I've noticed over the last 3 years or so that TDM, specifically T-1, access
and transport has been in a steady decline. Customers are moving to FTTH
and cable, or going WiMAX and Metro-Ethernet. Ethernet seems to have taken
an even bigger bite out of DS-3. The bigger pipes seem to favor ethernet. A
recent upgrade from OC-3 to GigE transport actually saved us a large chunk
of money.
I'm wondering if others are seeing the same behavior, if it's
market-dependant, or if I'm just imagining things. I'm working on building
new infrastructure and my current thoughts are to minimize my TDM
footprint. It would be useful to get a better feel if this is an overall
trend or something local.
Thoughts?
Thanks,
Rick Ernst expunged (nanog@shreddedmail.com):
I'm wondering if others are seeing the same behavior, if it's
market-dependant, or if I'm just imagining things. I'm working on building
new infrastructure and my current thoughts are to minimize my TDM
footprint. It would be useful to get a better feel if this is an overall
trend or something local.
You aren't imagining things. In fact, some large national networks have been designed to support solely ethernet. It comes down to cost, as always....
-Steve
Speaking of which, what is the state of voip-over-cellular as essentially the
last holdout of TDM? Will the new 4G stuff be able to support latencies, etc?
Has the work on handovers-over-IP matured enough that it's viable?
Mike
Speaking of which, what is the state of voip-over-cellular as essentially the
last holdout of TDM? Will the new 4G stuff be able to support latencies, etc?
Has the work on handovers-over-IP matured enough that it's viable?
Mike
Steve Meuse wrote:
I'm wondering if others are seeing the same behavior, if it's
market-dependant, or if I'm just imagining things. I'm working on building
new infrastructure and my current thoughts are to minimize my TDM
footprint. It would be useful to get a better feel if this is an overall
trend or something local.
You aren't imagining things. In fact, some large national networks have been des
igned to support solely ethernet. It comes down to cost, as always....
-Steve
Actually, a lot of people would be shocked at just how much VoIP is now
used to transport voice with TDM only occurring at the last mile and in
many cases at the last foot. Anyone designing a voice infrastructure
would be best to design it for VoIP. Your ROI is much much greater. If
you need to use TDM, then do so only at the edge as close to the TDM
equipment as possible.
Of course if you are going to use VoIP through-out an infrastructure it
certainly is a good idea to get familiar with QoS provisioning.
Bret
I tend to think this is market dependent. In major population centers, TDM service may well be on the decline, but I'd suspect that Ethernet based services have a much lower penetration in areas with lower population densities.
I don't see TDM going away entirely any time soon because it still comes in handy for things like out-of-band management, etc, plus nowadays there is lots of TDM gear on the secondary market that can be picked up dirt-cheap.
jms
I've noticed over the last 3 years or so that TDM, specifically T-1, access
and transport has been in a steady decline. Customers are moving to FTTH
and cable, or going WiMAX and Metro-Ethernet. Ethernet seems to have taken
an even bigger bite out of DS-3. The bigger pipes seem to favor ethernet. A
recent upgrade from OC-3 to GigE transport actually saved us a large chunk
of money.
I'm wondering if others are seeing the same behavior, if it's
market-dependant, or if I'm just imagining things. I'm working on building
new infrastructure and my current thoughts are to minimize my TDM
footprint. It would be useful to get a better feel if this is an overall
trend or something local.
Why I think it comes down to is do you want to use frame-relay, atm, sdh and ethernet when you can just use ethernet?
lan-phy ethernet has economies of scale that result in lower cost along virtually every dimension relative to the alternatives.
From: Rick Ernst [mailto:nanog@shreddedmail.com]
an even bigger bite out of DS-3. The bigger pipes seem to favor
ethernet. A recent upgrade from OC-3 to GigE transport actually saved
us a large
chunk of money.
We recently had exactly the opposite experience, unfortunately. During
relocation of our datacenter we asked our MPLS WAN provider to provide
GigE transport rather than the multiple OC-3s we were using today. To us
it seemed like it would be cheaper - GigE interfaces, even WAN-PHY rated
ones, were orders of magnitude cheaper than SONET. Unfortunately the
provider came back with absolutely outrageous costs for the port,
claiming they had to do non-standard agreements with incumbents to
provide the lines to us (despite the amount of circuits already in the
site).
This may be more a function of that particular provider, however.
Funny thing about this is we have been steadily getting rid of all of our t1 and ds3 circuits and replacing them with metro-e or cable based services at much better price/Mbs. However, when we went to VOIP and wanted to do sip trunking with qwest, they needed to deliver this over t1, otherwise is wasn't cost effective.
Dylan Ebner
Dylan Ebner expunged (dylan.ebner@crlmed.com):
Funny thing about this is we have been steadily getting rid of all of our t1 and ds3 circuits and replacing them with metro-e or cable based services at much better price/Mbs. However, when we went to VOIP and wanted to do sip trunking with qwest, they needed to deliver this over t1, otherwise is wasn't cost effective.
E911 was a sticky point for us. We ended up having to install some small boxes to do T1 handoff to our provider of choice, they didn't have any ethernet options (which we pushed for).
-Steve
What I've been hearing rumors of is ---
unregulated services (eg: gigaman, opteman) typically have a better price-point if you are going with the carrier of choice.
TDM services (DSn/OCn) where there is a standard interconnection method tend to have higher costs than ethernet services, but are available when you have multiple carriers involved. (eg: VZ/MCI/XO/QWEST to SBC/ATT) territory.
I see this as a two-fold issue, one, the carriers (ATT) are trying to provide an incentive for shifting away from the TDM based services. At the same time, it's more difficult to deliver service if you're not building to the market.
I would take into account the filing that ATT gave to the FCC recently asking to set a sunset date for their POTS (read: TDM) network elements. This will allow them to leave the markets that are unprofitable, while delivering the unregulated (ethernet/IP) services where it currently is profitable.
- Jared
All the same, the price of gear dropping through the floor on the secondary
market is usually a pretty good indication that the technology is played out,
because supply-and-demand says that demand for the gear will keep the price
propped up until the demand goes away - which usually means the tech is
played out.
Anybody got a counter-example? 
I don't know too many people who build their out-of-band management networks out of the latest and greatest gear. There are tons of Cisco 2511s, 2611s, etc that have been given second lives as terminal servers.
There are also lots of areas where Ethernet transports are either ridiculously expensive because the carrier wants me to partially subsidize their initial build costs, or it's just flat-out not available. There are also areas where I've seen it delivered over n x STS-1's bonded together, so the Ethernet frames are still riding over a SONET transport in many cases.
jms
Diversity, in case of locations meant to be used as each other's backup/failover
***Stefan Mititelu
http://twitter.com/netfortius
http://www.linkedin.com/in/netfortius