Asenkron transfer modu ve çerçeve aktarma
Asynchronous transfer mode and frame relay
- Tez No: 66720
- Danışmanlar: PROF. DR. GÜNSEL DURUSOY
- Tez Türü: Yüksek Lisans
- Konular: Elektrik ve Elektronik Mühendisliği, Electrical and Electronics Engineering
- Anahtar Kelimeler: Belirtilmemiş.
- Yıl: 1997
- Dil: Türkçe
- Üniversite: İstanbul Teknik Üniversitesi
- Enstitü: Fen Bilimleri Enstitüsü
- Ana Bilim Dalı: Elektronik ve Haberleşme Mühendisliği Ana Bilim Dalı
- Bilim Dalı: Belirtilmemiş.
- Sayfa Sayısı: 134
Özet
ÖZET Genişbandlı ISDN (Broadband ISDN ; B-ISDN) telefon, düşük hızda veri iletimi, faks ve teleks gibi dar-bandlı ve görüntÜlü teLEfon, yüksek hızda veri iletimi, video konferans, renkli faks, HDTV, yüksek çözünürlülüklü resim iletimi gibi geniş- bandli hizmetleri verbilen bir şebeke olarak tasarlanmıştır. B-ISDN, bağlaşmak, yan kalıcı, kalıcı, noktadan noktaya, noktadan çok noktaya bağlantılar içermektedir. Bu bağlantılar tek ve çoklu- ortam tipinde, bağlantı yönelimli ya da bağlantısız yapıda tek yönlü veya iki yönlü olabilir. Asenkron Transfer Modu (Asynchronous Transfer Mode ;ATM), B-ISDN' in üzerine kurulduğu temel teknolojidir. Devre ve paket bağlaşma tekniMerinin avantajlarım bünyesinde birleştirmektedir. ATM' de hücre aktarma (Cell Relay ;CR) tekniği kullanılmaktadır. Bu teknikte diğer paket bağlaşma tekniklerinden farklı olarak sabit uzunluklu hücre adı verilen paketlerin iletimi yapılmaktadır. ATM teknolojisinde kullanıcı verisi önce sabit uzunluklu bloklara ayrılır daha sonra ATM hücrelerinin içerisine yerleştirilerek başlık bilgisi eklenir ve diğer ATM hücreleri ile beraber Asenkron zaman bölmeli çoğullama yöntemi ile iletilir. Bu çalışmada incelenen ikinci bir teknoloji de Çerçeve aktarma (Frame Relay ; FR) teknolojisidir. Çerçeve aktarma, kamu ve özel veri şebekleri için geliştirilmiş standart temelli bir çözümdür. FR şebekesinde bilgi, paketler halinde gönderilir. Çerçeve adını alan bu paketler farklı uzunlukta olabilir ve bir uçtan bir uca, bandgenişliğinin maksimum kullanımı ile en kısa sürede gönderilir. FR teknolojisi ile düğüm seviyesinde ortaya çıkan transmisyon hatalarının saptanması mümkündür. X.25 şebekesinin aksine FR şebekesinde kaybolan çerçevenin yeniden alınması, geri getirilmesi işlemi uç birimlerdeki uygulamalara bırakılmıştır. İletimin bütünlüğünden kaynak ve hedefteki donanımlarda işletilen üst seviye protokolleri sorumludur. Gönderilecek çerçeveler bireysel olarak adreslenir ve seri olarak yığınsal halde gönderilir. Yöneltme ve tıkanma kontrol bilgisi de çerçeve içinde iletilir. Çerçeve sıralama bilgisi gönderilmez ; hata içeren bir çerçeve elenir. FR, diğer teknolojilere göre bandgenişliğinin dinamik olarak kullanımına imkan verdiği için daha verimlidir. FR şebekesinde bandgenişliği herhangi bir uygulamaya tahsis edilmiş değildir. Eğer yeterli bandgenişliği varsa FR şebekesinde tüm imkanlar daha hızlı ve verimli iletişim için kullanılabilir. FR, paket temelli bir teknoloji olduğundan dolayı ATM ile benzer yönleri bulunmaktadır. Teknolojiler arasındaki temel fark, FR'de değişken uzunluklu çerçevelerin, ATM 'de ise sabit uzunluklu hücrelerin iletilmesidir. FR teknolojisinde transfer gecikmesi X.25 'e göre daha azdır buna rağmen gerçek zamanlı hizmetler ancak küçük ve sınırlı şebekelerde verilebilmektedir. FR şebekesi aboneleri, ATM şebekesi aboneleri ile beraber çalışma fonksiyonu (IWF) aracılığı ile bağlantı kurabilmektedir. Ayrıca ATM şebekesi, FR şebekesi abonelerinin birbirleri ile bağlantı kurması için omurga şebeke olarak da kullanılabilmektedir.
Özet (Çeviri)
SUMMARY ASYNCHRONOUS TRANSFER MODE AND FRAME RELAY With popularization and increasing use comes a demand for more sophisticated services; the integration of voice and video along with the traditional text traffic. Thus the requirement for a technology capable of meeting these requirements in a scalable way. Over the past 25 years the Internet has evolved, using the technology most appropriate for the time. Along the way, leased lines, packet data services, satellites, and now ATM have appeared. By using ATM, we have a single technology capable of supporting the various voice, video, and data services to the home or business, supplanting existing dedicated networks. Even today, we see digitized broadcasts ditsributed to-TV-top decoders and Enterprise voice and data wide-area networks all based on ATM. In 1996, ATM is rightfully becoming the technology of choice for a number of environments, including workgroups in small and large businesses, high speed trunking within the service providers, and even video and information distribution to the residence. Over the next decade, ATM technology will increasingly provide the backbone network. Asynchronous Transfer Mode commonly known as ATM, is a networking technology capable of transposting all higher forms of intelligence voice, video, and data, from one user to another over a local or wide area. This universality is the“why”of ATM, a single platform and technology resulting in cost savings and quality of service improvements for both ens users and service provider. This is not saying that ATM is the ideal technology on which to base any one of these services. Many alternatives exist, but none are suitable for use in both the local and wide area, or have the widespread vendor and service provider support which ATM has gathered. In any case, within a production networking anvironment, ATM will find its place among other networing technologies as part of an end-to-end solution. The term“ATM”is used to describe the Broadband Integrated Services Digital Network (B-ISDN), having been adopted by the internet working and computer industry, as well as by the world press, to designate what is actually a combination of technology and services. The B-ISDN was and is a concept providing integrated services to the user, an outgrowth of 64 Kbps-ISDN (sometimes called Narrow band ISDN). ATM is a technology which resides directly above the physical infrastructure: the fiber, copper, or wireless transmission system. In fact, ATM operates at the physical layer, providing connection-oriented MAC like service to voice transport between PABX, video transport between video CODECs, and of course data transport. However, this forced classification usually breaks down in reality since ATM exhibits many characteristics of higher-layer protocols under actual deployment, looking at the data environment, the International Standardization Organization (ISO) has defined a layered internet working reference model known as the Open Systems Interconnect (OSI) reference model. This model is a method by which interactionsbetween various processes within and-systems (that is, PCs, workstations, and routers) and intermediate-systems (that is, ATM switches) may be mapped to each other in a logical way, helping to guarantee interoperability. An idea of where ATM fits within this reference model help in its understanding and capabilities, and tough the model is more relevant to the data networking world, its concepts may be applied to voice and video as well. ATM is capable of utilizing most deployed physical media types and transmission infrastructures. In the case of data networking, higher-layer services are based on well-defined data link, network, transport, session, and presentation layers, while video and voice support may rely video or voice encoding, and the ATM layer. Layer 3, the network layer, is sometimes known as the internetwork layer based on its relevance for end-to-end internetworking. This is one area where ATM does not map effectively to the layering model, since the ATM layer will contain many features commonly associated with the network layer, such as hierarchical addressing and routing. The most basic concept of ATM, and what differentiates it from past technologies, is the equal length cell structure. All higher-layer traffic, be it voice, video, or data, is broken into 48 byte blokes and then header is added, so ATM“cells”, 53-byte packets containing a header and data, payload is created. The header contains information identifying a particular virtual circuit, with many virtual circuits sharing a single physical interface, is no different than that found within X.25 and Frame relay. Unlike these two technologies, an equal lenght packet allows switches with very high throughput with almost all processing taking place in hardware. The ATM cell on its own, however, does not guarantee higher performance. This requires proper switch desgn, network design, and software within the ATM network supporting higher-layer services. In fact, ATM many have just the opposite effect on data traffic unless these guidelines are adhered to. As user traffic arrives at the ATM interface within a computer, router, video CODEC, or PABX, it is segmented into the 53-byte cells describes above. Over a single interface, these cells are interleaved and sent over the interface as they are generated. This contrasts to traditional Time Division Multiplexing (TDM) architectures where time slices are pre- allocated for the various traffic types, thus, the term Asynchronous Time Division Multiplexing may more adequately describe ATM. A second fundamental concept behind ATM is that it is connection-oriented. Unlike shared LAN technologies or the router-based segment of the Internet, where data is packetized and sent to its destination along a route calculated on a hop-by-hop basis, systems participating in ATM will signal a connection across the network, Once the connection has been established, it is expected to be able support the Quality of Service (QoS) requirements of the sources and destinations, to include parameters such as bandwidth, probability of discarded data, and delay, this QoS support is critical when ATM is used as a transport for services to include PABX interconnect or even reliable data transmission. Tough a datagram-oriented network may support QoS by making use of various quening and resource reservation techniques in the support of voice and video, it will not support the large installed base of more traditional voice and video qeuipment. Time division multiplexers support the services, but are not usually deployed in the local area. Thus, ATM is the only technology which truly support these requirements over both the local and wide area. xiA commonly asked question is whether or not ATM is the most efficient solution for data networking, if in fact alternative technologies exist which are capable of providing QoS support. The deployment of ATM must really be looked at in the context of multiple services: data, voice and video. In many environments, such as a high-speed campus backbone or high-performance workgroup, ATM will have definite advantages. It is the first networking technology which allows users to economically deploy a common physical infrastructure and signaling mechanism from end-station to end-station over local-and wide-area ATM network segments. Whereas existing networks translate from Ethernet, Token Ring, or Fiber Distributed Data Interface (FDDI) in the local area, with inherent complexity, ATM provides for a single end-to-end architecture where appropriate.“Appropriate”is defined by the network architect after an analysis of applications and of available local-and wide- area services, and tough this end-to-end transparency is a promise of ATM, actual implementations will combine many of the above technologies. The“why”of ATM therefore is probably best answered by looking at the business drivers behind its implementation and then comparing the promise of ATM with reality. Frame relay has its origins in the ISDN specifications, of the 1980s. The ISDN (sometimes now called to narrow band ISDN, or N-ISDN because of the emerging very high speed integrated networks) provides a single network interface to a user with the capability of connecting to many different types of network service. These include circuit switched data connections, voice switching services, access to packet switching services, and access to frame relay services. The frame relay service within the ISDN was designed to provide a very high speed packet switching style data service. This could then be used for the interconnection of devices which required high throughput for short durations (such as LAN routers). The ISDN as a technology has been very slow to take off in many countries, but the principles behind frame relay were recognized as having a use for applications outside the ISDN. Consequently, the protocol was developed for use in a stand-alone environment. Frame relay is based upon the principle of packet switching, not time division multiplexing, and consequently is better suited to data applications. In frame relay, data is divided into variable length frames (similar to the packets in packet switching), all of which include addressing information. These frames are then forwarded into the frame relay network, which attempts to deliver them to required destination. This philosophy, so far, is identical to that of packet switching. However, the main difference between frame relay and packet switching is in the implementation of the protocol itself. Packet switching operates at layer 3 of the OSI model, while frame relay operates 2 and even then does not implement all of the layer 2 functions. The major difference between frame relay and other forms of packet switching is that frame relay implements all of its required protocol functions within level 2 of the OSI model (also known as the frame or data link level). Even here it does not implement all of the functions traditionally associated with packet switching. The subset of the level 2 protocol that is implemented by frame relay is known as the“core”funtions. There are three main functions that are conducted within a frame relay switch. 1. Check that no bit errors have occurred within the frame by examining the frame check sequence. If there have been any errors, discard the frame. xii2. Read the addressing information within the frame, and route the input frame to the appropriate output link. 3. Check that the frame relay switch is not congested. If it is, set the congestion notification bits, or discard the frame. Unlike X.25, frame relay has no level 3 protocol. Instead to the switch having to access the level 2 and level 3 information, it only has to examine the level 2 information. Furthermore, the amount of processing that takes place at level 2 of frame relay is significantly less than that required at an interface such as X.25. A part of the X.25 protocol is the mechanism called flow control. This is the procedure which controls the rate at which the originating device issues packets into the switch, and is controlled at both levels 2 and 3. If the receiving switch is unable to accept any packets from the origination device, for reasons of congestion or lack of buffer space, it will not acknowledge receipt of packets, and issues a message saying“send no more packets”. When the congestion within the receiving switch clears, it will send the relevant“OK to send again”message to the originating device. This guarantees that the reeiving switch will never have to discard data for reasons of lack of buffer capacity. Frame relay has no equivalent process. It is perfectly possible for a frame relay access device to continue to send frames into the receiving switch even when there is insufficient buffer space to hold them. If this situation occurs, the receiving switch is able to discard any further received frames without notification, leaving the originating device to recover from the loss of data. Frame relay is an interface specification and only refers to the mechanism by which the user offers traffic into the network. The internal protocol within the network will be proprietary and vendor-dependent. This is a similar situation to X.25 networking. However, while both frame relay and X.25 rely upon proprietary network protocols, it is highly likely that they will use different protocols. X.25 network will probably use a derivative of the X.25 level 2 protocol inside the network, while frame relay networks will probably use a derivative of the level 2 core functions. One way in which we will realize ATMs promise as a multiservice offering is by internetworking native ATM users with those connected to a Frame Relay (FR) network. As described earlier, the interworking frame work by which users connected via an Fr-UNI on a Frame Relay or ATM switch may interworking may take to forms. The first, known as network interworking, segments the FR frame at a device known as an Interworking Function (TWF) but, preserves the native FR encapsulation based on RFC 1490. The IWF, though often depicted as a separate entity, will usually exist as part of the FR interface on an ATM switch or as part of an ATM interface on an FR switch. For this type of interworking, the ATM-UNI must also implement the FR encapsulation across one or more VCCs. This is supported by the FR-SSCS, and requires that the ATM system knows that the destination is reachable only across an FR network, and therefore uses NLPID for any VCCs directed to the IWF. Service working requires encapsulation conversion at the IWF from RFC 1490 to RFC 1483. The ATM user now does not care if the destination is reachable directly over ATM or via Frame relay,. When considering Frame Relay in conjunction with ATM, we often also look at a third function, that of using the ATM network to xiiiprovide for Frame Rlay transport. We first look at the simpler service interworking, and then introduce ATM as an FR transport. Frame Relay will succeed both as an access method for ATM and as a service offered over ATM. In the former case, the large installed base will create a demand for service interworking, whereby a user connected to a native ATM service is provided with transparent interworking with locations connected to Frame Relay. As a service offered over ATM, the service providers will realize economies of scale by deploying Frame Relay as one of many services across an ATM backbone. Of all the data services, Frame Relay is expected to be the most important in complementing ATM, as evidenced by projected 1996 service revenues of $1.6billion and equipment sales of $1.2 billion. This is by far the fastest uptake of any networking service to date. xiv
Benzer Tezler
- Çok-hızlı ISDN'de geniş bantlı çağırma kurma servisi ve LAN uygulamaları
Multirate ISDN wideband call processing and LAN applications
KENAN ŞAHİN
Yüksek Lisans
Türkçe
2000
Elektrik ve Elektronik Mühendisliğiİstanbul Teknik ÜniversitesiPROF.DR. GÜNSEL DURUSOY
- Yüksek hızlı yerel alan ağlarının tasarımı ve uygulamaları
Design and applications of high-speed local area networks
SAİT ESER KARLIK
Yüksek Lisans
Türkçe
1999
Elektrik ve Elektronik MühendisliğiUludağ ÜniversitesiElektronik Mühendisliği Ana Bilim Dalı
DR. GÜNEŞ YILMAZ
- ATM (asenkron transfer modu) şebekelerde trafik kontrolü
Trafic control in ATM (asynchronous transfer mode) networks
GAZİ KARAKUŞ
Yüksek Lisans
Türkçe
2004
Elektrik ve Elektronik Mühendisliğiİstanbul Teknik ÜniversitesiElektronik ve Haberleşme Mühendisliği Ana Bilim Dalı
PROF.DR. GÜNSEL DURUSOY
- Quality of service for voice over ip with various queueing mechanisms on point to point links
Uçtan uca bağlantılarda değişik sıralama mekanizmaları kullanarak ip üzerinden ses iletiminde hizmet kalitesi sağlama
DEVRİM YENER KÜÇÜK
Yüksek Lisans
İngilizce
2000
Elektrik ve Elektronik MühendisliğiBoğaziçi ÜniversitesiElektronik Mühendisliği Ana Bilim Dalı
DOÇ. DR. HAKAN DELİÇ
- ATM networks and ATM switching architectures
ATM (Asenkron transfer modu) ağları ve ATM seçici mimarileri
TOLGA AÇIKÖZ
Yüksek Lisans
İngilizce
1998
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrolİstanbul Teknik ÜniversitesiKontrol ve Bilgisayar Mühendisliği Ana Bilim Dalı
DOÇ. DR. MEHMET BÜLENT ÖRENCİK