The Atm Interfaces And Devices

Published Date: 02 Nov 2017

Sushant Kumar

Abstract — ATM networks were designed to support a range of services such as telephone and internet under one at reasonable cost. This technology is similar to frame relay and X.25 as it also breaks the data into small chunks or packet to transfer .These packets are of fixed size called cells.ATM is a connection-oriented technology in which end-to-end connection is established between users .It also support multiplexing (multiple connection) over a single physical interface with high level of control and quality of services.

Introduction

ATM scheme was developed in 1991by ITU (International Telecommunication union) for packet switching networks which is designed to offer voice and data signals services works on bottom two layers of OSI reference model. Transmits data in a fixed size cells, these cells are transmitted asynchronously. Each cell is 53 bytes with 48 bytes of data and 5 bytes of header.ATM is a connection-oriented service means a connection is first established between two end users and subsequently each cell follow the same path to destination.

ATM is independent of transmission medium it can transfer data across optical fiber, coaxial cable, twisted pair etc. Much of the ATM devices operated at 155Mbps (OC3) and 622Mbps (OC12) speeds.

ATM Network architecture and devices

ATM Networks conspires of two kind devices basically-

a)ATM Switches- These devices accepts the cell either from one ATM switch or end devices and forward it to other ATM switches.

b)ATM end devices- These devices may include workstation or router that are directly connected to ATM switches .

ATM uses different type of interfaces that connect ATM devices over point to point link-

a)UNI- Is user network interface connects ATM end devices with an ATM switch . UNI is also subdivided into public and private UNI

Figure .1 ATM interfaces and devices

b)NNI –Stands for network node interface it connects two ATM switches both ends are in network field. It is also categories as public and private.

c)B-ICI – Is Broadband inter-carrier a third type of interface in ATM networks connects two switches from two different networks.

ATM layer by layer approach

Figure 1.2- Simplistic view of ATM layers

ATM Adaptation Layer – This layer is responsible for communication with higher layers of system. It is further subdivided into two sublayers.

a)CS –convergence sublayer provide different services to different applications which may include CBR(constant bit rate) , ABR(available bit rate) etc.

b)SAR-Segmentation and reassembly sublayer- It tears down the message from upper layer into cells of 48 byte each and reassembles it at destination end.

ATM layer - This layer images to network layer of OSI model and also show some characteristics data link layer . It control many functions such as transmission, congestion control, call header, sequential delivery.

Physical layer – Transport ATM cells in physical medium also control bit rate.

ATM Cell structure

ATM uses fixed size cells to transmit source information to destination. These small sized fixed cells give many advantages to this system like it reduces queuing delay to high-priority cells and switching can be implemented more efficiently producing high data rates for ATM this is one of the reasons it is used for video and data information.

ATM Cell is 53 byte (octets) in length. Payload or user data is of 48 bytes (octets) and header is of 5 bytes.

(a)

(b)

Figure 1.3- (a) Cell format of ATM (b) Expanded version of cell format

ATM HEADER – 5 bytes of header field is subdivided into several fields.

(a)Generic Flow control (GFC)- These 4bits are used for local flow control only between the network interface and more then one end users .This provides the multiple access to users but it reduces the transmission rate of the node(usually ignored)

(b)Virtual path identifier (VPI)-It is 8 bit for user-network interface and as GFC fireld is not of use in network-network interface so it is of 12 bits . Its identifies the next destination of the cell in ATM network .

(c)Virtual channel identifier(VCI)-It works along with VPI, to find the next destination of the cell. It defines the virtual channel for the specified virtual path .

(d)Payload type identifier (PTI)- It is a 3-bit field identifier

If first set to 0 shows user data and bit set to 1 indicates operation data.

Second bit shows congestion states and third bit is to find out last cell of the frame(0 not last cell /1 last cell of the frame)

(e)Cell loss priority(CLP)-This field is a 1 bit field used as priority indicator if it is set to ‘0’ means cell have high priority and only should be discarded when there is no other alternative. If it is set to ‘1’ it means cell has low priority and can be discarded .

(f)Header error control(HEC)-It is a 8-Bit field ,it can detect can correct all 1-bit error using cyclic redundancy check(CRC), but only can detect multi-bit errors.

Logical connections in ATM

In the connection-oriented approach of ATM a logical path is created between two end users for communication. There are two elements of this approach -

(a).Virtual channel connections –This is a connection channel between setup between two nodes and is identify by VCI field. Full-duplex in nature.

(b)Virtual path connections – It consist of large number or bundles of virtual channel with same end points. It provide efficient use of channel.

Figure 1.4- Virtual paths and virtual channel

Controlling signals- These are used to establish initial connection between two user , as in case of Circuit switching .

ATM architecture protocols

Protocol reference model of ATM has three planes, each plane use same reference model i.e. use same layers as used by other planes.

Figure – 1.5- "Protocol architecture"

(Referred from –William stalling –Data and computer communication –Ed.8 –page number 309, chapter 11)

Three separate planes in ATM model-

(a)User Plane – This plane is responsible for managing Data transfer.

(b)Control Plane- This plane manages the call control to setup the virtual circuit and control the connection.

(c)Management Plane- Is responsible for the coordination between all planes and all layers.

Conclusion

I would like to conclude that when is included ATM technology gave the networking world a new era, providing high speed rate and multiplexing, for voice video and data signals. Concept of virtual path and channel provides the high efficiency of single channel i.e. bandwidth utilization.

References -

1)http://docwiki.cisco.com/wiki/Asynchronous_Transfer_Mode_Switching

2) http://www.cellsoft.de/telecom/atmconcepts.htm

3) http://technet.microsoft.com/en-us/library/cc976978.aspx

4)http://www.cisco.com/en/US/products/hw/switches/ps1893/prod_troubleshooting_guide_chapter09186a00800f4fb9.html

5) William Stalling -Data and computer communication Ed.8 – chapter 11

References for Figures

Figure1.3 http://www.cisco.com/en/US/products/hw/switches/ps1893/prod_troubleshooting_guide_chapter09186a00800f4fb9.html

Figure1.4

http://docwiki.cisco.com/wiki/Asynchronous_Transfer_Mode_Switching

FIBER CHANNEL

Abstract- Fiber channel technology is standardized by T11- technical committee which can support speed from 2-16 Gigabits. It was designed for supercomputer at first, but with its advantages it is now in use for storage area networks (SAN). It support wide variety of topologies and protocols, is very highly efficient because for less overheads and no congestions is offered in this technology.

Introduction

Fiber channel fuse the characteristics of two technologies. One is the speed of I/O channels. I/O channels (manages transfer between processers and peripheral devices) is used to send high speed devices over short distance and other is network communication which allow interconnectivity and flexibility. Fiber channel uses fiber optics media for long distance communication (10 KM), and support coaxial and twisted pair over short distance. Fiber also support several protocols likesuch as SCSI(Small Computer System Interface) and IP (Internet Protocol)

Topologies and Physical media

In fiber channel there are two basic elements .Node which are the high speed end devices, other one is fabric these are the switching elements in the network. At Node there are N_ports and multiple F_ports on fabric.

Figure 2.1 – "Fiber network and elements"

(Referred from –William stalling –Data and computer communication –Ed.8 –page number 479, chapter 16)

In Figure 2.1 working cluster, mainframes are end devices and these are referred to as Nodes. Fiber channel switching includes fabric elements.

Physical Media- Fiber Channel support wide range of media depending upon distance. Optical fiber is used for long distance with 3.2 Gbps from 175m to 10km. Coaxial and twisted pair uses over short distance ranging from 20m to 80m.

Protocol Architecture

Fiber channel architecture follow a hierarchical approach or layered approach with 5 basic layers.

Figure 2.2 – layers in Fiber channel

FC-0 Physical layer- is bottom layer of fiber channel deals with transmitter, receiver ,transmission rate and electrical parameter of signals.

FC-1 layer – It deals with transmission protocol which include coding and decoding (8B/10B). 8B/10B means 8-Bit internal byte is encoding into 10-bit transmission character and vice versa on receiver side. 10-Bits give 1024 different combination as compare to 256 for 8-Bit but to maintain dc level only several combinations are used , one named comma is for Byte synchronization.

FC-2 layer-This layer is multipurpose. This layer defines both framing and signaling protocol. Breaking of sequence data into frames which is assembly and disassembly. Also check the flow control.

Several in build features in this layers -

Ordered set- It is a 4 Byte Word doing management, signaling between end devices and error recovery. Three type of ordered set is defined by

(a)Frame Delimiter- To detect startup and end bit of a frame .

(b)Primitive signal-Are of two type idle and receive ready (R_RDY) used during frame transmission.

(c)Primitive sequence - To determine right order of frames.

FC-3 Layer-A layer that provides common services to Node ports (N_port) some of these services are Stripping to use signal information in multiple link .Hunt group to enable different ports to respond to same address and multicast to deliver single message to multi-destination.

FC-4 layer- Is a mapping layer which maps the higher level protocols to fiber channel layers, and support protocols such as SCSI,IPI-3 .

Addressing in Fiber channel

A fiber channel address is of 4 byte and is assigned dynamically. Frame includes source and destination address. These address ranges from hex’FFFFF0’ hex’FFFFFE’ .Before any transmission address at N_port is undefined and set as default hex ‘000000’. All F’s is a broadcast address (FFFFFF).

CONCLUTION

Fiber channel is one of the fastest interfaces we have in today world

supports speed up to 20 -250 time any other interface. Because

of the high speed feature of I/O interfaces channel and flexibility and interconnection of communication channel