Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: [HSSG] 10 x 4 = 40


Although I agree that the mechanisms allow multiple projects to go 
forward simultaneously, I think there is a big disconnect.

For 100G, we have seen multiple end users come in and state that they 
need 100G as soon as we can deliver it. On the other hand the 40G demand 
is presented as a possibility further into the future for a different 
market. It seems that, given the timing of expected demand and the state 
of the respective proposals, the project for 100G should move forward 
with the proposed media support and the idea of a project for 40G, 
probably aimed at data centers or something like that, should be 
incubated and brought forward some time in the next 2 - 3 years.

I do not see any reason to delay the project authorization for 100G 
based on some speculation that 40G might be the ideal speed for an 
application where demand is too far out to qualify.


Dove, Dan wrote:

>Its my experience that the IEEE has the means to allow multiple projects
>to operate at the same time. This has been demonstrated numerous times
>whether we are talking about different PMDs for a given MAC, or
>different MACs.
>I believe there will be a solution to this challenge whereby we can move
>forward on both projects and satisfy both unique market sets
>-----Original Message-----
>From: Hays, Robert [mailto:robert.hays@INTEL.COM] 
>Sent: Monday, May 21, 2007 10:00 AM
>Subject: Re: [HSSG] 10 x 4 = 40
>Dan - Your points below support the case we've been making for 40GbE AND
>100GbE objectives...  We need 40GbE server connections for
>cost-optimized storage I/O and 100GbE for network aggregation and
>long-haul connections.  Two different markets, two different sets of
>end-user purchase criteria.
>All - We need to figure out how to move forward with a portfolio of
>standards that meet the needs of the diverse (and growing) applications
>that Etherent is being used for today.  It's not sufficient for IEEE to
>target future speeds of Ethernet at only network aggregation & long haul
>while sacraficing other high-volume, cost-sensitive applications in the
>enterprise.  There seems to be concensus among HSSG participants that
>serve the enterprise server market that 40GbE is a good opporunity in
>addition to 100GbE.  The opposition is coming from folks serving other
>markets that won't benefit from 40GbE.
>How can we move forward together without stalling these standards?  I
>haven't heard any reasonable proposals from the folks opposed to 40GbE.