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

RE: [10GBASE-T] Economic feasibility



Title: Message
Pat,
 
What are your thoughts on the future trends/needs for 10GBASE-T as the connection to wireless access points?  I know currently most APs run 10/100 into them, but I'm trying to figure out if the trends in the wireless market (to things like higher bandwidths, direction antennae, greater user support, etc.) that there will be a growing need for 10G into these "concentrators".
 
If this is a potential market, then we're looking at data center connections (server to switch and switch to switch), concentrator connections (wireless APs), and storage connections (iSCSI, etc.).  Blade systems could also make use of this.  With the growing trend in the blade environment, the concentration is done in the system and multiple 10GBASE-T ports may be required.
 
Thoughts?
 
Thanks,
Brad
-----Original Message-----
From: pat_thaler@agilent.com [mailto:pat_thaler@agilent.com]
Sent: Friday, August 01, 2003 12:16 PM
To: GEisler@aol.com; btolley@cisco.com; yousefi@broadcom.com; Booth, Bradley; stds-802-3-10gbt@ieee.org
Subject: RE: [10GBASE-T] Economic feasibility

George,
 
Economic Feasibility doesn't address market. Broad market potential does.
 
If one bases the broad market potential on connecting aggregation elements, the economic feasibilty argument should show the technology has feasibility for that market. If one bases broad market potential on desk top, then it better have economic feasibiity for that market (which requires lower cost than the aggregation interconnect market).
 
The 5 criteria and objectives aren't each isolated items. They need to be consistant with each other. If they aren't, one is likely to produce a standard that doesn't get used.
 
With 10GBASE-T, I don't think either of those is the basis for broad market potential.
 
If we are successful, then some 10GBASE-T will be used for aggregation interconnects, but most of the aggregation will be of a distance that requires fiber or within the closet at short enough distances for CX-4. I don't think the niche between those two in the aggregation interconnects space is big enough to justify broad market potential. 10GBASE-T will presumably be too expensive for desktops for some time and it will be quite a while before desktops need 10 Gig. Also, there isn't much point in putting a 10 Gig adapter into a desktop until desktops start having PCI Express slots (or one of the other high speed alternatives). 
 
The market where 10GBASE-T will be very useful initially is the data center to connecting servers and, with the help of iSCSI and other IP storage initiatives, storage devices.
 
The objectives, technical feasibility and economic feasibility should address the needs of that market.
 
Regards,
Pat
-----Original Message-----
From: GEisler@aol.com [mailto:GEisler@aol.com]
Sent: Thursday, July 31, 2003 10:06 PM
To: btolley@cisco.com; pat_thaler@agilent.com; yousefi@broadcom.com; bradley.booth@intel.com; stds-802-3-10gbt@ieee.org
Subject: Re: [10GBASE-T] September interim meeting

Guys,

I would like to point at the Economic Feasibility bullets:

- Cost factors known, reliable data
- Reasonable cost for performance expected
- Total installation costs considered

None of these are concerned with how many units will be sold or whether there will be payback on development costs, etc. I think the word Economic is being misinterpreted in the present discussion.

Further, the need for 10Gig arises from increasing data speeds and volume. With 1000BASE-T about to be rolled out to desktops, the aggregation network elements need to go faster. That's the way it has worked at previous levels; why not now?
And if not, why was the ae standard created and adopted with four different PHY's?

George Eisler