ATM Vs Fast Ethernet.

      By,
       M. V. Vinod and Preetham M. P, Final Year BE ( Comp Sci ), NIE, Mysore.

1. Introduction to Networks

In the past there have been three distinct types of networks:  voice, video, and data.  Voice has always been the responsibility of the telephone companies.  Their networks were designed for the maximization of voice traffic.  Like voice, video has always been the responsibility of television stations and cable companies.  Their networks consisted of analog broadcast over airwaves, as in the case of the local television station, or over the cable company's coax and fiber distribution system.  The television network was designed for half-duplex operation.  The third type of network, data networks, has had a relatively short history.  This history started out with the problem of remote access to mainframes, which was solved using IBM's SNA utilizing terminals, cluster controllers, and front ends.  All of these networks provided a specific type of application.  The optimization of these networks used proven analytical tools to determine network size and topology.  The assumptions that these analytical tools made are no longer true.

The lines that once separated the three network types are becoming less and less clear.  For example, cable companies are now starting to carry data via cable modems.  This new capability is costly because it requires the redesigning of the entire cable network to carry full duplex traffic.  And Telephone companies are also racing to provide data service as well as their usual voice service.  ISDN (Integrated Services Digital Network) is evidence of the telephone companies' attempts to offer higher bandwidth and more reliable data communication than possible with analog.  Digital video service is not yet in wide use, but with the popularity of the Internet, it looks to be popular in the near future.  Designing networks that offer all three type of traffic is a monumental challenge for network engineers.  The rules for designing networks in the past do not apply anymore with customers demanding more bandwidth, QoS (Quality of Service), and higher utilization while expecting the network to handle voice, video, and data.  There are many technologies for carrying this digital multi-media traffic, but we will only look at two of these in this paper, ATM (Asynchronous Transfer Mode) and Fast Ethernet (100BaseT).  Both of these are positioned to be significant players in the future of information networks but each is driving to out perform the other, and this  paper will look at that race and its consequences.

2. Ethernet

2.1. History

With the rush to provide some means to transfer data, Ethernet was quickly instituted as the standard data network.  This standard has evolved from several different levels with its newest level being 100BaseT Fast Ethernet.  Fast Ethernet has not changed the standard of Ethernet it has only made it faster.  100BaseT Fast Ethernet essentially took 10BaseT Ethernet and multiplied its bit rate by ten.  The hopes for this new, faster Ethernet are that it will revolutionize current Ethernet so it can continue to be the network topology of choice.  100BaseT brings to the desktop what is desired the most, more speed, but what are the costs of this improvement and is it enough to ward off potential ATM fanatics?

2.2. The Basics

High-speed networking is a hot topic in the LAN industry, which is why Fast Ethernet was developed.  100BaseT Fast Ethernet uses the same architecture that 10BaseT Ethernet uses and also uses the same collision detection scheme, CSMA/CD (Carrier Sense Multiple Access/Collision Detection.  This means it is just like Ethernet only faster.  Therefore, to understand Fast Ethernet we must first look at Ethernet. Ethernet is the most used network topology in industry today and it is for this reason Ethernet has the most integrated software and hardware there is in networking today. A typical Ethernet network might look like the following.

Figure 1. Example of Standard Ethernet.

This simple example shows how four computers might be connected, creating a network.  Computers 1 and 2 are connected via a hub, which simply takes information from one of two connected computers and broadcasts it within the connections to the hub (Computers 3 and 4 are also connected in this same manner).  In this example there are only two adjoining computers so the broadcast is really only sent to one computer but in much more complex networks several more computers would be connected to just one hub.  The hub simply acts as an intermediary between computers that communicate frequently.  What happens when computers are not connected to the same hub but need to communicate?  This is where the second intermediary, the router, plays its role.  In our example,  this is illustrated by computer 1 needing to communicate with computer 3 or computer 4 (the same illustration is made with computer 2 as well).  When computer one places its data out onto the network, the router recognizes that the destination is an adjoining network and therefore transfers the data. This simple example illustrates the basics of Ethernet and the basis under which Fast Ethernet evolved.

2.3. Evolution

The above example illustrated a simple network, but most networks today are not so simple.  There are sometimes hundreds of pieces of equipment.  In the large Ethernet networks of today, much of the system potential is not utilized.  A 10BaseT network might have a maximum bit rate of 10Mbs, but in actual usage, there are probably only 40 to 50 percent of its capability being utilized.  To increase the efficiency of Ethernet switches were added.  This type of network has come to be known as Switched Ethernet.

 

Figure 2. Example of Switched Ethernet

Switched Ethernet is a relatively new concept that simply takes the simple network design and adds the benefits of switching to it.  The simple network can be rightfully termed "dumb" because traffic is just basically broadcast to adjoining computers resulting in poor use of the network.  Switched Ethernet adds switches to the basic network so the utilization of each resource is increased.  A switch is basically a fancy hub where the frames are "switched" to the appropriate destination instead of broadcasting to everybody.  A switch actually knows where the information is going and where the destination is, which means a link between the sender and the receiver is set up by the switch.  This means a switch enhances the flow of network traffic and increases utilization.  It is important to point out here that Switched Ethernet does not run faster than typical Ethernet.  Both of these networks are bound to the line speed of 10Mbps but Switched Ethernet utilizes this line speed much better than Ethernet.  The 10Mbps boundary is an obstacle for Ethernet and Switched Ethernet, which is why a hurdle was made to 100BaseT Fast Ethernet.

Fast Ethernet is virtually identical to 10BaseT Ethernet.  The same protocols and methods of transport are used in both versions, but the line bandwidth is what separates them.  Fast Ethernet increased bandwidth remedies the bottlenecks found in standard 10BaseT due to a lack of bandwidth.  As with Switched Ethernet, switches can be added to Fast Ethernet to created Switched Fast Ethernet to allow better utilization of the network.

3. ATM (Asynchronous Transfer Mode)

3.1. History

ATM is a relatively new type of network, which is still in the development stages.  The ATM Forum, a body of industry experts, is still finalizing the standards that will bring much needed compatibility to ATM.  Currently, vendors are incorporating set standards into their ATM switches and adapters, but they rely on proprietary information for standards, which the Forum has not yet finalized.  These compatibility problems along with the uncertainty of purchasing non-standardized network equipment are currently significant hindrances in the acceptance of ATM.

3.2. The Basics

Unlike Standard 100BaseT Ethernet, ATM is a switched, connection-oriented networking technology that allows for dedicated, high-speed communications.  ATM is defined by a set of standards that offer the following advantages: These characteristics make ATM suitable for a wide range of environments and can be used to carry a wide variety of applications.

 

3.3. Evolution

Unlike Fast Ethernet, ATM has not had a long time to evolve over time.  A complete set of standards has yet to be finalized by the ATM Forum and the IEEE.

4. Fast Ethernet vs. ATM Architecture

4.1. Fast Ethernet

Figure 3. Ethernet Packet Description.

Fast Ethernet uses the same standards as Ethernet, which has a 26 byte header that specifies destination address and source address along with other overhead information.  With Ethernet the packet is broadcast onto the network via the hub and each computer looks as the packet header for the destination address.  If the destination address matches the address of one of computers connected to the hub then it will be picked up by that computer.  Once the destination computer picks up this packet, 46-1500 bytes will have been received which is the payload of an Ethernet Packet.

A Fast Ethernet network is limited in several respects.  The bandwidth of the network is at the mercy of every user on the network.  If one sender requires the use of all the bandwidth allocated on the network then when other users place information onto the network it is buffered until it can be sent.  This is a huge network management concern because if too much information is placed onto the network at once the network could crash.  What this basically says is, shared-media LANs, such as Fast Ethernet, have characteristics that cause the average bandwidth available to each user to decline with each additional user.  This is the main concern when discussing the use of Fast Ethernet for multimedia applications because these applications require such a large portion of the bandwidth on the network.  This type of architecture is what is plaguing Ethernet when it comes to using such applications as desktop videoconferencing.

 

Figure 4  Bandwidth vs. Number of Users.

4.2. ATM

 ATM traffic travels in cells not in packets.  These cells are 53 bytes, which was discussed above, with a header of only 5 bytes.  This means that the payload is only 48 bytes.  This cell format is illustrated below:

Figure 5. ATM Cell Description.

The payload is small but this is part of the design.  In Fast Ethernet the packet size is largely based on the information that is being sent.  For example, voice and video applications in Fast Ethernet create large packets that are then sent over the network.  These large packets can cause congestion in the network or in the buffers of the network.  ATM cells are fixed length, which means they are always 53 bytes wide and so the problems associated with the variable length Ethernet packets are avoided in ATM.  The model shown below describes ATM architecture:

 

.

Figure 6. ATM Layer Architecture

The ATM standard consists of three areas of functionality that are shown in the model:

The ATM adaptation layer is a crucial portion of the ATM architecture it consists of, The main role of AAL3-5 is to map the variable length information packets into the payloads of the fixed-length ATM cells.

Figure7 SVC Network Model.

The purpose of the ATM layer is to provide a common set of services to support higher layer protocol functions such as data, voice, or video applications.  These services include multiplexing multiple data streams and switching data over a network to the correct destination. Virtual paths are set up throughout the ATM network, which provide a guaranteed amount of bandwidth requested by the application.  If this request cannot be guaranteed by the network then the request is refused.  This type of virtual path is called a SVC (Switched Virtual Path).  It is set up only for the duration of each transmission session.  The following figure illustrates a network with SVC:

This function is apart of the traffic management scheme that ATM uses.  ATM traffic management has many service categories that have been defined by the ATM Forum.  The following tables give a description of these categories:
 
Constant Bit Rate (CBR) Variable Bit Rate, Real Time (VBR-RT)
- traffic pattern is deterministic  - traffic pattern is known and burst 
- cannot tolerate jitter or delay  - minimal cell delay variation 
- examples, voice traffic, circuit emulation, video  - example, compressed video 
 
Variable Bit Rate, Non-Real Time (VBR-NRT) Available Bit Rate (ABR)
Traffic pattern is known and burst  Burst traffic 
Can tolerate more delay than VBR-RT, but delay should still be bounded.  Dynamic 
Can tolerate any cell delay variation  Delay tolerant 
Example, frame relay  Example, TCP/IP LAN traffic 
 

Table 1  Traffic Management Service Categories  
Unspecified Bit Rate (UBR)
- No guaranteed quality of service 
- Relies on higher level protocols to provide flow control 
- Example, TCP/IP LAN traffic 
 

The picture below shows probable service categories for ATM applications.

 

Figure 8  ATM Applications.

The purpose of these traffic management schemes is to balance high quality service with maximum network utilization.

5. Multimedia Performance Comparison

5.1. Voice

The standard digital voice channel currently being used by the telephone networks using eight bit samples sampled at 4kHz which yields a bit rate of 64kbps.  When a call is placed, a fixed circuit is set up which guarantees a 64kbps path through the network for the duration of the call.  The customer pays for this allocated bandwidth.  In order to have uninterrupted speech that is comprehensible the network must be able to guarantee that at least most of the voice packets/cells make it to the destination in real time.  Telephone customers will not tolerate significant delays in signal propagation.

Fast Ethernet has made huge progress over 10BaseT Ethernet in delivering these voice packets to the destination quickly.  This improvement is due primarily to the increased bandwidth of the lines.  For a customer who just upgraded from 10BaseT this would provide good voice channels because the network utilization would be quite low.  As the traffic on the network increases, however, voice communication would become degraded because packets would begin to be stored in buffers until the network allows for their transmission.  Because of the continuous bit rate associated with voice communication, the Fast Ethernet packet would be large in comparison with an ATM cell.  In the case of lost or damaged packets/cells, a loss for delayed Fast Ethernet will have a significant impact on voice communication performance compared to a small ATM cell.  For this reason, Fast Ethernet is not a good solution for good communication on a congested network.

The structure of ATM lends itself well to the implementation of voice traffic over the network.  ATM will set up a SVC (Switched Virtual Channel), using the AAL1 protocol, throughout the network at the initiation of a session and eliminate this SVC at the end of the session.

The SVC specifies a path and allocated bandwidth through the network.  This means that a voice session will have all of its packets delivered through the same path and in a timely fashion.  If the network is too congested to guarantee the above criteria then the request will be denied.  The SVC is very similar to the physical switching done by the telephone network and is therefore very adept to carry good quality voice traffic.

5.2. Video

Video traffic is very similar to voice traffic because a constant stream is needed to provide good quality pictures from source to destination.  This, again, is a problem for Fast Ethernet because the link from sender to receiver is not guaranteed.  If video is being sent through the network and there is not enough bandwidth allocated to this traffic then portions of the video stream will be buffered.  This virtually causes portions of the picture to be lost because video needs a constant flow in order to maintain quality.  When portions are buffered they will arrive at the destination too late and the picture will appear skewed or delayed.  Fast Ethernet is able to provide video only by good network management.  In other words, parts of the network that have video traveling back and forth are connected together so the video stream will not have to flow over the backbone where the most congestion occurs and bandwidth is usually limited.

As in voice, the structure of ATM lends itself well to the implementation of video traffic over the network.  The SVC that is set up by the network will assure a constant stream that is necessary for good quality.  The SVC system does not mean that there is no buffering in the ATM network.  There is buffering but it is not on the network, like in Fast Ethernet, but is at the card level where network performance is not sacrificed.  This does not mean network management is not an issue with ATM because good network management always connects computers together (in the same domain) that are frequent communicators, but it is not such a crucial element to allow for quality video as it is in Fast Ethernet.

5.3. Data

Data traffic is well suited for use on Fast Ethernet.  This type of traffic is bursty in nature and not sensitive to delays in transmission.  On a fully utilized 100BaseT network, data traffic can be buffered during transmission without the destination node experiencing unacceptable delays.  Since data traffic does not require the constant flow of packets, as does video and voice, this buffering does not sacrifice quality.  This means the current topology of choice, Ethernet, is perfectly acceptable for data networks.

ATM networks still set up a SVC for data traffic but this is not a necessity for data traffic since real time transmission is not crucial.

6. Why Not ATM?

ATM networks have proven themselves to be more efficient, faster, and more reliable when it comes to video and voice applications.  So why isn't ATM becoming the network of choice?  The biggest reason is cost.  The network of choice for many, many years has been Ethernet, which means most of the current network resources are for use with Ethernet.  100BaseT technology offers unparalleled ease of migration.  You can decide how fast to upgrade, in what steps and stages, without massive "fork-lift" upgrades.  Most 100BaseT network interface cards (NICs) will run at 10BaseT and 100BaseT.  This means that a slow move to Fast Ethernet, from Ethernet, can be made; companies can first purchase the needed NICs and run them at 10BaseT until they can afford to purchase the other needed equipment to upgrade to 100BaseT.  This ease of migration is not in place for the move to ATM networks.

ATM equipment is not too much more expensive, except for the adapters, but the cost in implementing this equipment is very high.  The current network basically has to be thrown out to make room for an entire ATM network.  This means the current Ethernet equipment would be rendered obsolete and placed out of service.  This is a pricey task especially for those who have enormous Ethernet networks already setup.  It is not entirely true that there is no ease of migration for ATM.  In view of the hardship involved in changing to an ATM network most developers of ATM equipment have come out with Ethernet switches that will allow an Ethernet network to be changed into an ATM network;  an Ethernet- to-ATM switch.

This switch would allow companies to slowly change to ATM, but we have already discussed the standards issues that plague this market right now.  Which company do you buy your ATM equipment through because if you buy from one, you might not be able to integrate that equipment with equipment from another company.  It is a very risky move, and there are too many considerations to be taken into account for such a move which is why ATM will not become the LAN of choice for some time to come.  The chart below illustrates the cost differences in the adapters for Fast Ethernet and ATM as well as a few others.
 

100BaseT ATM
Speed 100 Mbps 155 Mbps
Cost <  $295 ~  $1,300
 
                                            Table 2  Adapter Cost Comparison

7. Where is ATM Now?

Even though Ethernet continues to be the network of choice there is a big move to ATM.  This move is shown in the backbone of networks or in Wide Area Network Applications (WANs).  Many companies are making the change to ATM on their backbone to increase the performance of their Ethernet networks.  There is a great deal of congestion at the backbone, which is a great limitation for Fast Ethernet because this slows down the network, which was mentioned above.  With an ATM backbone, information can be pushed through much faster which helps eliminate some of the congestion.  This ease of congestion can lead to better performance from Fast Ethernet networks when using multimedia applications such as video and voice.  With WANs, ATM is very useful simply because of its speed and utilization.  The faster the network traffic can arrive the better the network and ATM provides the speed needed for these networks.

8. Conclusion

Telecommunications networks have come a long way in the last couple of decades.  What used to be decisively different networks are now coming together in a reformation of telecommunications topologies and architectures.  This change is being driven by the ultimate goal of integrating voice, video, and data all into one giant multimedia network.  Fast Ethernet and ATM are current technologies that will take us further on this quest for integration.  Fast Ethernet is currently a bargain as far as price for bandwidth, but in the future ATM has the tools to make it the obvious choice for multimedia.  ATM's price will eventually come down and should be able to compete with Fast Ethernet in the LAN market.  As time moves on, Fast Ethernet and ATM will probably make way for a new technology that will take us into the distant future of telecommunication networks.

9. References

10. Acknowledgement

The authors of this paper thank Dr. M. S. Shiva kumar, Head, Department of Computer Science & Engineering of The National Institute of Engineering, Mysore for his valuable guidance in writing this paper.

Top

Back To Home Page