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

Re: [HSSG] Topics for Consideration



Eric,

The reaches are definitely open to debate.  The long-haul and metro
spaces will be easier for HSSG as these are SMF environments and are
likely to only require one port type.  The data center is going to be a
bit tougher as the environment is a mix of MMF and SMF, and there are a
couple cabling standards that list the maximum reaches.  As you observed
though, the 10GBASE-CX4 solution has done very well for the <= 15m
reach.  There is probably a better segregation based on the usage in a
data center.  For example, intra-rack, inter-rack, horizontal runs and
vertical risers.  The hope would be that a single low-cost solution
might be able to span a couple of the reaches/market requirements.

Thanks,
Brad

-----Original Message-----
From: Eric B. Grann [mailto:ebgrann@ONP-OMRON.COM] 
Sent: Wednesday, August 09, 2006 9:22 AM
To: STDS-802-3-HSSG@listserv.ieee.org
Subject: Re: [HSSG] Topics for Consideration

Brad,

Your initial stab at distance criteria is a great start.

Long-Haul	--> 100+ km
Metro		--> 10+ km
Data Center --> 50m & 300m

If we can focus on gathering information to validate these distances, we
will be one step closer.  The 10km and 300m distance targets have a fair
amount of history from 802.3ae.  Do we have enough information from 10G
Ethernet port installations to validate these distances now?

For the long haul, we had many system providers pushing for 80km versus
40km at 10G.  Is 100km the mark now?  

In the data center room, it seems that 50m is long.  Based on 10G
copper's success in addressing the short distance data center needs, it
seems that 25m might be a better target.  Can the data center managers
comment on distribution of fiber run distances here.

Regards

Eric

Eric Grann
Omron Network Products

-----Original Message-----
From: Brad Booth [mailto:bbooth@QUAKETECH.COM] 
Sent: Tuesday, August 08, 2006 4:36 PM
To: STDS-802-3-HSSG@listserv.ieee.org
Subject: Re: [HSSG] Topics for Consideration

While it is useful to discuss the technologies that may be used, the
goal of the HSSG is to create a set of  objectives, 5 critters and
PAR(s).  For that, it is important to understand the market that is
looking at deploying higher speed Ethernet.  From some of the
information shown to date, there appears to three markets: long-haul
wide area, metro, and data center.  The long-haul wide area seems
interested in the 100+ km reach.  Metro appears to be the 2-10 km reach.
Data center has a bit of a mix on reach requirements with something like
70% of the cabling runs in data centers being less than 50 m, but also
the need for longer runs possibly up to 300 m.

If the HSSG can agree on reaches that will satisfy the market demands,
without getting too worried about media and technology at this point in
time, then it will certainly help narrow the scope of work relative to
the objectives and 5 criteria.  It can be difficult to leave the media
and technology discussion out.  Heck, we're engineers and there's a
problem to be solved.  Speaking from past experience though, it can make
the study group's job easier to tackle the bigger questions before they
have to get into the nitty gritty details.

Cheers,
Brad

-----Original Message-----
From: OJHA,JUGNU [mailto:jugnu.ojha@AVAGOTECH.COM] 
Sent: Tuesday, August 08, 2006 3:53 PM
To: STDS-802-3-HSSG@listserv.ieee.org
Subject: Re: [HSSG] Topics for Consideration

Or, as some of us have been suggesting, we just use existing PMDs.  

-----Original Message-----
From: Roger Merel [mailto:roger@LUXTERA.COM] 
Sent: Tuesday, August 08, 2006 3:49 PM
To: STDS-802-3-HSSG@listserv.ieee.org
Subject: Re: [HSSG] Topics for Consideration

This pre-supposes that there will be PMD technology solutions at these
distance "break points" which account for substantially different
economics.

If a single PMD works for up to 5km with reasonable economics, there may
not be a need for 3 of the PMDs.  We need presuppose how many until we
start to understand the technologies in contention, the distances they
support, and their relative economics.

Then we just need a 2nd PMD to cover 5-40km.

This seems quite reasonable and likely.




-----Original Message-----
From: Menachem Abraham [mailto:mabraham@COLUMBUSADVISORS.COM] 
Sent: Monday, August 07, 2006 5:00 PM
To: STDS-802-3-HSSG@listserv.ieee.org
Subject: Re: [HSSG] Topics for Consideration

Greg,

Based on your input it looks like we need to develop VSR, SR, LR, VLR
all at once so that we can address all of the "first wave" application
spaces shown in the CFI presentation.

If we had to prioritize our work in 802.3 what would be your view on the
relative importance of these?

Regards,
Menachem
Sent from my Blackberry 


-----Original Message-----
From: "OJHA,JUGNU" <jugnu.ojha@AVAGOTECH.COM>
Date:         Tue, 8 Aug 2006 13:46:45 
To:STDS-802-3-HSSG@listserv.ieee.org
Subject: Re: [HSSG] Topics for Consideration

Greg, 

Should campus backbone include LR and VLR?  I could see it needing a few
kms, but beyond this, doesn't it really become a "Corporate Backbone"?  

Jugnu 

-----Original Message-----
From: Greg Hankins [mailto:ghankins@FORCE10NETWORKS.COM] 
Sent: Tuesday, August 08, 2006 1:38 PM
To: STDS-802-3-HSSG@listserv.ieee.org
Subject: Re: [HSSG] Topics for Consideration

Based on my experience with IX and ISP customers I see these reach
targets
in each application.

(1) Internet Exchange = VSR/SR, ISP Backbone = VSR/SR/LR/VLR , Content
Providers = VSR/SR/LR

(2) ISP Aggregation = VSR/SR/LR, Supercomputing / R&D = VSR/SR,
Broadband Aggregation = VSR/SR/LR

(3) Campus Backbone = SR/LR/VLR, Corporate Backbone = SR/LR/VLR,
Corporate Data Center = VSR/SR

Greg

-- 
Greg Hankins <ghankins@force10networks.com>
Technical Marketing Engineer
Force10 Networks
+1 404 542 5530

-----Original Message-----
Subject: Re: [HSSG] Topics for Consideration
Date: Tue, 8 Aug 2006 07:38:19 -0700
From: "Menachem Abraham" <mabraham@COLUMBUSADVISORS.COM>
To: <STDS-802-3-HSSG@listserv.ieee.org>

John,

 

In the CFI presentation there was a chart showing the expected market
adoption of a higher speed Ethernet over time.

 

First wave shown was: Internet Exchanges, ISP Backbones, Content
Providers

 

Second wave shown was: ISP Aggregation, Supercomputing / R&D, Broadband
Aggregation

 

Third wave shown was: Campus Backbone, Corporate Backbone, Corporate
Data Center

 

In the same presentation there were categories for reach targets: 

 

Backplane up to 1m

VSR up to 25 m

SR up to 100 m

LR up to 40 Km

VLR beyond 40Km

 

Assuming that the market adoption projection above is correct we should
match the reach definitions to these applications and see what that
means regarding the priorities for our work, i.e. :

 

(1) Internet Exchange = SR, ISP Backbone = ? , Content Providers = SR

 

(2) ISP Aggregation = ?, Supercomputing / R&D = SR, Broadband
Aggregation = ?

 

(3) Campus Backbone = LR, Corporate Backbone = VLR, Corporate Data
Center = SR

 

What are the answers to the question marks above? 

 

Regards,

Menachem

 

 

 

_____  

From: John DAmbrosia [mailto:jdambrosia@FORCE10NETWORKS.COM] 
Sent: Wednesday, August 02, 2006 8:44 AM
To: STDS-802-3-HSSG@listserv.ieee.org
Subject: [HSSG] Topics for Consideration

 

All,

I would like to thank all those who have joined the reflector and will
be bringing information and requirements to the table in the future.  I
am sure we are going to have a lot of "fun" in the future.

 

During our first meeting, I anticipate spending a lot of time focusing
on objectives.  At the closing plenary I highlighted two issues /
objectives that the SG would have to consider:

 

*         Tradition of 10x leap in speed

*         Multiple Reach Targets

 

It was also presented that the focus of this effort wasn't for a desktop
application, and that the cost model needs to be considered.

 

I want to encourage everyone to start thinking about these areas and to
use the reflector for discussion.  This will assist everyone in
providing well-thought out presentations, which will make for a better
and more productive meeting in September.  :-)

 

John D'Ambrosia

Force10 Networks

Phone - 717-460-8535