Frame Relay versus Asynchronous Transfer Mode: A Comparative Study and Simulation
Автор: Ibrahim Ali Ibrahim Diyeb, Sharaf A. Alhomdy
Журнал: International Journal of Computer Network and Information Security(IJCNIS) @ijcnis
Статья в выпуске: 10 vol.9, 2017 года.
Бесплатный доступ
Frame Relay and Asynchronous Transfer Mode (ATM) are two famous technologies in wide-area networks that use the virtual circuits for connecting and multiplexing many branches and networks. Those technologies can be used to carry different types of information such as real-time information which needs high data rate, low delay and an efficient throughput of the network's data communication devices. This paper is a study based on simulation for comparative performance evaluation between the Frame Relay and ATM to understand the overview and the nature of these technologies focusing on real-time applications using the OPNET simulation tool for analyzing the performance of voice application. The evaluation parameters are traffic sent, traffic received, delay, Jitter and the end-to-end delay which have been used to compare the performance of the ATM and Frame Relay in high-speed networks. The simulation results demonstrate that the ATM has high traffic sent, high traffic received, and less delay compared to the Frame Relay. In addition, this paper may be considered as an insight for the new researchers to guide them to an overview, essentials, and understanding of the virtual connections for Frame Relay and ATM.
Frame Relay, Asynchronous Transfer Mode, Traffic, OPNET, Voice, Real-time Applications
Короткий адрес: https://sciup.org/15011934
IDR: 15011934
Текст научной статьи Frame Relay versus Asynchronous Transfer Mode: A Comparative Study and Simulation
Published Online October 2017 in MECS DOI: 10.5815/ijcnis.2017.10.04
There are many techniques and protocols of the wide area communication for transferring data between remotely networks and devices. But, to carry out the realtime data with high accuracy and with minimum delay in high-speed network needs high-speed technologies. The Frame Relay and ATM are two well-known technologies that are used in these days.
They are considered as the variations of the legacy X.25 protocol to eliminate the data transmitted overhead and to provide best controlling of signals by carrying the signals on different channels and paths using the virtual connections[1].
Both Frame Relay and ATM are connection oriented protocols, support switching and multiplexing of multiple logical connections on the single physical link and they provide dynamic bandwidth allocation for efficient traffic management, but there is no processing of an error correction .
In addition, Frame Relay has the characteristics of packet switching network, variable packet size, more cheap cost than the ATM and other network's types like leased lines. It appeared in the early 1990s as the successful technique for X.25 by overcoming physical errors (e.g. noise lines) that were found in X.25. Hence it is taken as better controlled congestion, high performance, throughput and high speed compared to X.25 for the lowest two layers of OSI model (Link-layer and the physical layer)[1][2].
Whereas, ATM is known as cell relay, it was designed in the 1980s, and fit to standardize in ITU-T at 1987. ATM combines the circuit switching and packet switching of small size. It is designed for low latency in real-time applications like video and voice and high throughput like file transfer. But in this method there is minimum flow control and error control capabilities. The ATM’s virtual circuits are established for both the source and the destination devices [1][2].The data is transferred as fixed cell of 53 bytes long of both header (5 bytes) and data payload (48 bytes) [3].
Moreover, in the ATM a virtual channel is a logical connection similar to virtual circuit. The virtual channels which have the same end points can be grouped into virtual paths. Hence, all the circuits in virtual paths are switched together. This offers increased efficiency, architectural simplicity, and the ability to offer enhanced network services.
In general, the real-time applications such as voice over IP and video on demands need high data rate and throughput for data transmission from source to destination. The Frame Relay has less data rate than the ATM and it provides low quality for communication of those applications. Hence, this paper provides and analyses the comparative study between the Frame Relay and ATM to understand the overview and the nature of these technologies in Wide Area Network (WAN) for real-time applications. The analysis of voice application was done using the OPNET simulation tool based on the various parameters(traffic sent, traffic received, delay, Jitter and the end-to-end delay) to compare the performance of the ATM and Frame Relay in high-speed networks.
-
II. Frame Relay
Frame Relay appeared in early 1990s. it is the backbone of a communication network for more companies and it is also provided by Internet Service Provider (ISP) for the remote intranet branches on different remote cities.
Moreover, it is an enhanced network of X.25 for reducing the End-to-end delay, equipment cost, and overhead of X.25. At the beginning, Frame Relay is developed to work on ISDN and now it is expanded to other public and private non-ISDN network interfaces. It is also considered as the backbone to provide services to upper layer protocols that have the network layer like IP. Furthermore, it has typical data rate from 64Kbps reach to 45Mbps (or T3).There are three types of frames show in table 1.
Table 1. The Frames Type
SN |
Type |
Description |
1 |
Information Frames |
Carry data as well as Next Send (NS) and Next Receive (NR) counts |
2 |
Supervisory Frames |
Controls flow of data with Receiver Ready (RR), Receiver Not Ready (RNR), and Reject (REJ) frames |
3 |
Unnumbered Frames |
Establish and maintain communications with Set Asynchronous Balanced Mode (SABM), Unnumbered Acknowledgment (UA), Disconnect (DISC), Disconnect Mode (DM) and Frame Reject (FRMR) |
The data transfer in Frame Relay, which works typically on the data link layer, occurs in packets that are named frames. The virtual connections are recognized by the unique identifier that is called Data link connection identifier (DLCI). The DLCI connections in Frame Relay devices are called 'ports'. All Data Terminal Equipment’s (DTEs) devices that connected to frame relay network need a port that is unique for every remote device. However, there are 10 bits for DLCI's field value in the Frame Relay header that provides 1,024 different virtual connections or DLCIs.
Furthermore, the data terminal equipment (DTE) that connected to Frame Relay network needs a port that is unique for every remote device. There are two types of connections in Frame Relay such as the Switched Virtual Circuit (SVC) and the Permanent Virtual Circuit (PVC) which are defined by ANSI and ITU-T. The PVC has two stages of operational states as data transfer and idle. Whereas, the SVC has four operational states as call setup, data transfer, idle, and call-termination. The DTE’s devices establish the virtual connections via call setup and then they transfer the data between DTE’s devices. The idle stage means that the connection is active, but there is no data transferring. The call-termination, as the last stage, is tearing the virtual connection off [1][2][3][4].
There are two types of establishing connection between devices in frame relay which are the point to point and point to multipoint sub-interfaces connection. It is considered as the Non-Broadcast Multi-Access (NBMA) network that does not provide broadcast between its virtual channels by default and it provides multi access across PVC or SVC channels [4].
However, there are some limitations such as the function of Frame Relay is just to detect errors at the data link layer and there is neither error control nor hop to hop flow control of data. Furthermore, the Frame Relay is weak in control and administration of the congestion for the data traffic (e.g. drop packets when congestion occurs).Whereas, the upper layers are responsible for the error correction and flow control of an entire payload [1].
The author of paper[5] showed such limitations by comparing the Frame Relay with other networks like MPLS which has low latency and a labeling feature of packets instead of the huge process of routing that is based on IP addresses. He used OPNET simulator tool according to certain parameters as latency, Ethernet delay and traffic received.
-
III. ATM
ATM is a node or a switching device which routes cells. The cell is a basic transmission unit that is always fix length. Therefore, an ATM is also known as a cell relay.
The ATM has features of both circuit switching and packet switching techniques. Such ATM is implemented as controlled hardware and it is designed for using realtime applications with low latency, and also for using high throughput in case of file transfer. The same ATM uses an external or isochronous clocking for transferring data faster[3], and it can be deployed in private, public or hybrid networks either Local Area Network (LAN), Campus Area Network (CAN) or WAN.
The ATM header has two types of formats such as: User-Network Interface (UNI) and Network-Network
Interface (NNI) formats. Each type has two fields in each ATM cell header as Virtual Path Identifier (VPI) and Virtual Channel Identifier (VCI) that are reserved for the virtual connections.
The VPI field works as routing and it uses the virtual values for switching the cells between ATM’s networks. In UNI interface the VPI field has 8-bits value to permit up to 256 virtual path identifiers (i.e. from 0 to 255 values). While, in NNI interface format the VPI field has 12-bits value to allow for 4,096 virtual path identifiers (from 0 up to 4,095).
Whereas, the VCI field is used as switching for the end users. This field has 16-bits value for both UNI and NNI interface formats which can allow to get 65,536 virtual channels. The actual values of such range from 32 to 65,535 are used for virtual circuit connections in ATM network, whereas the other VCIs values are reserved by ATM Forum and ITU-T. The combining of both VPI and VCI is referred to as virtual channel connection (VCC).The VCC is defined as allocating connection of VPI and VCI known as the connection identifier (CI) [1][6][7].
There are some previous studies in the area of the Frame Relay and ATM networks that are summarized and conducted as follows:
Список литературы Frame Relay versus Asynchronous Transfer Mode: A Comparative Study and Simulation
- Kiran R. Barapatre and et.al, "A Performance Comparison of X.25, Frame Relay and ATM in High-Speed Networks-A Review, "I. J. of Innovative Research in Computer and Communication Eng., Vol.5, Issue 3.March 2017.
- Sajad Ahmed Mir, Manmohan Sharma, "A Comparative Study of X.25, Frame Relay and ATM in High-Speed Networks, International Journal Research in Applied Science and Eng.Tech., ISSN: 2321-9653, Vol. 2, Issue 4, April 2014.
- Todd Lammle, CCNA Routing and Switching, Study Guide, 2013.
- Gursimran Singh and Jatinder Singh, "Performance Evaluation and Optimization of TCP Parameters over Frame Relay Network," International Journal on Computer Science and Eng., Vol. 3, 0975-3397, No. 10, October 2011.
- Shahad H. Zwayen and Mustapha B. Ibrahim, "Evaluating the Performance of MPLS and Frame-Relay using OPNET Modeler," International Journal of Computer Applications, ISSN: 0975 – 8887, Vol. 108, No. 12, December 2014.
- BehrouzA. Forouzan, Data communications and Networking, Fifth Edition, 2013.
- Niraj Nake, "An Overview of Network Architecture and Security Framework of Asynchronous Transfer Mode," International Journal of Scientific and Research Publications, Vol. 3, ISSN: 2250-3153, Issue No. 4, April 2013.
- Mr. Adnan Sarwar, and et.al, “Performance Analysis of Data Traffic over TCP/IP, Frame Relay and ATM Networks”, International Journal of Science and Advanced Technology, Vol. 1, ISSN: 2221-8386, Issue No. 8, October 2011.
- Zhonghui Yao and David C. Blight, “Modeling and Simulation of ATM Networks,” unpublished.
- Ehab Aziz Khalil, “Comparative Performance of High Speed Networks Carrying Multimedia”, International Journal of Engineering Sciences & Emerging Technologies, Vol. 3, pp: 9-21 ©IJESET, Issue 1, ISSN: 2231 – 6604, August 2012.
- Erik Chromy, Tibor Misuth, and Adam Weber, “Aapplication of Erlang Formulae in Next Generation Networks”, International Journal of Computer Network and Information Security, DOI: 10.5815/ijcnis.2012.01.08, 2012.
- ArpanBenerjee and et.al, “A Comparative study on CBR and UBR”, International Journal of New Innovations in Engineering and Technology, Vol. 1, Issue 1, June 2012.
- Sandhyarani.S and Chandrasekha R.G, "Network Delay: Network Analyzer and OPNET Simulation Tool," I.J. of CS and IT, Vol. 2, Issue 6, pp. 2620-2626, June 2011.
- Priyanka Grover and Meenakshi Chawla, “Performance Analysis of VOIP Codec’s With QoS Parameters,” International Journal of Advance Foundation and Research in Computer, Vol. 2, Issue 5, ISSN: 2348 – 4853, May 2015.
- Therdpong DAENGSI et.al, IP Telephony: Comparison of Subjective Assessment Methods for Voice Quality Evaluation, Walailak J Sci & Tech 2014; 11(2): 87-92.