Re: [mpls] Draft submitted - MPLS Entropy Block

["Joel M. Halpern" <[e-mail]>]

Jeff, you assertion that the authors are ignoring existing work seems to 
border on mind-reading.  Binding SIDs (ala 8662) are a way to reduce 
label stack depth.  It is a useful tool.  Being able to remove the 
Entropy Label Indicator would also be beneficial.

The proposal does recognize that one may well need more than one entropy 
label (and thus labels aimed at different targets) in the stack.  The 
initial portion of the document does focus on the single entropy label 
case to make the idea clear.  The authors (i am not one of them) are 
well aware of the work on advertising various attributes of label depeth 
for generation and processing.

Yours,
Joel

On 6/2/2020 4:54 PM, Jeff Tantsura wrote:
> Hello co-authors,
> 
> The solution proposed ignores all the previous work done to address this 
> particular issue. Without going into more details
> Please read RFC8662 to better understand all the different pieces involved.
> draft-ietf-isis-mpls-elc/draft-ietf-ospf-mpls-elc specifically to 
> understand control plane involved.
> 
> Cheers,
> Jeff
> On Jun 2, 2020, 1:47 AM -0700, Jiang He 
> <jiang.he=[e-mail]>, wrote:
>>
>> Hello All,
>>
>> I just submitted a draft, 
>> https://datatracker.ietf.org/doc/draft-jiang-mpls-entropy-block/ .  it 
>> describes an approach for MPLS load balance, by  allocating a specific 
>> MPLS label block for entropy usage.
>>
>> The purpose is to mitigate the limitation of the number of labels can 
>> be pushed and the number of labels can be inspected when looking for 
>> entropy information.
>>
>> Your comments are highly appreciated!
>>
>> Regards,
>>
>> Jiang
>>
>> _______________________________________________
>> mpls mailing list
>> [e-mail]
>> https://www.ietf.org/mailman/listinfo/mpls
> 
> _______________________________________________
> mpls mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/mpls
>