Dear authors, WG, i carried the shepherd review of this document, please find below some comments and issues that should be fixed as part of the WG last call.
BR Daniele - ID nits: Summary: 14 errors (**), 0 flaws (~~), 17 warnings (==), 2 comments (--). Please note that there is a number of too long lines in the document and that the references need to be fixed (missing references, unused references, downref and outdated reference. - Intended Status (front page): it should be Standard Track - Front page: Ramon's affiliation is not properly formatted - Section 3: suggested change for improved readability o OLD In this document, it is assumed that wavelength converters require electrical signal regeneration. Consequently, WSONs can be transparent (A transparent optical network is made up of optical devices that can switch but not convert from one wavelength to another, all within the optical domain) or translucent (3R regenerators are sparsely placed in the network). o NEW WSONs can be transparent or translucent. A transparent optical network is made up of optical devices that can switch but not convert from one wavelength to another, all within the optical domain. On the other side translucent networks include 3R regenerators sparsely placed. In this document only wavelength converters that require electrical signal regeneration are considered. - Section 3: Expand LSC at first use - Section 3: s/need not operate/do not need to operate - Section 3: o OLD Two optical paths that share a common fiber link cannot be assigned the same wavelength. To do otherwise would result in both signals interfering with each other o NEW Two optical paths that share a common fiber link cannot be assigned the same wavelength otherwise both signals would interfere with each other - Section 3 - Figure 1: what do X and LSC stand for in the figure? Reading further it says: "If the LSC LSP induced a Forwarding Adjacency / TE link, the switching capabilities of the TE link would be [X X] where X < LSC (PSC, TDM, ...)" but it is not yet clear what it means. - Section 3: s/wavelength MUST be/wavelength must be - Section 4.1 o OLD (a) By means of Explicit Label Control (ELC) where the PCE allocates which label to use for each interface/node along the path. in the sense that the allocated labels MAY appear after an interface route subobject. o NEW (a) By means of Explicit Label Control (ELC), where the PCE allocates the label to be use on each interface/node along the path. In this case that the allocated labels may appear after an interface route sub-object. - Figure 3: 31 bits reserved for flags and only 1 used is a significant waste. Suggestion: whi don't you split the field into 16 reserved and 16 flags? - Section 4.2: Drop "to ''explicit'' and could add "The Walelngth Selection TLV MUST NOT be used when the M bit is cleared" - Section 4.4 s/ include/includes - Section 4.4 drop "however this is not a MUST". - Section 4.4.1: What is the X field at the beginning of the object? Type = X seems to be referring to that. Moreover indication of how to encode the "length" field is missing. - Section 4.4.1 :Attribute: if it can only assume 2 values why using 8 bits? If other future values can be defined I'd suggest to rephrase saying something like: "actually the following 2 values are defined....". - Figure 6: maybe you should add Type and Length? This TLV is encoded, ulikely the analogous WA object, using Reserved instead of flags. - Section 5: the text says that link identifier and assigned wavelength fields have variable length, the figure should be updated accordingly. - Section 5.2: s/One new bit flag are/One new flag is - Section 6.2: is it still needed?
_______________________________________________ Pce mailing list [email protected] https://www.ietf.org/mailman/listinfo/pce
