the description I understand that BSSP is mainly a "smart" CLA able to select the most appropriate of two available (reliable/unreliable) CLA's.
I think this could be highlighted better and reference to Annex B of CCSDS 734.2-B-1 would be helpful.

Is there the intention of defining BSSP in a normative book (e.g. Blue). If yes, state it in this green book.

Annex A misses BP , despite it is obvious what BP means :o)
Annex A defines "CLA convergence layer adaptor" but the term ADAPTER is used in CCSDS 734.2-B-1 ==> please align to Blue Book
This book sometimes uses the term "convergence-layer" while no hyphen is used in in CCSDS 734.2-B-1 for convergence layer ==> please align to Blue Book
----------------------------
Last but not least I think that a feedback effect (e.g. Technical corrigendum) to CCSDS 734.2-B-1 is required for Annex B y=that contains the Convergence Layer Adapters (CLAs).
Clause in B2.1 states "Compliant implementations shall implement at least one of the CLAs in this section." but, as the (normative) Annex B defines the following 3 CLAs
- LTP CONVERGENCE LAYER ADAPTER
- UDP CONVERGENCE LAYER ADAPTER
- CCSDS ENCAPSULATION SERVICE CONVERGENCE LAYER ADAPTER
then BSSP would not be not BP compliant.....
Therefore it looks wise adding there the definition of a TCP CLA and/or stating something in B2.1 that the list of CLAs in Annex B is not exhaustive.
I also noticed that (normative) section A6 BASIC REQUIREMENTS mentions adapters referring to sections B3.1.2 and B4 but not B5 (i,e. the CCSDS ENCAPSULATION SERVICE CLA defined in B5 and the LTP CLA for Unreliable transmission defined in B3.1.3 should be added to the table in A6).

Of course this may be taken separately from the poll for this Green Boo, but it would be very nice aligning the books (almost) at the same time.

Jonathan Wilmot (Approve with Conditions): General comment, book is a discussion of use cases and not a requirements document with "shalls". Is the document name appropriate? Note: comments below avoid duplication with other CESG review comments.

Condidtions:

1) Section 4 states "CFDP (as specified in 5.3)" Section 5.3 states "(CFDP) should be investigated" 5.3 does not seen to specify use case. Add more specification or reword reference to 5.3. It would be useful to say CFDP Class 1 over DTN would provide reliable multi-hop video file transfers.

2) Operations concept for file transfer in section 4 page 19 should not be recomended due to potential duplication storage requirements. Reword to state streaming of CFDP files should pause in LOS conditions to avoid data duplication between bundles/LTP and video file system.


Total Respondents: 6
All Areas responded to this question.

SECRETARIAT INTERPRETATION OF RESULTS:  Approved with Conditions
PROPOSED SECRETARIAT ACTION:            Generate CMC poll after conditions have been addressed
*****
Rodney Grubbs
NASA Imagery Experts Program Manager
MSFC HP27
256-544-4582
256-603-3270 (cellular, text message capable)
Follow me on Twitter @rod4dtv
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.ccsds.org/pipermail/sis-mia/attachments/20180226/06ee7ab1/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 730x2g0_CESG_Approval with notes and highlights_rg_wl.docx
Type: application/vnd.openxmlformats-officedocument.wordprocessingml.document
Size: 984172 bytes
Desc: 730x2g0_CESG_Approval with notes and highlights_rg_wl.docx
URL: <http://mailman.ccsds.org/pipermail/sis-mia/attachments/20180226/06ee7ab1/attachment.docx>