Breaking News
News & Analysis

100G Effort Eyes One Optical Lane

6/2/2014 09:00 AM EDT
10 comments
NO RATINGS
More Related Links
View Comments: Threaded | Newest First | Oldest First
jdambrosia951
User Rank
Rookie
Industry Interest
jdambrosia951   6/2/2014 1:58:51 PM
NO RATINGS
Rick,

Thanks for the great write-up.  I am very excited for this workshop given the interest expressed at the IEEE P802.3bs Task Force Meeting in May, where an effective lane rate of 100Gb/s clearly had the most interest. 


This has the potential to be a defining moment in the development of network technology.

 

John D'Ambrosia

Chair, IEEE P802.3bs 400GbE Task Force

Chairman, Ethernet Alliance

Chief Ethernet Evangelist, Dell

rick merritt
User Rank
Blogger
Re: Industry Interest
rick merritt   6/2/2014 7:52:54 PM
NO RATINGS
I'll look forward to what comes out of the event...meanwhile I'd welcome other opinions about the opportunities and challenges here.

Bert22306
User Rank
CEO
Re: Industry Interest
Bert22306   6/2/2014 9:20:00 PM
NO RATINGS
It's always the same story, Rick. The standard usually starts out with a single serial link, then to support faster speeds, it parallels several of the old serial links. Then a faster single serial is developed because multiple parallel ones become a problem to synchronize. Then speeds go higher still, and this newer serial link is again connected in parallel.

You can see this in optical and copper interconnects. In fact, 10 Gb/s Ethernet allows both single and parallel link options even for fiber optic links (where coarse WDM is used to create the parallel lanes, e.g. 10GBASE-LR compared with 10GBASE-LX4). Happens over and over again.

Single lane beats parallel lanes, but requires faster electronics and, at these speeds, would require exclusive use of single-mode fiber to achieve any range at all.

jdambrosia951
User Rank
Rookie
Re: Industry Interest
jdambrosia951   6/2/2014 10:37:16 PM
NO RATINGS
Bert,

The response below doesn't seem in line with where Ethernet has gone.  First, i haven't heard many individuals refer to 10GBASE-LX4 as a success.  Next while it is true that 40G was based on an aggregate of 4x10G, i.e. a prior speed, that is not what happened at 100G.  While 100GBASE-SR10 used 10G, -SR4, -LR4, and -ER4 are based on 4 lanes of 4x25G.  Furthermore, syncronization was solved via lane markers with the IEEE P802.3ba standard.  

THe jump to a 100G serial speed at this time is a very significant one.  SO while you are partially correct with the mix of serial / parallel speeds, we are somewhat more challenged this time to jump to the next serial speeed.

 

 

DaisyCSage
User Rank
Rookie
Re: Industry Interest
DaisyCSage   6/3/2014 6:13:41 PM
NO RATINGS
I­m mak­ing ­over $­1­3k a month working ­part tim­e. I kept hea­ring other p­eople tell me­ how much m­oney they can ma­ke­ online­ so I d­ecide­d to look int­o­ it. Wel­, it was all tru­e and ha­s totally ch­anged­ my life. ­This is­ wha­t I do,

➜➜➜➜➜➜➜➜➜➜➜➜➜➜➜ W­­­W­­W.J­­­O­B­S­7­­­5.C­­­O­M
GO TO THE SITE AND CLICK NEXT TAB FOR MORE INFO AND HELP

krisi
User Rank
CEO
Re: Industry Interest
krisi   6/4/2014 6:48:26 AM
NO RATINGS
John: is there any indication that you can build 100 GHz circuits in CMOS? I remain highly sceptical...Kris

jdambrosia951
User Rank
Rookie
Re: Industry Interest
jdambrosia951   6/4/2014 8:37:50 AM
NO RATINGS
Kris,

I am referring to an effective data rate through some form of modulation, not running at 100 GHz or what i believe you mean 100Gb/s NRZ.

jdambrosia951
User Rank
Rookie
Re: Industry Interest
jdambrosia951   6/4/2014 8:43:01 AM
NO RATINGS
Kris,

Perhaps you want to consider coming to the workshop.  I am chairing the session on the technical issues related to coming up with a solution.  More information can be found on the Ethernet Alliance / OIDA workshop at http://bit.ly/1fNDC8q.

 

Never say never!

krisi
User Rank
CEO
Re: Industry Interest
krisi   6/4/2014 1:56:42 PM
NO RATINGS
thank you John for teh invitation...I was invloved in 10 GHz circuits design while back but for years doing something very different so will have to pass...good luck in designing this, amazing challenge! Kris

krisi
User Rank
CEO
Re: Industry Interest
krisi   6/3/2014 9:54:41 PM
NO RATINGS
I agree Bert and would even go father...100Gb/s single lane is mission impossible...high-speed IO speeds do not follow Moore's law, I am sorry...there is no way you can design 100 GHz CDR in CMOS, even at 7nm...Kris

Flash Poll
Radio
LATEST ARCHIVED BROADCAST
Join our online Radio Show on Friday 11th July starting at 2:00pm Eastern, when EETimes editor of all things fun and interesting, Max Maxfield, and embedded systems expert, Jack Ganssle, will debate as to just what is, and is not, and embedded system.
Like Us on Facebook

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)
EE Times on Twitter
EE Times Twitter Feed
Top Comments of the Week