27 April 2010

This gets better as it proceeds.

From Coldwarcomms Mail List: http://groups.yahoo.com/group/coldwarcomms/

To: coldwarcomms[at]yahoogroups.com
From: Mike Cowen <mcowen[at]bolderconcepts.com>
Date: Mon, 26 Apr 2010 16:34:17 -0700
Subject: [coldwarcomms] ATT vs. VZ Reality Check

Need some help here. I read this paragraph in an MSN article about 
ATT vs. VZ. Have I lost my marbles, or is it impossible to get 18 
MBPS through a T1 line? For that matter 18 MBPS is an odd speed in 
the first place. Also, what about a 1 GBPS "limit" over fiber? I 
thought fiber speed was only dependent on how much you invested in 
the electronics at either end. It seems the author's gone 
nonlinear. What say you experts? Thanks!

Mike

"One area AT&T is focusing on is supplying faster connections to its 
cell towers. Currently, 38% of its fastest 3G towers are fed by 
fiber-optic cable. The company expects this number to rise to 90% by 
2011, which should alleviate some of the pressure that so annoys AT&T 
subscribers. Fiber offers connections speeds of up to 1 gigabit per 
second, or gbps, compared with the 18 megabits per second (a megabit 
is one-thousandth of a gigabit) that more common T1 connections top out at."

http://articles.moneycentral.msn.com/learn-how-to-invest/att-vs-verizon-
which-stock-is-better.aspx



To: coldwarcomms[at]yahoogroups.com
From: Jake <jakematic[at]bellsouth.net>
Date: Mon, 26 Apr 2010 19:53:06 -0400
Subject: Re: [coldwarcomms] ATT vs. VZ Reality Check

Last I checked T1 was 1.5Mbps.

Both single and multi-mode fibre are limited in speed by distance, laser, and 
electronics.

WDM changes the equation some. Here is the list I use to plan site to site link 
speed requirements for DR:

Topology Rated Speed MBps
T1/E10.15
T3/E35
OC16
OC-319
OC-1277
OC-48310
OC-96620
OC-1921,240
DWDM12.5 TBps

-Jake



To: coldwarcomms[at]yahoogroups.com
From: Mark Shostak <shostakmark[at]gmail.com>
Date: Mon, 26 Apr 2010 19:11:37 -0500
Subject: Re: [coldwarcomms] ATT vs. VZ Reality Check

Mike,

I figure the author didn't become a stock analyst because he knew which end
of a flux capacitor to plug into the wall.

The information he's regurgitating may be from a source that's factoring in
knowledge of the specific equipment in the mobile back-haul solution.
Another words the equipment may only terminate 12 T1s, resulting in an
aggregate bandwidth of 18Mbps "for T1s". Similarly, the equipment may be
using GbE over fiber, which limits the fiber to 1Gbps.

Technically, you are correct, a T1 formatted signal is still 1.5544 Mbps and
fiber without a specific format is limited by the number of lambdas and
equipment that happens to be terminating each one.

HTH,
-Mark



To: coldwarcomms[at]yahoogroups.com
From: Mike Harpe <mike[at]mikeharpe.com>
Date: Mon, 26 Apr 2010 20:17:17 -0400
Subject: Re: [coldwarcomms] ATT vs. VZ Reality Check

A T-1 is a 1.544 MB/sec circuit. They are probably referring to the number
of T-1s that it makes sense to bundle together before it makes economic
sense to pull fiber to the towers. Fiber pulls to some towers, especially in
rural areas, can be very, very expensive.

Fiber optic cable speeds can run into many gigabits per second. Plus, A T-1
used for Internet service can carry a relatively small number of connections
while a fiber backbone can carry hundreds of thousands of simultaneous
connections. Fiber technologies also allow a multitude of possible
configurations of channels and routes inside one physical cable. The
configurations can get very sophisticated.

Yes, the equipment to do really high speeds is quite expensive but this
isn't what controls the speeds. One really interesting technology used in
fiber optics is called DWDM for Dense Wave Division Multiplexing. Put
simply, each possible color of the light going through a cable can represent
a channel! The equipment for this technology is quite expensive. Other more
mature technologies don't cost as much,

AT&T needs fiber to the towers to support the incredible traffic load
generated by iPhones and other mobile devices. Verizon also uses a slightly
different implementation of 3G that doesn't require quite as much bandwidth.
Little differences mean a lot when you're talking about all those phones out
in the wild.

Mike Harpe, CCNA, Network Engineer
Sellersburg, IN



To: <coldwarcomms[at]yahoogroups.com>
From: Richard Wells <w2hn[at]hotmail.com>
Date: Mon, 26 Apr 2010 20:20:28 -0400
Subject: RE: [coldwarcomms] ATT vs. VZ Reality Check

I believe the capacity of a 3G site is around 8 Mbps in the download direction.  
That's the capacity limitation, not the connecting links.  To expand capacity 
requires a modulation scheme that makes more efficient use of available 
bandwidth or increased bandwidth. 

W2HN



To: coldwarcomms[at]yahoogroups.com
From: "David I. Emery" <die[at]dieconsulting.com>
Date: Mon, 26 Apr 2010 21:07:35 -0400
Subject: Re: [coldwarcomms] ATT vs. VZ Reality Check

On Mon, Apr 26, 2010 at 08:20:28PM -0400, Richard Wells wrote:

> I believe the capacity of a 3G site is around 8 Mbps in the download
> direction. That's the capacity limintation, not the connecting links. 
> To expand capacity requires a modulation scheme that makes more
> efficient use of available bandwidth or increased bandwidth. 

I think this would be more accurate if you said "channel" or
"carrier". I believe one site can use multiple downlink carriers in
different channels on different frequencies and directions... depending
on what spectrum the carrier controls...

So an actual cell site (with directional antennas for illuminating
different cells) might be well over 8 Mb/s.



To: coldwarcomms[at]yahoogroups.com
From: Bill Smith <brscomm[at]yahoo.com>
Date: Mon, 26 Apr 2010 20:11:21 -0700 (PDT)
Subject: Re: [coldwarcomms] ATT vs. VZ Reality Check

Typical downlink speeds for EVDO are 400kbps to 3Mbps for Rev A. 



To: coldwarcomms[at]yahoogroups.com
From: David <wb8foz[at]nrk.com>
Date: Mon, 26 Apr 2010 23:37:27 -0400
Subject: [coldwarcomms] ATT vs. VZ Reality Check

I know a guy who brings up err "why resistor" make that "unencrypted" brand 
3G sites. [You can figure it out...]

He mentioned that everything in the site was built for 20Mb/s per sector, 
and best case one mobile could suck 12. Sites had different #'s of sectors, 
but the beancounters had built up MANY sites; THEN said "gee, we need 
backhauls, don't we?"

And guess what, Verizontal was then falling all over themselves to not 
help. So he had $MANY sites ready to go, but no pipes. They brought up some 
cities on 3-4 DS1's/tower.....

Independently, I heard that even though Comcast *owns* part of resistor, 
and had fiber going by many locations; there was some political disconnect 
so that nope, it was not used for any backhauls.

Last move was they were going to buy into shorthaul tower to tower 
microwave, a technology that "resistor" people knew nothing about.



To: coldwarcomms[at]yahoogroups.com
From: blitz <blitz716[at]sonic.net>
Date: Tue, 27 Apr 2010 03:32:01 -0400
Subject: Re: [coldwarcomms] ATT vs. VZ Reality Check

Thats why you pull fiber to a central site where you can link to other 
sites via microwave. Microwave is by far cheaper in the long run, for 
those who are bright enough to implement it.

Mike Harpe wrote:
> A T-1 is a 1.544 MB/sec circuit. They are probably referring to the number
> of T-1s that it makes sense to bundle together before it makes economic
> sense to pull fiber to the towers. Fiber pulls to some towers, especially in
> rural areas, can be very, very expensive. [Snip]



To: coldwarcomms[at]yahoogroups.com
From: blitz <blitz716[at]sonic.net>
Date: Tue, 27 Apr 2010 03:40:39 -0400
Subject: Re: [coldwarcomms] ATT vs. VZ Reality Check

I dont know where those speeds came from but they are just plain WRONG!

Ref:
http://en.wikipedia.org/wiki/Optical_Carrier_transmission_rates

Optical Carrier classifications are based on the abbreviation OC 
followed by a number specifying a multiple of 51.84 Mbit/s: n × 51.84 
Mbit/s => OC-n. For example, an OC-3 transmission medium has 3 times the 
transmission capacity of OC-1.
[edit] OC-1

OC-1 is a SONET line with transmission speeds of up to 51.84 Mbit/s 
(payload: 50.112 Mbit/s; overhead: 1.728 Mbit/s) using optical fiber.
[edit] OC-3 / STM-1x

OC-3 is a network line with transmission speeds of up to 155.52 Mbit/s 
(payload: 148.608 Mbit/s; overhead: 6.912 Mbit/s, including path 
overhead) using fiber optics. Depending on the system OC-3 is also known 
as STS-3 (electrical level) and STM-1 (SDH).

When OC-3 is not multiplexed by carrying the data from a single source, 
the letter c (standing for concatenated) is appended: OC-3c.
[edit] OC-3c

OC-3c ("c" stands for "concatenated") concatenates three STS-1(OC-1) 
frames into a single OC-3 look alike stream. The three STS-1 (OC-1) 
streams interleaved with each other such that the first column is from 
the first stream, the second column is from the second stream, and the 
third is from the third stream. Concatenated STS(OC) frames carry only 
one column of path overhead because they cannot be divided into finer 
granularity signals. Hence, OC-3c can transmit more payload to 
accommodate a CEPT-4 139.264 Mbit/s signal. The payload rate is 149.76 
Mbit/s and overhead is 5.76 Mbit/s.
[edit] OC-12 / STM-4x

OC-12 is a network line with transmission speeds of up to 622.08 Mbit/s 
(payload: 601.344 Mbit/s; overhead: 20.736 Mbit/s).

OC-12 lines are commonly used by ISPs as WAN connections. While a large 
ISP would not use an OC-12 as a backbone (main link), it would for 
smaller, regional or local connections. This connection speed is also 
often used by mid-sized (below Tier 2) internet customers, such as web 
hosting companies or smaller ISPs buying service from larger ones.
[edit] OC-24

OC-24 is a network line with transmission speeds of up to 1244.16 Mbit/s 
(payload: 1202.208 Mbit/s (1.17 Gbit/s); overhead: 41.472 Mbit/s). 
Implementations of OC-24 in commercial deployments are rare.
[edit] OC-48 / STM-16x / 2.5G SONET

OC-48 is a network line with transmission speeds of up to 2488.32 Mbit/s 
(payload: 2405.376 Mbit/s (2.349 Gbit/s); overhead: 82.944 Mbit/s).

With usually cheap interface prices and being faster than OC-3, OC-12 
connections, and even surpassing gigabit Ethernet, OC-48 connections are 
used as the backbones of many regional ISPs. Interconnections between 
large ISPs for purposes of peering or transit are quite common. As of 
2005, the only connections in widespread use that surpass OC-48 speeds 
are OC-192 and 10 gigabit Ethernet.

OC-48 is also used as transmission speed for tributaries from OC-192 
nodes in order to optimize card slot utilization where lower speed 
deployments are used. Dropping at OC-12, OC-3 or STS-1 speeds are more 
commonly found on OC-48 terminals, where use of these cards on an OC-192 
would not allow for full use of the available bandwidth.
[edit] OC-192 / STM-64x / 10G SONET

OC-192 is a network line with transmission speeds of up to 9953.28 
Mbit/s (payload: 9621.504 Mbit/s (9.396 Gbit/s); overhead: 331.776 Mbit/s).

A standardized variant of 10 gigabit Ethernet (10GbE), called WAN-PHY, 
is designed to inter-operate with OC-192 transport equipment while the 
common version of 10GbE is called LAN-PHY (which is not compatible with 
OC-192 transport equipment in its native form). The naming is somewhat 
misleading, because both variants can be used on a wide area network.

As of 2005, OC-192 connections are most common for use on backbones of 
large ISPs.
[edit] OC-768 / STM-256x

OC-768 is a network line with transmission speeds of up to 39,813.12 
Mbit/s (payload: 38,486.016 Mbit/s (37.584 Gbit/s); overhead: 1,327.104 
Mbit/s (1.296 Gbit/s)).

On October 23, 2008, AT&T announced the completion of upgrades to OC-768 
on 80,000 fiber-optic wavelength miles of their IP/MPLS backbone 
network.[3] OC-768 SONET interfaces have been available with short-reach 
optical interfaces from Cisco since as early as 2006. Infinera made a 
field trial demonstration data transmission on a live production network 
involving the service transmission of a 40 Gbit/s OC-768/STM-256 service 
over a 1,969 km terrestrial network spanning Europe and the U.S. In 
November 2008, an OC-768 connection was successfully brought up on the 
TAT-14/SeaGirt transatlantic cable,[4] with the longest hop being 7,500km.
[edit] References

1. ^ "Synchronous Optical Network (SONET)". Web ProForums. 
International Engineering Consortium. 
http://www.iec.org/online/tutorials/sonet/topic01.html. Retrieved 
2007-05-25.
2. ^ Cartier C, Paynetitle T (2001-07-30). "Optical Carrier levels 
(OCx)". 
http://searchnetworking.techtarget.com/sDefinition/0,,sid7_gci212685,00.html. 
Retrieved 2007-05-25.
3. ^ "AT&T Completes Next-Generation IP/MPLS Backbone Network, 
World's Largest Deployment of 40-Gigabit Connectivity". AT&T press 
release, October 23, 2008. Accessed 28 January 2009.

Jake wrote:
> Last I checked T1 was 1.5Mbps. [Snip]



To: coldwarcomms[at]yahoogroups.com
From: Jake <jakematic[at]bellsouth.net>
Date: Tue, 27 Apr 2010 09:42:31 -0400
Subject: Re: [coldwarcomms] ATT vs. VZ Reality Check

My list is in MBps, not Mbps. 1 MBps = 8 Mbps.

Since I deal with data rates of change and the pipe size required for 
synchronous replication, it really should be in GBps these days.

Sorry for the confusion.
 
-Jake

On Apr 27, 2010, at 3:40 AM, blitz wrote:

> I dont know where those speeds came from but they are just plain WRONG! [Snip]



To: coldwarcomms[at]yahoogroups.com
From: "widebandit" <widebandit[at]yahoo.com>
Date: Tue, 27 Apr 2010 23:41:49 -0000
Subject: [coldwarcomms] Re: ATT vs. VZ Reality Check

> I dont know where those speeds came from but they're just plain WRONG!

Ah, but then you're forgetting that an STS-1 payload envelope (PE) contains 
a column of nine path-overhead (POH) bytes, reducing the usable PE bandwidth 
to 86 columns per frame or 51.84 x (86/90) = 49.536 Mb/s - 6.1 MB/s.

Concatenating three STS-1s into an STS-3c eliminates two POH columns so that 
260 columns are available for payload. The STS-3c payload rate is 155.52 x 
(260/270) = 149.76 Mb/s - 18.7 MB/s.

When four STS-3c's are concatenated into an STS-12c all four POH columns remain, 
but three are empty. The STS-12c payload rate is 149.76 x 4 = 599.04 Mb/s - 74.9 MB/s. 
In the early days of SONET, many ISPs got dinged on their service agreements by 
guaranteeing 600 Mb/s through-put on an OC-12 pipe but delivering only 599 Mb/s. 

Even wikipedia doesn't quite get SONET payload rates right.

The treatment of POH bytes in the concatenation process brings STS-3c frames into 
alignment with European SDH STM-1. The SDH (Synchronous Digital Hierarchy) standard 
does not define an STS-1 equivalent. SDH line rates are:

STM-1 155.52 Mb/s
STM-4 622.08 Mb/s
STM-82,488.32 Mb/s
STM-169,953.28 Mb/s

The 5-ESS is SDH capable - it can deliver E-1s (2.048 Mb/s) in STM-X containers.

Europe adopted STM-1 as their first level to provide a PE that could transport the 
CEPT (European Conference of Postal and Telecommunications - French syntax) E-4 line 
rate of 139.264 Mb/s.

The STS-3c/STM-1 alignment also explains why STS-1's can only be concatenated as 
integer multiples of the number three - 1-3, 4-6, 7-9, 10-12, 46-48… and not 2-4, 
11-13, or 45-47.

Americans see SDH as a subset of SONET, Europeans see SONET as a subset of SDH, but 
they were developed together.

Transporting Gb Ethernet over SONET requires 21 STS1s in virtual concatenation – 
a payload rate of 1,048.32 Mb/s. They need not be contiguous within the OC48/OC192 
but they must be combined by the 3-rule above.

The very first long-haul fiber gear that I worked (now I'm dating myself) was the 
Rockwell-Collins LTS-3139; transporting 3 DS3s at an optical line rate of 139.264 Mb/s 
so it could operate through European E-4 optical repeaters. R-C ultimately produced 
PDH optical gear running at 565 Mb/s and 1,130 Mb/s before converting to SONET and 
being absorbed by Alcatel. The 1,130 Mb/s gear was a tentative step into the synchronous 
world - it was actually two 565 Mb/s multiplexers synchronized to a common clock - the 
two streams were probably bit-interleaved.

By the time SONET came along, most vendors had been producing PDH optical gear running 
at hundreds of Mb/s so there was no need for OC-1. First-generation SONET gear ran at 
OC-3 and was usually named something like ADM-150 or ADM-155. ADM - Add-Drop Multiplexer - 
SONET's ability to add/drop single DS-1s without de-muxing the whole payload - now that 
DS3 muxes are pizza boxes, that capability is relatively unused.

PDH - Plesiochronus Digital Hierarchy.
Plesiochronous - almost but not quite not nearly unlike synchronous (with thanks to 
Douglas Adams)....waw...

> Ref:
> http://en.wikipedia.org/wiki/Optical_Carrier_transmission_rates [Snip]