& cplSiteName &

MPLS Argument Leads to Split Standard

Craig Matsumoto
2/28/2011

The MPLS operations, administration and management (OAM) effort officially has an offshoot, which ought to make Alcatel-Lucent (NYSE: ALU) and Huawei Technologies Co. Ltd. happy while possibly raising some blood pressures at Cisco Systems Inc. (Nasdaq: CSCO) and Ericsson AB (Nasdaq: ERIC).

Last week, International Telecommunication Union, Standardization Sector (ITU-T) Study Group 15 apparently passed a recommendation for using the ITU Y.1731 standard for operations, administration and management (OAM) in transport networks. Light Reading didn't have a formal announcement handy at press time; we're inferring the details from an Internet Engineering Task Force (IETF) and Internet Society (ISOC) statement denouncing the ITU-T decision, and from a happy but vague e-mail sent to us by Huub van Helvoort, a Huawei representative to the ITU-T.

Why this matters
Analysts had started feeling like this debate would end up this way, but it's still an interesting, dramatic turn in what's become a heated argument over standards. The side that got their way includes AlcaLu, China Mobile Ltd. (NYSE: CHL), Huawei and Telecom Italia (TIM) ; they had campaigned to get Y.1731 included in the MPLS-TP discussion.

Last fall, those companies were accused of disrupting standards meetings by bringing this issue up again. Many participants considered Y.1731 a dead issue, bypassed in favor of an MPLS-TP approach that was to be jointly developed by the IETF and ITU-T. But Y.1731 is already implemented in some equipment and has been deployed by China Mobile, and its supporters clearly weren't willing to let go.

The next move would apparently belong to the MPLS-TP camp, which includes big names of its own. Cisco and Ericsson, in particular, have campaigned about the supposed superiority of MPLS-TP, leaning on its compatibility with MPLS.

What's wrong with having two approaches? The IETF/ISOC statement says it runs counter to the open nature of the Internet, because it "ensures that the two product groups [those based on Y.1731 and those based on MPLS] will not work together."

For more
If you didn't catch the fun a few months ago, here's our coverage of the Y.1731 debate, with the most recent article listed first:

— Craig Matsumoto, West Coast Editor, Light Reading

(37)  | 
Comment  | 
Print  | 
Newest First  |  Oldest First  |  Threaded View        ADD A COMMENT
Page 1 / 4   >   >>
Huub_van_Helvoort
Huub_van_Helvoort
12/5/2012 | 5:11:53 PM
re: MPLS Argument Leads to Split Standard


Hello Graig,


In my message to you I did not claim I was happy, I wrote that I had exciting news and that I was too tired to give you more detailed information, which I did send you minutes ago, before I found this new Blog.


There was a lot of excitement in the ITU-T SG15 meeting but not everybody was happy.


Were you not surprised by the speed at which the IETF statement appeared on the ISOC website? To me it looks like they already anticipated this determination and instead of working on a compromise worked on this press release.


BR, Huub.


 


 

jcurranarin
jcurranarin
12/5/2012 | 5:11:52 PM
re: MPLS Argument Leads to Split Standard


In light of ITU resolution 101 (which calls for "ITU-T to continue its collaborative activities on IP-based networks with ISOC/IETF and other relevant recognized organizations, in respect of interconnectivity with existing telecommunication networks and migration to NGN") and the presence of a ITU-T/IETF Joint Working Team agreement (RFC 5317) on this specific work that specified MPLS-TP should be undertaken by the IETF and called for "Termination of the work on current T-MPLS", it appears that the ITU-T leadership faces a very difficult decision by the actions of SG-15.


If the ITU-T lets the matter stand, then they will have effectively shown that they are unable to participate in good faith with another standard body, in this case the IETF (which is the preeminent standards organization for the Internet) and this could have far reaching ramifications for cooperation in other areas between ITU and the Internet community.  As the ITU seeks relevance within the Internet space, the Internet standards community has reached out in good faith to engage in any and all areas that might be of mutual interest.  This level of "enhanced cooperation" has been predicated on trust and open processes, and it now appears to have been a misplaced belief in the case of at least one ITU-T study group.


/John


John Curran


President and CEO


American Registry for Internet Numbers (ARIN)

Sterling Perrin
Sterling Perrin
12/5/2012 | 5:11:51 PM
re: MPLS Argument Leads to Split Standard


The reality is that Huawei and Alcatel-Lucent combined represent roughly 50% of the optical transport market - meaning they have a very large constituency. I don't see how the operators represented by these two suppliers could realistically be ignored.


To place the blame fully on SG15 for refusing to cooperate with the IETF here is unfair. The debate, as everyone here knows, has been highly politicized, and there is plenty of motive for layer 3 suppliers to promote an option that places them in the strongest position and their competitors in the weaker position. ALU, Huawei, Cisco, Ericsson, etc. are all competitors looking to gain an advantage. John Curran's post takes the perspective that the IETF position was the unbiased/neutral position in the debate. This is not true.


All sides need to share the blame in a failure to reach a compromise. In the absence of a compromise, an offshoot of y.1731 was inevitable.


Sterling

jcurranarin
jcurranarin
12/5/2012 | 5:11:51 PM
re: MPLS Argument Leads to Split Standard


Sterling - 


   Alas, I'd agreed with you if this was simply the decision of two working groups in two different standards bodies working in an uncoordinated fashion.  We all recognize that happens from time to time.  


   In this case, there was an actual agreement between ITU-T and the IETF regarding how this matter would be handled.  That agreement allows both organizations to coordinate their efforts to be complimentary and avoid conflicts precisely of this nature.  


   In light of a prior agreement of coordination, the breach that has occurred is quite problematic.  Asked another way, how should the IETF view any future coordination agreements from the ITU-T?


/John


 

gigeguy
gigeguy
12/5/2012 | 5:11:50 PM
re: MPLS Argument Leads to Split Standard


Sterling,


As you can probably guess from my handle, I'm an Ethernet person, and Y.1731 provides awesome OAM for ETHERNET networks! However, it was not created with MPLS in mind or for use on MPLS(-TP) networks, and only provides at best a limited subset of the OAM functionality needed to meet the MPLS-TP requirements as agreed to between ITU-T SG15 and the IETF. In this case, the IETF-developed OAM provides the required functionaity, and having two solutions just hurts interoperability and drives up costs for everybody.


What was really going on here was that China Mobile deployed pre-standard products from its mostly Chinese vendors (plus ALU), and is unwilling to pay the price any operator has to if it bets on a particular pre-standard technology and the bet goes sour. Plus the vendors that implemented the Y.1731-based approach don't want to lose their own time-to-market advantage. Stir in some posturing from the Chinese goverment to support their local operators and vendors, and you end up with last week's fiasco.


 

Sterling Perrin
Sterling Perrin
12/5/2012 | 5:11:49 PM
re: MPLS Argument Leads to Split Standard


Gigeguy-


Thanks for your comments. My understanding is that the support for y.1731 goes beyond a single operator or even the Chinese operators. ie, there are big European operators in the camp, which are being served by ALU, i believe.


If it is just a case of China Telecom - or the 3 Chinese operators - pushing the agenda, then i would agree with you. But i'm still thinking that the support is broader than China.


Sterling

Charles_C
Charles_C
12/5/2012 | 5:11:49 PM
re: MPLS Argument Leads to Split Standard


Short-sighted and stupid

chechaco
chechaco
12/5/2012 | 5:11:49 PM
re: MPLS Argument Leads to Split Standard


Politics aside, neither tool set IETF's or ITU's addresses all possible constructs of MPLS-TP. The scope of the porposed adaptation ofY.1731 is limited to point-to-point co-routed LSPs only. Scope of IETF's tool set broader though not consistant. Some elements of IETF's OAM address all possible types of connections of MPLS-TP (LM/DM and FM) while CC/CV limited to bi-directional p2p LSPs, both associated and co-routed. Thus, ITU's scope is more narrow at this time even if that is what deploying customers can live at the moment.

Huub_van_Helvoort
Huub_van_Helvoort
12/5/2012 | 5:11:48 PM
re: MPLS Argument Leads to Split Standard


Hello John,


The title of your comment suggests that it only the ITU-T that has to make a decision. I disagree, also the IETF should take a decision: revive the cooperation or continue to run away from a compromise that satisfies the needs of the whole industry.


Regards, Huub.

Huub_van_Helvoort
Huub_van_Helvoort
12/5/2012 | 5:11:47 PM
re: MPLS Argument Leads to Split Standard


Hello Gigeguy,


You wrote:


As you can probably guess from my handle, I'm an Ethernet person, and Y.1731 provides awesome OAM for ETHERNET networks! However, it was not created with MPLS in mind or for use on MPLS(-TP) networks"


Please note that Y.1731 was developed with a Transport Network in mind. To provide the service providers with OAM comparable to the OAM they were used to in their SDH/SONET/OTN networks.


"and only provides at best a limited subset of the OAM functionality needed to meet the MPLS-TP requirements as agreed to between ITU-T SG15 and the IETF."


Would you be so kind to exactly indicate what the limitations are.


In this case, the IETF-developed OAM provides the required functionaity, and having two solutions just hurts interoperability and drives up costs for everybody.


The IETF has no MPLS-TP OAM RFCs.





What was really going on here was that China Mobile deployed pre-standard products from its mostly Chinese vendors (plus ALU), and is unwilling to pay the price any operator has to if it bets on a particular pre-standard technology and the bet goes sour.


See my previous comment on the availability of OAM, that is the fact.


Plus the vendors that implemented the Y.1731-based approach don't want to lose their own time-to-market advantage. Stir in some posturing from the Chinese goverment to support their local operators and vendors, and you end up with last week's fiasco.


As Sterling already pointed out: there are also major European service providers supporting the standardisation of Y.1731 based OAM, as well as service providers in Korea and Japan.





Page 1 / 4   >   >>
Featured Video
Upcoming Live Events
October 22, 2019, Los Angeles, CA
November 5, 2019, London, England
November 7, 2019, London, UK
November 14, 2019, Maritim Hotel, Berlin
December 3-5, 2019, Vienna, Austria
December 3, 2019, New York, New York
March 16-18, 2020, Embassy Suites, Denver, Colorado
May 18-20, 2020, Irving Convention Center, Dallas, TX
All Upcoming Live Events