Tuesday, June 10, 2008

Is Ethernet Really The Answer?

There appears to be a lot of hype promoting Ethernet everywhere for Internet and private networks. However, reality can be quite different. There are a limited number of Ethernet-enable lit buildings in the US or globally.

If your office is in a lit building, colo, or carrier hotel, great! You may be able to get 100 Mbps FastE, GigE or even 10GigE ports. Many businesses are not in lit buildings, but are close enough to get 10 Mbps Ethernet over Copper (EoC) or DS1 (EoDS1) or up to 100 Mbps Ethernet over multiple DS3s. Locations further away from Ethernet POPs may be limited to more traditional options such as DSL, T1, NxT1 or DS3.

Do not think that Ethernet is such a big revelation. This has 2 distinct annoyances. The first.....due to one big "step" it is risky and generates a lot of unknowns. The second.....the fact that Ethernet tends to service the same customer needs demanded 10-15 years ago. If you would go back and check......you will recognize a lot of the same promises compared to ATM and ISDN.

The standard Ethernet protocol did solve, for example, the small packaged sizes of ATM. But, it is lacking solutions in the control plain which are more developed in ATM or even MPLS.

However there are two big differences compared to "back then". The Ethernet hardware technology currently used is relatively cheaper than ATM 10-15 years ago. Telcos are more pressed to drop old hardware. Plus, the TCO rises closer and closer to the total income, for example the 5ESS and PP8600.

From the current communication technologies available the real revolution is technologies like MPLS..... or more so, the 3G mobile phone control plain networks. Where the service is presented at the customer location..... and the core supports a significant management plan to control the services.... it's not as difficult to provide the services. This would be a revolution..... due to the fact that the customer can determine what services they want by the customer location unit. With this step it no longer matters if Ethernet or POTS technologies are used. Its the service that matters.....as in QOS and SLA.

Today.....the largest wireless carriers are all but demanding carrier Ethernet from their wire-line counterparts as part of a growing dilemma around delivering large data/media content to the handset. Over the past couple of years, wireless providers discovered that the data bottleneck was no longer the handset, but the delivery system to the cell site locations. Traditional T-1 architecture is becoming a costly pill that wireless providers don't want to swallow. The alternative is Radio Access Network aggregation to Pseudo Wire (PWE) technology for the current RAN systems..... and eventually native packet delivery over later versions of RAN architecture. All of which would be carried on Ethernet.

Here are a few topics of concern -

1) Will executive management of wire-line companies see past T-1 delivery for the wireless carriers as a whole? Does it make financial sense for them or will they continue to force traditional TDM architecture.

2) Will bonded copper applications have a place in carrier Ethernet delivery, or must we expect fiber to be the only transport medium?

3) Will PWE3 standards be proven as a reliable T-1 emulation and delivery system?

4) Will carrier grade Metro-E find a place in the LEC portfolio? Will latency and jitter issues prevent wireless carriers from moving to this transport medium in an effort to smooth out growing transport costs.

5) Will wire-line companies have to build dual networks or will they be able to support carrier Ethernet over existing architecture.

6) Are their any large local loop carriers today who are positioned well to support dedicated carrier grade Ethernet.

7) If carrier Ethernet, with QOS standards, can be provided as wireless carriers are requesting, are they delusional in expecting the costs to compare with enterprise Ethernet costs? Seems to me that if you want carrier grade, you will have to pay for it.

Remember, there is hardly any significant difference between the ATM phase and the Ethernet phase. The only real differentiation is the price and the amount of influence that cost has on the implementation of the products. The Ethernet products currently still differ a lot...... as did so ATM and DSL once upon a time. Eventually you'll likely see a consolidation of some sort of the several solutions. But by then......history tells us it's probable a new technology will emerge. When that happens......only those operators with services limited to the edge will be flexible enough to change to the "newest" low costs and lower impacts technology.

Michael is the owner of FreedomFire Communications....including DS3-Bandwidth.com and Business-VoIP-Solution.com Michael also authors Broadband Nation where you're always welcome to drop in and catch up on the latest BroadBand news, tips, insights, and ramblings for the masses.

Packages Home Basic Or Business Broadband
Packages Home Basic Or Business Broadband
Streamyx Combo
Streamyx
Streamyx TMpoint
Streamyx Online Registration
Register Online%20home%20or%20business
Streamyx
Streamyx Combo
Register Online%20home%20or%20business
Streamyx Installation Guide
Self%20installation%20guide
Self%20installation%20guide
Register Online%20home%20or%20business
#top Home

0 Comments:

Post a Comment

<< Home