One of the coolest things at N+1 this year was a demonstration of 10GBASE-T technology. If you happened to go by the Ethernet Alliance's pavilion, you probably saw some very cool interoperability demonstrations. Support for the interoperability was provided by the UNH Interoperability Lab. 10GBASE-T technology is based on the IEEE standard 802.3an® and supports 10 Gigabits per second (Gbps) operations on 100m of copper cabling.
This standard is likely to revolutionize the data center and will certainly provide a means to move 10G to the desktop down the road. With the increase in bandwidth available, applications now have room to grow to their potential. For years we have had to compress, add memory, add more network cards and so on to get around a slow network for the applications we use today. But since applications double every 18 months (Gates' Law), and the complexity and functionality change and grow, who knows what this bandwidth can mean to us as end users.
It could mean full collaboration with video, voice and data, as provided by the new Cisco TelePresence meeting application (which, by the way, was a fantastic demo if you were lucky enough to see it). Or perhaps medicine will advance to where results are instantaneous and provide accuracy above and beyond what is available or possible to date. Who knows what the future holds?
While on the subject of the Ethernet Alliance, for those of you who don't know much about this organization, "The Ethernet Alliance mission is to promote industry awareness, acceptance, and advancement of technology and products based on both existing and emerging IEEE 802 Ethernet standards and their management." Their Web site is full of non-vendor-specific information on standards and applications, and there is an excellent university program. I encourage you to make use of this valuable resource.
Now back to 10GBASE-T -- don't think your organization will ever use it? You could be making a mistake in thinking so. For years, people have modeled networks based on percent utilization -- but think back 10 or 15 years. At that time, we were using mainframe green-screen applications and DOS with network speeds of 4-16Mbps token ring and 10Mbps Ethernet. Clearly, systems have evolved, and so should you if you're planning to be around another 10 years. Preparing for 10GBASE-T applications will prove invaluable if you are considering a long road ahead. When looking at percent utilization, you need to be able to determine what is actually going on behind the numbers. For instance, is the user at his desk? Is he using the network, or is it simply a mail client going back and forth checking mail? Is the user working locally and saving to the network only once in a while because it is slow? Unless you know all of these factors, you are missing a large part of the equation.
With 10GBASE-T, the NIC cards will auto-negotiate between the various Ethernet speeds. These NICS will therefore achieve a longer lifecycle when compared with fiber, which requires a hardware change to migrate from one speed to another. Currently there are various cabling standards surrounding 10GBASE-T, and the only published standard supporting 10GBASE-T for a full 100m is the ISO standard for Category 7/class F in 11801 (which was published at the same time as the Category 6 standard). There are two new pending standards for Category 6A systems -- a shielded and an unshielded version -- but these will probably not publish until next year. As they are written today, they are channel specifications, meaning that channel components from one manufacturer to another cannot be interchanged. The entire channel must be purchased from a single manufacturer.
There is some support for legacy Category 6 cabling over certain distances to address some legacy installations, but it is important to note that a significant amount of mitigation will need to take place to eliminate alien crosstalk (cable-to-cable noise) in order for it to work properly. Also, with UTP systems, additional field testing will be required to determine whether the noise is cancelled. These problems do not exist with shielded systems, and for that reason many companies today are moving toward shielded systems.
I hope you had the opportunity at N+I to stop by the Ethernet Alliance booth and see this, as well as Power over Ethernet, 10GBASE-LRM, and other new technologies at work.
In an interview, Brad Booth, president of the Ethernet Alliance, shared his thoughts on the Alliance and the direction of technology. His responses follow:
What was the thinking behind the creation of the Ethernet Alliance?
The idea to create the Ethernet Alliance came at a time when the 10GEA was winding down. Just as the 10GEA closed up shop, a couple of other 10G Ethernet projects started up. They realized that forming an alliance for each standards project was just too costly and inefficient. This was the first time that the industry tried to create an alliance to support all IEEE 802 Ethernet projects, so the rules and concepts from the past had to be re-written.
How does this organization differ from others, such as the previous Gigabit
Past alliances were very focused on one (usually large) standards project. While this was effective for those projects, some of the smaller projects were overshadowed due to lack of a marketing alliance. Power over Ethernet is a good example. This market is just in its infancy, and Ethernet Alliance members are working diligently to explain and demonstrate this remarkable technology.
With the number of mergers and acquisitions, do you think we will continue to see new players
in Ethernet development? In what area?
Mergers and acquisitions happen all the time. There was a flurry of them in the Gigabit Ethernet days, and that tends to happen with each new technology introduction. One of the primary drivers is Ethernet's growth into new markets. Many companies use the M&A strategy to gain access to new technologies instead of developing a homegrown solution. As for the area, I expect that it will be a broad range.
Do you think last week's Interop event helped create more interest in 10GBASE-T technologies?
Who do you think will be the first adopters?
I believe that Interop brought a lot of interest in a number of Ethernet technologies, and 10GBASE-T was certainly one of them. The data center market has been watching and waiting for 10GBASE-T, as the volume of Ethernet deployed worldwide is UTP-based.
With work being done on 100G and potentially 40G, what do you think the impact will be on
applications in the future? What do you think tomorrow's desktop will look like, and what killer
applications will it be running?
The next speed of Ethernet is intended to alleviate the bottlenecks at those points in the network where multiple links of 10G are in use today. We now live in a society where Internet access is commonplace, and the bandwidth demand from homes is outpacing that from business. What this means is that no one application is truly the "killer application." Rather, the combined demand and the expected response times, whether it be to view a Web page or download a movie, are going to become more demanding. To meet these needs, companies are going to have to find ways to eliminate the bottlenecks.
About the author:
Carrie Higbie has been involved in the computing and networking industries for 25+ years. As the Global Network Applications Market at The Siemon Company, Carrie supports the end-user and electronics communities. She participates with the IEEE, TIA and various consortiums for standards acceptance. She has extensive background in all aspects of networking and application development as a consultant, project manager, and Fortune 500 executive and has taught at a collegiate level. She speaks at industry events and has published several articles and whitepapers globally. Carrie holds an MBA and MSBA. Carrie is an expert in TechTarget's Searchnetworking.com, SearchEnterpriseVoice.com and SearchDataCenter.com forums and is on the board of advisors. She writes a weekly column on a variety of topics. She is the President of the BladeSystems Alliance. Carrie has won the "Communication News" Editor's Choice Award for the last two years.
This was first published in May 2007