10-Gigabit Ethernet




10-Gigabit Ethernet

As soon as gigabit Ethernet was standardized, the 802 committee got bored and wanted to get back to work. IEEE told them to start on 10-gigabit Ethernet. This work followed much the same pattern as the previous Ethernet standards, with standards for fiber and shielded copper cable appearing first in 2002 and 2004, followed by the standard for copper twisted pair in 2006.

10 Gbps is a truly prodigious speed, 1000x faster than the original Ethernet. Where could it be needed? The answer is inside data centers and exchanges to connect high-end routers, switches, and servers, as well as in long-distance, high bandwidth trunks between offices that are enabling entire metropolitan area networks based on Ethernet and fiber. The long distance connections use optical fiber, while the short connections may use copper or fiber.

All versions of 10-gigabit Ethernet support only full-duplex operation. CSMA/CD is no longer part of the design, and the standards concentrate on the details of physical layers that can run at very high speed. Compatibility still matters, though, so 10-gigabit Ethernet interfaces autonegotiate and fall back to the highest speed supported by both ends of the line.

The main kinds of 10-gigabit Ethernet are listed in Fig. 4-22. Multimode fiber with the 0.85μ (short) wavelength is used for medium distances, and singlemode fiber at 1.3μ (long) and 1.5μ (extended) is used for long distances. 10GBase-ER can run for distances of 40 km, making it suitable for wide area applications. All of these versions send a serial stream of information that is produced by scrambling the data bits, then encoding them with a 64B/66B code. This encoding has less overhead than an 8B/10B code.

10-Gigabit Ethernet

The first copper version defined, 10GBase-CX4, uses a cable with four pairs of twinaxial copper wiring. Each pair uses 8B/10B coding and runs at 3.125 Gsymbols/second to reach 10 Gbps. This version is cheaper than fiber and was early to market, but it remains to be seen whether it will be beat out in the long run by 10-gigabit Ethernet over more garden variety twisted pair wiring.

10GBase-T is the version that uses UTP cables. While it calls for Category 6a wiring, for shorter runs, it can use lower categories (including Category 5) to allow some reuse of installed cabling. Not surprisingly, the physical layer is quite involved to reach 10 Gbps over twisted pair. We will only sketch some of the high-level details. Each of the four twisted pairs is used to send 2500 Mbps in both directions. This speed is reached using a signaling rate of 800 Msymbols/sec with symbols that use 16 voltage levels. The symbols are produced by scrambling the data, protecting it with a LDPC (Low Density Parity Check) code, and further coding for error correction.

10-gigabit Ethernet is still shaking out in the market, but the 802.3 committee has already moved on. At the end of 2007, IEEE created a group to standardize Ethernet operating at 40 Gbps and 100 Gbps. This upgrade will let Ethernet compete in very high-performance settings, including long-distance connections in backbone networks and short connections over the equipment backplanes. The standard is not yet complete, but proprietary products are already available.

Retrospective on Ethernet

Ethernet has been around for over 30 years and has no serious competitors in sight, so it is likely to be around for many years to come. Few CPU architectures, operating systems, or programming languages have been king of the mountain for three decades going on strong. Clearly, Ethernet did something right. What?

Probably the main reason for its longevity is that Ethernet is simple and flexible. In practice, simple translates into reliable, cheap, and easy to maintain. Once the hub and switch architecture was adopted, failures became extremely rare. People hesitate to replace something that works perfectly all the time, especially when they know that an awful lot of things in the computer industry work very poorly, so that many so-called ‘‘upgrades’’ are worse than what they replaced.

Simple also translates into cheap. Twisted-pair wiring is relatively inexpensive as are the hardware components. They may start out expensive when there is a transition, for example, new gigabit Ethernet NICs or switches, but they are merely additions to a well established network (not a replacement of it) and the prices fall quickly as the sales volume picks up.

Ethernet is easy to maintain. There is no software to install (other than the drivers) and not much in the way of configuration tables to manage (and get wrong). Also, adding new hosts is as simple as just plugging them in.

Another point is that Ethernet interworks easily with TCP/IP, which has become dominant. IP is a connectionless protocol, so it fits perfectly with Ethernet, which is also connectionless. IP fits much less well with connection-oriented alternatives such as ATM. This mismatch definitely hurt ATM’s chances.

Lastly, and perhaps most importantly, Ethernet has been able to evolve in certain crucial ways. Speeds have gone up by several orders of magnitude and hubs and switches have been introduced, but these changes have not required changing the software and have often allowed the existing cabling to be reused for a time. When a network salesman shows up at a large installation and says ‘‘I have this fantastic new network for you. All you have to do is throw out all your hardware and rewrite all your software,’’ he has a problem.

Many alternative technologies that you have probably not even heard of were faster than Ethernet when they were introduced. As well as ATM, this list includes FDDI (Fiber Distributed Data Interface) and Fibre Channel,† two ringbased optical LANs. Both were incompatible with Ethernet. Neither one made it. They were too complicated, which led to complex chips and high prices. The lesson that should have been learned here was KISS (Keep It Simple, Stupid). Eventually, Ethernet caught up with them in terms of speed, often by borrowing some of their technology, for example, the 4B/5B coding from FDDI and the 8B/10B coding from Fibre Channel. Then they had no advantages left and quietly died off or fell into specialized roles.

It looks like Ethernet will continue to expand in its applications for some time. 10-gigabit Ethernet has freed it from the distance constraints of CSMA/CD. Much effort is being put into carrier-grade Ethernet to let network providers offer Ethernet-based services to their customers for metropolitan and wide area networks (Fouli and Maler, 2009). This application carries Ethernet frames long distances over fiber and calls for better management features to help operators offer reliable, high-quality services. Very high speed networks are also finding uses in backplanes connecting components in large routers or servers. Both of these uses are in addition to that of sending frames between computers in offices.



Frequently Asked Questions

+
Ans: At the same time that switches were becoming popular, the speed of 10-Mbps Ethernet was coming under pressure. At first, 10 Mbps seemed like heaven, just as cable modems seemed like heaven to the users of telephone modems. But the novelty wore off quickly. view more..
+
Ans: In any broadcast network, the key issue is how to determine who gets to use the channel when there is competition for it. To make this point, consider a conference call in which six people, on six different telephones, are all connected so that each one can hear and talk to all the others. It is very likely that when one of them stops speaking, two or more will start talking at once, leading to chaos. view more..
+
Ans: We have now finished our discussion of channel allocation protocols in the abstract, so it is time to see how these principles apply to real systems. Many of the designs for personal, local, and metropolitan area networks have been standardized under the name of IEEE 802. A few have survived but many have not, as we saw in Fig. 1-38. Some people who believe in reincarnation think that Charles Darwin came back as a member of the IEEE Standards Association to weed out the unfit. view more..
+
Ans: 10 Gbps is a truly prodigious speed, 1000x faster than the original Ethernet. Where could it be needed? The answer is inside data centers and exchanges to connect high-end routers, switches, and servers, as well as in long-distance, high bandwidth trunks between offices that are enabling entire metropolitan area networks based on Ethernet and fiber. view more..
+
Ans: Wireless LANs are increasingly popular, and homes, offices, cafes, libraries, airports, zoos, and other public places are being outfitted with them to connect computers, PDAs, and smart phones to the Internet. view more..
+
Ans: The 802.11 standard defines three different classes of frames in the air: data, control, and management. Each of these has a header with a variety of fields used within the MAC sublayer. view more..
+
Ans: The main wireless LAN standard is 802.11. We gave some background information on it in Sec. 1.5.3. Now it is time to take a closer look at the technology. view more..
+
Ans: At this point you may be thinking: why devise a new standard? Why not just use 802.11 or 3G? In fact, WiMAX combines aspects of both 802.11 and 3G, making it more like a 4G technology. view more..
+
Ans: All MAC frames begin with a generic header. The header is followed by an optional payload and an optional checksum (CRC), as illustrated in Fig. 4-33. The payload is not needed in control frames, for example, those requesting channel slots. view more..
+
Ans: In 1994, the L. M. Ericsson company became interested in connecting its mobile phones to other devices (e.g., laptops) without cables. Together with four other companies (IBM, Intel, Nokia, and Toshiba), it formed a SIG (Special Interest Group, i.e., consortium) in 1998 to develop a wireless standard for interconnecting computing and communication devices and accessories using short-range, low-power, inexpensive wireless radios. view more..
+
Ans: Bluetooth defines several frame formats, the most important of which is shown in two forms in Fig. 4-36. It begins with an access code that usually identifies the master so that slaves within radio range of two masters can tell which traffic is for them. view more..
+
Ans: Many organizations have multiple LANs and wish to connect them. Would it not be convenient if we could just join the LANs together to make a larger LAN? In fact, we can do this when the connections are made with devices called bridges. view more..
+
Ans: To increase reliability, redundant links can be used between bridges. In the example of Fig. 4-43, there are two links in parallel between a pair of bridges. This design ensures that if one link is cut, the network will not be partitioned into two sets of computers that cannot talk to each other view more..
+
Ans: In the early days of local area networking, thick yellow cables snaked through the cable ducts of many office buildings. Every computer they passed was plugged in. No thought was given to which computer belonged on which LAN. view more..
+
Ans: The network layer is concerned with getting packets from the source all the way to the destination. Getting to the destination may require making many hops at intermediate routers along the way. view more..
+
Ans: The main function of the network layer is routing packets from the source machine to the destination machine. In most networks, packets will require multiple hops to make the journey. The only notable exception is for broadcast networks, but even here routing is an issue if the source and destination are not on the same network segment. view more..
+
Ans: When a routing algorithm is implemented, each router must make decisions based on local knowledge, not the complete picture of the network. A simple local technique is flooding, in which every incoming packet is sent out on every outgoing line except the one it arrived on. view more..
+
Ans: Distance vector routing was used in the ARPANET until 1979, when it was replaced by link state routing. The primary problem that caused its demise was that the algorithm often took too long to converge after the network topology changed (due to the count-to-infinity problem). view more..




Rating - NAN/5
454 views

Advertisements