Dear Colleagues,
Thank you for the review of the IEEE 802.3 PARs and CSDs currently under consideration. The responses to the comments received from IEEE 802.1 and IEEE 802.11 can found below, along with the URLs to updated PARs and CSDs (both compare and clean) where
appropriate.
Best regards,
David
IEEE 802.3cy PAR and CSD
========================
IEEE 802.11 IEEE 802.3cy PAR comment 1
Comment:
PAR 2.1 – The use of “Automotive Electrical Ethernet” was undefined. Should this just be “Automotive Ethernet”? Or should this be defined in the scope of the project.
Response:
ACCEPT IN PRINCIPLE
IEEE P802.3cy: Physical Layer Specifications and Management Parameters for greater than 10 Gb/s Electrical Automotive Ethernet
IEEE P802.3cz: Physical Layer Specifications and Management Parameters for Multi-Gigabit Optical Automotive Ethernet
IEEE P802.3cy and IEEE P802.3cz have coordinated this joint response.
IEEE 802.11 IEEE 802.3cy PAR comment 2
Comment:
5.2.b – add “architectures” after “zonal”. Change “(centralized architecture)” to “(centralized)”.
Response:
ACCEPT IN PRINCIPLE
The text referenced in the comment is found in PAR item 5.5, not 5.2.b. As a result the text '... zonal (centralized architecture) using Ethernet ....' in PAR item 5.5 has been changed to read '... zonal architectures using Ethernet ...'.
The automotive industry has settled on zonal as the preferred terminology. All references to centralized have been removed from the PAR and CSD.
IEEE 802.11 IEEE 802.3cy PAR comment 3
Comment:
5.6 –“Tier 1 and below (top-level and below)” this seems to include all “automotive suppliers”. Suggestions: 1. delete “(top-level and below)” or 2. change “automotive Original Equipment Manufacturers (car makers) and Tier 1 and below (top-level and below)
automotive suppliers” to “automotive Original Equipment Manufacturers (car makers), automotive suppliers”
Response:
ACCEPT IN PRINCIPLE
In PAR item 5.6 the text 'Tier 1 and below (top-level and below)' has been changed to read 'Tier x' and the text 'For 5.6: Tier x refers to the various levels of suppliers to Original Equipment Manufacturers (e.g., car manufacturer). A Tier 1 supplier
for example supplies components or subsystems directly to the Original Equipment Manufacturer.' has been added to PAR item 8.1.
IEEE 802.1 IEEE 802.3cy CSD comment 1
Comment:
Compatibility: While 802.1 is very supportive of this project, the standards defined by 802.1 have not focused on operation over asymmetric speed links. We suggest to mention that the use of asymmetric links may require further study of the set of protocols
defined in IEEE Std 802.1Q.
Response:
REJECT
For an observer above the MAC, a link with asymmetric data rates is indistinguishable from a symmetric link with asymmetric traffic flow. Our understanding is that the latter is something that has never been an issue for IEEE 802.1.
IEEE 802.1 IEEE 802.3cy CSD comment 2
Comment:
Economic Feasibility
- “The balance of costs between infrastructure and attached stations is not applicable to the automotive environment.”
- This statement is unclear to 802.1. Infrastructure includes bridges and routers, and we believe there are both infrastructure and attached stations within the automotive environment. Furthermore, we believe the balance of costs between these components
of the solution is critical to the success of 802 technologies in the automotive environment.
- Please clarify your evaluation of the balance of costs.
Response:
REJECT
Subclause 14.2.5 'Economic Feasibility' of the IEEE 802 Operations Manual reads 'Balanced costs (infrastructure versus attached stations)'. Subclause 3.1 'Definitions' of IEEE Std 802-2014 IEEE Standard for Local and Metropolitan Area Networks: Overview
and Architecture defines 'station' as 'An end station or bridge. See also: bridge; end station.' and 'end station' as 'A functional unit in an IEEE 802® network that acts as a source of, and/or destination for, link layer data traffic carried on the network.'.
As a result, the IEEE 802 Overview and Architecture defines both a bridge and a router (as a source and destination for link layer data traffic carried on the network) as a station. Since this criteria includes the text 'infrastructure versus attached
stations', and since both a bridge and a router is a station, and therefore cannot be infrastructure, we do not believe the statement 'Infrastructure includes bridges and routers ...' is correct.
Neither infrastructure as defined, nor the inclusion of bridges and routers into infrastructure, apply to automotive networks because they are very different from enterprise networks and other network types where IEEE 802 technologies have traditionally
been applied. An automotive network is precisely engineered from qualified subsystems (e.g., cable harness, electronic control units, etc.); and then replicated millions of times for installation in individual vehicles. Unlike an enterprise network, bridge
and router functionality may not be a distinct device in an automotive network, e.g., it may be one function of an electronic control unit which may also include functions that might be considered a server in an enterprise network.
IEEE 802.11 IEEE 802.3cy CSD comment 1
Comment:
References to “Zonal (centralized) architecture” vs “(zonal or central architecture)” vs in the PAR “zonal (centralized architecture)” The references should be consistent.
Response:
ACCEPT IN PRINCIPLE
In Broad Market Potential, under 'Broad Sets of Applications' the text ' Recent automotive electrical architecture changes, (zonal or central architecture) require ...' has been changed to read ' Recent automotive electrical architecture changes, e.g.
to zonal architectures require ...'. In Economic Feasibility 'Zonal (centralized) architecture, enabled by ...' has been changed to read 'Zonal architecture, enabled by ...'.
The automotive industry has settled on zonal as the preferred terminology. All references to centralized have been removed from the PAR and CSD.
IEEE 802.11 IEEE 802.3cy CSD comment 2
Comment:
For PAR 5.6, after we provided the suggestions, we reviewed 802.3cz, and the description used in 5.6 and 8.1 seemed a better solution. Please consider using their text for 5.6 and 8.1.
Response:
ACCEPT IN PRINCIPLE
See response to IEEE 802.11 IEEE 802.3cy PAR comment 3 above, the text has been added to 8.1.
IEEE P802.3cz PAR and CSD
=========================
IEEE 802.11 IEEE 802.3cz PAR comment 1
Comment:
PAR 2.1 Title: “Optical Automotive Ethernet” for 802.3.cy it was “Automotive Electrical Ethernet” suggest making the titles more consistent. – Both TF would need to be involved in discussion. Possible title : Physical Layer Specifications and Management
Parameters for Multi-Gigabit Optical Ethernet for the Automotive Environment”
Response:
ACCEPT IN PRINCIPLE
'Automotive Ethernet' is an industry term <
https://events.weka- fachmedien.de/automotive-ethernet-congress/home/>, <
https://standards.ieee.org/events/automotive/> and will be found
in magazine and technical papers. We want to use the term, but need to differentiate between the electrical and optical projects. As a result the IEEE P802.3cy title has been updated to read as follows (see responses to IEEE P802.3cy comments above), the unchanged
IEEE P802.3cz title is provided for reference.
IEEE P802.3cy: Physical Layer Specifications and Management Parameters for greater than 10 Gb/s Electrical Automotive Ethernet
IEEE P802.3cz: Physical Layer Specifications and Management Parameters for Multi-Gigabit Optical Automotive Ethernet
IEEE P802.3cy and IEEE P802.3cz have coordinated this joint response.
IEEE 802.11 IEEE 802.3cz CSD comment 1
Comment:
CSD: version submitted was watermarked “DRAFT”. Consider updating when submitting to the IEEE 802 LMSC.
Response:
ACCEPT IN PRINCIPLE
The updated IEEE P802.3cz CSD being submitted has the watermark 'DRAFT' removed.
IEEE 802.1 IEEE 802.3cz CSD comment 1
Comment:
Economic Feasibility
- “The balance of costs between infrastructure and attached stations is not applicable to the automotive environment.”
- This statement is unclear to 802.1. Infrastructure includes bridges and routers, and we believe there are both infrastructure and attached stations within the automotive environment. Furthermore, we believe the balance of costs between these components
of the solution is critical to the success of 802 technologies in the automotive environment.
- Please clarify your evaluation of the balance of costs.
Response:
REJECT
Subclause 14.2.5 'Economic Feasibility' of the IEEE 802 Operations Manual reads 'Balanced costs (infrastructure versus attached stations)'. Subclause 3.1 'Definitions' of IEEE Std 802-2014 IEEE Standard for Local and Metropolitan Area Networks: Overview
and Architecture defines 'station' as 'An end station or bridge. See also: bridge; end station.' and 'end station' as 'A functional unit in an IEEE 802® network that acts as a source of, and/or destination for, link layer data traffic carried on the network.'.
As a result, the IEEE 802 Overview and Architecture defines both a bridge and a router (as a source and destination for link layer data traffic carried on the network) as a station. Since this criteria includes the text 'infrastructure versus attached
stations', and since both a bridge and a router is a station, and therefore cannot be infrastructure, we do not believe the statement 'Infrastructure includes bridges and routers ...' is correct.
Neither infrastructure as defined, nor the inclusion of bridges and routers into infrastructure, apply to automotive networks because they are very different from enterprise networks and other network types where IEEE 802 technologies have traditionally
been applied. An automotive network is precisely engineered from qualified subsystems (e.g., cable harness, electronic control units, etc.); and then replicated millions of times for installation in individual vehicles. Unlike an enterprise network, bridge
and router functionality may not be a distinct device in an automotive network, e.g., it may be one function of an electronic control unit which may also include functions that might be considered a server in an enterprise network.
IEEE P802.3da PAR and CSD
=========================
IEEE 802.11 IEEE P802.3da PAR comment 1
Comment:
PAR 2.1 missing the word “network” -suggest add to title: “Physical Layer Specifications and Management Parameters for 10 Mb/s Operation over Single Balanced Pair Multidrop Network Enhancements”
Response:
ACCEPT IN PRINCIPLE
The comment suggested the addition of Network. The group thought that network would not be the proper term as we are specifying enhanced operation over a segment of the network and not the network as a whole.
As a result the PAR title has been changed to read 'Amendment: Physical Layer Specifications and Management Parameters for Enhancement of 10 Mb/s Operation over Single Balanced Pair Multidrop Segments'.
In addition PAR item 5.2.b was modified to be consistent with the title change (see below).
IEEE 802.11 IEEE P802.3da PAR comment 2
Comment:
5.2.b suggest add text from the CSD “This amendment specifies optional power delivery supporting multiple powered devices on the mixing segment.”
Response:
ACCEPT IN PRINCIPLE
The text '... and optional power delivery to enhance 10Mb/s multidrop single balanced pair networks' in PAR item 5.2.b has been changed to read '... and optional power delivery supporting multiple powered devices on the 10 Mb/s mixing segment.'.
IEEE 802.11 IEEE P802.3da PAR comment 3
Comment:
5.5 “from legacy networks to Ethernet” what are “legacy networks” in this context? Suggest “legacy non-Ethernet networks”.
Response:
ACCEPT
The text '... from legacy networks ...' in PAR item 5.5 has been changed to read '... from legacy non-Ethernet networks ...'.
IEEE 802.11 IEEE P802.3da CSD comment 1
Comment:
CSD: 1.2.2 Broad Market Potential: suggest same change as in PAR 5.5.
Response:
ACCEPT
The text '... from legacy networks ...' in CSD Broad Market Potential has been changed to read '... from legacy non-Ethernet networks ...'.
IEEE P802.3db PAR and CSD
=========================
IEEE 802.11 IEEE P802.3da CSD comment 1
Comment:
CSD Technical Feasibility: “IEEE 802.3 has already established 100 Gb/s, 200 Gb/s, and 400 Gb/s MAC specifications suitable for 100 Gb/s per wavelength PHY operation in IEEE Std 802.3bs-2017 and IEEE Std 802.3cd-2018.”
And Economic Feasibility “Higher speed 100 Gb/s signaling leads to reduced lane counts, reduced fiber and component counts, reduced complexity, and lower cost than previously standardized PMDs based on 50 Gb/s signaling”
One points out that it is already done, and one points out that it is being developed.
Is there a consistency issue?
Response:
ACCEPT IN PRINCIPLE
We do not believe there is a consistency problem. The first bullet under Technical Feasibility is intended to indicate that the requisite MAC protocols for 100, 200, and 400 Gb/s have already been defined. This project will define additional 100 Gb/s PHYs
to operate with the existing MAC specifications. However, in order to clarify this, we are modifying the text 'Higher speed 100 Gb/s signaling leads to reduced ...' in the fourth bullet in Economic Feasibility to read 'Higher speed 100 Gb/s signaling over MMF
will lead to reduced ...' to highlight the work yet to be done.