ATM - asynchronous transfer mode tutorial

- an overview or tutorial of the basics of asynchronous transfer mode, used in many data networking applications large and small.

The Asynchronous Transfer Mode (ATM) was developed to enable a single data networking standard to be used for both synchronous channel networking and packet-based networking. Asynchrnonous transfer mode also supports multiple levels of quality of service for packet traffic.

In this way, asynchronous transfer mode can be thought of as supporting both circuit-switched networks and packet-switched networks by mapping both bitstreams and packet-streams. It achieves this by sending data in a series or stream of fixed length cells, each of which has its own identifier. These data cells are typically sent on demand within a synchronous time-slot pattern in a synchronous bit-stream. Although this may not appear to be asynchronous, the asynchronous element of the "Asynchronous Transfer Mode", comes from the fact that the sending of the cells themselves is asynchronous and not from the synchronous low-level bitstream that carries them.

One of the original aims of Asynchronous Transfer Mode was that it should provide a basis for Broadband Integrated Services Digital Network (B-ISDN) to replace existing PSTN (Private … ). As a result of this the standards for Asynchronous Transfer Mode standards include not only the definitions for the Physical transmission techniques (Layer 1), but also layers 2 and 3.

In addition to this, the development of Asysnchronous Transfer Mode was focussed heavily on the requirements for telecommunications providers rather than local data networking requirements, and as a result it is more suited to large area telecommunications applications rather than smaller local area data network solutions, or general computer networking.

While Asynchronous Transfer Mode is widely used for many applications, it is generally only used for transport of IP traffic. It has not become the single standard for providing a single integrated technology for LANs, public networks, and user services.

Basic asynchronous transfer mode system
There are two basic elements to an ATM system. Any system can be made up a number of each of these elements:

• ATM switch     This accepts the incoming cells or information "packets" from another ATM entity which may be either another switch or an end point. It reads and updates the cell header information and switches the information cell towards its destination



• ATM end point     This element contains the ATM network interface adaptor to enable data entering or leaving the ATM network to interface to the external world. Examples of these end points include workstations, LAN switches, video codecs and many more items.

ATM networks can be configured in many ways. The overall network will comprise a set of ATM switches interconnected by point-to-point ATM links or interfaces. Within the network there are two types of interface and these are both supported by the switches. The first is UNI and this is used to connect ATM end systems (such as hosts and routers) to an ATM switch. The second type of interface is known as NNI. This connects two ATM switches.

ATM operation
In ATM the information is formatted into fixed length cells consisting of 48 bytes (each 8 bits long) of payload data. In addition to this there is a cell header that consists of 5 bytes, giving a total cell length of 53 bytes. This format has been chosen because time critical data such as voice packets is not affected by very long packets being sent. The data carried in the header comprises payload information as well as what are termed virtual-circuit identifiers and header error check data.

ATM is what is termed connection orientated. This has the advantage that the user can define the requirements that are needed to support the calls, and in turn this allows the network to allocated the required resources. By adopting this approach, several calls can be multiplexed efficiently and ensuring that the required resources can be allocated.

There are two types of connection that are specified for asynchronous transfer mode:

• Virtual Channel Connections - this is the basic connection unit or entity. It carries a single stream of data cells from the originator to the end user.



• Virtual Path Connections - this is formed from a collection of virtual channel connections. A virtual path is an end to end connection created across an ATM (asynchronous transfer mode) network. For a virtual path connection, the network routes all cells from the virtual path across the network in the same way without regard for the individual virtual circuit connection. This results in faster transfer.

  The idea of virtual path connections are also used within the ATM network itself to route traffic between switches

ATM networks can be configured in many ways. The overall network will comprise a set of ATM switches interconnected by point-to-point ATM links or interfaces. Within the network there are two types of interface and these are both supported by the switches. The first is UNI and this is used to connect ATM end systems (such as hosts and routers) to an ATM switch. The second type of interface is known as NNI. This connects two ATM switches.