I. Cidon and I. Gopal "PARIS: An approach to integrated high- speed private networks", International Journal of Digital Analog Cabled Systems, AprilJune 1988, pp. 77--86  Home/Search   Document Not in Database   Summary   Related Articles   Check

....online algorithms, preemption, call control, call admission AMS subject classifications. 68M20, 68Q20, 68Q25, 90B12, 90B35 1. Introduction. Bandwidth allocation is one of the most important problems in the management of networks that have guaranteed bandwidth policy (e.g. ATM [5] PARIS [9], IBM BBNS [10] In such networks the application has to reserve in advance sufficient bandwidth for its communication. The guaranteed bandwidth policy is contrasted with the more traditional policy (e.g. TCP IP) where information packets are routed as they come to the network without prior ....

I. Cidon and I. Gopal. PARIS: An approach to integrated high-speed private networks. Int. Journal of Digital and Analog Cabled Systems, 1(2):77--86, 1988.

....guarantees. Assuring QoS requires reservation of resources. As a result, B ISDN will likely allocate resources in terms of virtual circuits. Examples of broadband networking technology that uses a virtual circuit based approach are ATM (Asynchronous Transfer Mode) deP91, Bou92] and PARIS PlaNET [CG88, CGG91] The Internet al..so seems to be moving in the direction of using virtual circuits. Traditionally, nodes communicating on the Internet may use a different path for each data packet. As a result, there are no performance guarantees and no resource reservations. However, new routing ....

I. Cidon and I. S. Gopal. PARIS: An approach to integrated high-speed private networks. International Journal of Digital & Analog Cabled Systems, 1(2):77--86, April-June 1988.

....transmission of a message triggers a computation in the receiving node, and communication is only allowed to immediate neighbors. The problem with the above model is that it fails to capture the way distributed algorithms are implemented in many modern network designs, e.g. high speed networks [4] and active networks [5] Cidon et al. 6] were the first to address this problem in the context of high speed networks. They separated the delay associated with transmission and the delay associated with computation, and allowed messages to cutthrough a node with no computation delay penalty. ....

Israel Cidon and Inder Gopal, "PARIS: an approach to integrated high-speed private networks," International Journal of Digital and Analog Cabled Systems, vol. 1, no. 2, pp. 77 -- 86, April-June 1988.

....transmission of a message triggers a computation in the receiving node, and communication is only allowed with immediate neighbors. The problem with the above model is that it fails to capture the way distributed algorithms are implemented in many modern network designs, e.g. high speed networks [3] and active networks [14] Cidon et al. 4] were the first to address this problem in the context of high speed networks. They separated the delay associated with transmission and the delay associated with computation, and allowed messages to cut through a node with no computation delay penalty. ....

I. Cidon and I. Gopal. PARIS: an approach to integrated high-speed private networks. International Journal of Digital and Analog Cabled Systems, 1(2):77 -- 86, April-June 1988.

.... protocols for wireless mobile networks are: the Lightweight Mobile Routing Protocol [3] the Destination sequenced distance vector protocol [4] the MurthyGarcia Luna Aceves protocol [5] and the temporally ordered routing algorithm [7] Examples of systems that have used source routing are PARIS [8], Ring Network Bridges [7] and more recently DSR [6] Although the source routing vector method leads to a larger overhead, it has several advantages that make it preferable in scatternet systems. If connections are based on tables at the relays, we need protocols to construct the table entries ....

....routing vector field (RVF) see Fig. 2) The value BF=0 defines an inter piconet unicast packet, BF=1 defines the packet as an inter piconet broadcast packet. The broadcast case will be discussed in Sec. VI. B. Routing of unicast packets The protocol is similar to the source routing protocol of [8], 9] When BF=0, the RVF contains a sequence of alternating id s LocID, MacAddr, LocID, MacAddr,etc. where LocID is a 3 bit local identifier and MacAddr is a 3 bit MAC address. When the packet is sent by the source node, the latter writes in the RVF field the sequence of id s corresponding to the ....

I. Cidon and I. S. Gopal, "Paris: An approach to integrated high-speed private networks," Intern. J. on Digital and Analog Cabled Systems, vol. 1, pp. 77--85, 1988.

....transmission of a message triggers a computation in the receiving node, and communication is only allowed with immediate neighbors. The problem with the above model is that it fails to capture the way distributed algorithms are implemented in many modern network designs, e.g. high speed networks [3] and active networks [14] Cidon et al. 4] were the first to address this problem in the context of high speed networks. They separated the delay associated with transmission and the delay associated with computation, and allowed messages to cut through a node with no computation delay penalty. ....

I. Cidon and I. Gopal. PARIS: an approach to integrated high-speed private networks. International Journal of Digital and Analog Cabled Systems, 1(2):77 -- 86, April-June 1988.

.... CCITT has recommended that the ATM networks be connection oriented [17] 17 The very limited processing time for each packet leads to a need for developing a dedicated hardware switch that has the ability to route packets very quickly, without consuming the processing capacity [26] According to [26, 25], this can be achieved by building a network node in two parts: the switching subsystem (SS) and the Network Control Unit (NCU) The SS is a fast hardware switch with relatively limited functionality. The NCU is a slower but more sophisticated processor. Packets that are only relayed through the ....

....handled by the SS directly without involvement of the NCU. Packets that require more complex processing, e.g. packets that carry call set up and routing information, are forwarded from the SS to the NCU. Source routing, where every packet carries its own path information, is also recommended in [26, 25] so that packets can be quickly switched by the SS at transit nodes. Although processing requirements can be decreased by doing virtual circuit routing and switching normal packets via a switching subsystem, significant burdens will still be placed on the processing elements in future high speed ....

[Article contains additional citation context not shown here]

Cidon, I. and Gopal, I. PARIS: An Approach to Integrated High-speed Private Networks. International Journal of Digital & Analog Cabled Systems, pages 77-- 86, April-June 1988.

....without disruption of the real time service and thus violating the performance guarantees. Even in somewhat easier case of uni cast (point to point) communication rerouting is considered difficult [GKR94] and it is not used in the current of Gigabit rate networks (e.g. PARIS plaNET [ACG 90, CG88, CGG91, CGG 93] or in ATM standards [XVI88, Bou92] We also will not consider preemption or rejection of connections [GG92, GGK 93] Because of the unfeasibility of the rerouting or preemption options, the routing decisions are in effect irreversible . Since connections arrive in an ....

I. Cidon and I. S. Gopal. PARIS: An approach to integrated high-speed private networks. International Journal of Digital & Analog Cabled Systems, 1(2):77--86, April-June 1988.

....results are presented in section 5. Finally, section 6 summarizes this paper. 2 2 Network Models We consider a connection oriented high speed network consisting of N nodes, N 2, with each node having some number of incoming and outgoing links. We adopt the network node structure described in [16, 5], in which a node consists of two components: the switching subsystem (SS) and the network control unit (NCU) The SS is a fast hardware switch with relatively limited functionality. The NCU is a slower, but more sophisticated, processor. Packets (or cells) that need only be relayed through the ....

I. Cidon and I. Gopal, "PARIS: An Approach to Integrated High-speed Private Networks," International Journal of Digital & Analog Cabled Systems, pp. 77--86, April-June 1988.

....administered by The Israel Academy of Science and Humanities and by a grant of the Israeli Ministry of Science and Technology. 1 1 Introduction Bandwidth allocation is one of the most important problems in the management of networks that have guaranteed bandwidth policy (e.g. ATM [5] PARIS [9], IBM BBNS [10] In such networks the application has to reserve in advance sufficient bandwidth for its communication. The guaranteed bandwidth policy is contrasted with the more traditional policy (e.g. INTERNET) where information packets are routed as they come to the network without prior ....

I. Cidon and I. Gopal. PARIS: An approach to integrated high-speed private networks. Int. Journal of Digital and Analog Cabled Systems, 1(2):77--86, 1988.

....ways to reduce the overhead of information distribution, other than reducing the frequency of global distribution of updates. They did not address the critical issues of scale discussed here. The PARIS network architecture developed at IBM uses source routing in a fast packet switching network [7]. Switches maintain a database of current link utilizations. Routes are chosen based on the link utilizations, and available bandwidth is verified during setup to guarantee that the requirements of real time traffic can be accommodated. The scheme is dependent on fast global distribution of link ....

Israel Cidon and Inder S. Gopal. PARIS: An approach to integrated high-speed private networks. International Journal of Digital and Analog Cabled Systems, 1(2):77--85, 1988.

....on the computational complexity of these problems and on how well simple heuristic procedures can approximate the optimal strategy. 1 Introduction It is generally accepted that future high speed networks will be based on packet switching technology. Various alternatives (e.g. ATM [ATM] PARIS [CG]) have been proposed and are in various stages of development. Underlying all such design alternatives is the assumption that high speed networks will have to support a wide variety of traffic types. Each of these traffic types will have very different requirements in terms of delay, throughput, ....

I. Cidon and I. Gopal, "PARIS: An approach to integrated high-speed private networks," Int. Journal of Digital and Analog Cabled Systems, 1(2), pp. 77-86, 1988.

....link utilization information available at the source at the time, and a set up packet is sent on that path to establish a connection. The set up packet is followed after a short delay, much shorter than the end to end round trip delay required by wait for reservation types of protocols (e.g. 1] [2]) by the data packets, in this way avoiding the pretransmission delay associated with end to end reservations. If the capacity available at a preferred intermediate link is insufficient to accommodate the session, the set up packet and the data packets that follow it may have to be routed over a ....

I. Cidon and I. Gopal, "PARIS: An approach to integrated high-speed private networks," Int. J. Digital Analog Cabled Syst., vol. 1, no. 2, pp. 77--86, Apr.--June 1988.

....in which few cells may be in transit between Leland Window Based Congestion Management 2 one source and its destination. As an alternative approach to access control, rate based windows [3] leaky buckets or input throttles ) have recently received much attention in the industrial [4] [5] [6] 7] and academic [8] 9] 10] research communities. A leaky bucket forces a traffic source not to exceed some specified average rate of traffic input (over some interval) and some specified burst length (maximum sequence of consecutive cells at the peak input rate) Within those ....

Israel Cidon and Inder S. Gopal, "PARIS: An Approach to Integrated High-Speed Private Networks," International Journal of Digital and Analog Cabled Systems, Vol 1 (1988).

....is the job of the communication network to deliver the message from its source to its destination. In many cases, the actual topology of the network can be ignored, since passing a message between any pair of processors takes roughly the same time. For some examples of networks of this type, see [1,2,3,4]. To analyze systems of this type, Bar Noy and Kipnis [5] have developed the postal model, in which it is assumed that processors are connected in a complete network and there is simply a communication latency factor which measures the inverse ratio of the time it takes for a processor to send ....

I. Cidon and I. Gopal, PARIS: an approach to integrated high-speed private networks, Intl. J. of Digital and Analog Cabled Systems, 1(2) (1988) 77--85.

No context found.

I. Cidon and I. Gopal "PARIS: An approach to integrated high- speed private networks", International Journal of Digital Analog Cabled Systems, AprilJune 1988, pp. 77--86

....[Tur86] A second issue, the need for an additional performance and functionally in the network control layer, is much less explored. In this paper we explore this second issue, specifically focusing on the lessons that we have learned during the design and implementation of the PARIS network ([CG88]) We believe that most of our conclusions are general and can be applied to any high speed packet network. The early stage of PARIS was described in [CG88] A subsequent paper [CGG 92] describes a successor to PARIS called plaNET. For clarity, we shall only refer to PARIS though plaNET is ....

....this second issue, specifically focusing on the lessons that we have learned during the design and implementation of the PARIS network ( CG88] We believe that most of our conclusions are general and can be applied to any high speed packet network. The early stage of PARIS was described in [CG88]. A subsequent paper [CGG 92] describes a successor to PARIS called plaNET. For clarity, we shall only refer to PARIS though plaNET is largely similar as far as distributed control is concerned. Both works also include some initial ideas regarding distributed control. In the current paper we ....

[Article contains additional citation context not shown here]

I. Cidon and I. S. Gopal. Paris: An approach to integrated high-speed private networks. International Journal of Digital & Analog Cabled Systems, 1(2):77--86, April-June 1988.

....tree edges in the transient states. Loop Freedom: At all times, the set of tree edges does not contain a cycle. 7 Path Preservation: An edge ceases being a tree edge only in the case that it fails. Loop Freedom is essential in the environment of hardware based fast packet switching [Tur88, CG88, ACG 90] CGK88, CS88] and asynchronous transfer mode (ATM) XVI88] which are the most dominant proposals for the future commercial integrated data voice video networking systems (B ISDN) Path Preservation is important in virtual circuit switching environment. The properties of loop freedom ....

I. Cidon and I. S. Gopal. Paris: An approach to integrated high-speed private networks. International Journal of Digital & Analog Cabled Systems, 1(2):77--86, April-June 1988.

....this paper appeared in Proc. 2nd Israel Symp. on the Theory of Computing and Systems, Natanya, Israel, June 1993. IBM T.J. Watson Research Center, P.O. Box 704, Yorktown Heights, NY 10598. Computer Science Department, Tel Aviv University, Israel. 1 Introduction High speed networks (e.g. [13, 5, 9, 8]) as providers of multimedia services, will have to support a wide variety of traffic types. Each of these traffic types will have very different requirements in terms of required duration and throughput, and tolerable delay and message loss. Many of these requirements will only be met if ....

I. Cidon and I. Gopal, "PARIS: An approach to integrated high-speed private networks, " Int. Journal of Digital and Analog Cabled Systems, 1(2), pp. 77-86, 1988.

....of the project is plaNET. The basic components are a Gigabit sec WAN composed of plaNET switches and a Gigabit sec MAN LAN called ORBIT which acts as a feeder or local access mechanism for the backbone WAN. To clear up a frequent confusion, plaNET and ORBIT are based on, respectively, the PARIS [14, 19, 20] and METARING [27, 42] projects. Thus, in many other technical papers, these names are used instead of plaNET and ORBIT. In this paper we present an overview of the plaNET ORBIT system. We focus on the functions provided by the network and on the essential hardware and software components. There ....

I. Cidon and I. S. Gopal, "PARIS: An Approach to Integrated High-Speed Private Networks," Int'l. J. Digital and Analog Cabled Sys., vol. 1, no. 2, pp. 77-86, April-June 1988.

....years. One factor has been the dramatic increase in the capacity of the communication links. The advent of fiber optic media has pushed the transmission speed of communication links to over a Gigabit sec, representing a significant increase over typical links in today s packet switched networks [5]. A second factor is the altered nature of traffic transmitted through these networks. It is now accepted that packet switched networks [1] 5] or variants of packet switching like ATM) will form the basis for multimedia high speed networks that will transmit voice, data and video through a ....

....speed of communication links to over a Gigabit sec, representing a significant increase over typical links in today s packet switched networks [5] A second factor is the altered nature of traffic transmitted through these networks. It is now accepted that packet switched networks [1] [5] (or variants of packet switching like ATM) will form the basis for multimedia high speed networks that will transmit voice, data and video through a common set of backbone nodes and links. Both these factors have a significant impact on the design of the protocols and control procedures within ....

[Article contains additional citation context not shown here]

I. Cidon and I. S. Gopal, "PARIS: An Approach to Integrated High-Speed Private Networks," International Journal of Digital & Analog Cabled Systems, Vol. 1, No. 2, pp. 77-86, April-June 1988.

....burst transmission. The suggested bypass algorithm is simple enough to be implemented by cheap hardware. Before proceeding, we describe two routing approaches for high speed networks with which our algorithm can be used: source routing and ATM. Source routing, or Automatic Network Routing (ANR) [5], is a routing method where each packet carries in it the entire route it should traverse. In our discussion, we will assume that the route is placed in the header as a list of port IDs (or link IDs) and each node along the packet route strips the ID it uses from the head of the list (in ....

.... route is placed in the header as a list of port IDs (or link IDs) and each node along the packet route strips the ID it uses from the head of the list (in practice, there are other methods for handling the source route that only differ in technicalities and can be integrated with our algorithm [5]) In networks that employ source routing, the route for the session is computed at the source node using data that is distributed by a topology update algorithm. It is therefore plausible that routes thus computed are not optimal (and may not even be feasible) Changing the route on the fly ....

[Article contains additional citation context not shown here]

Israel Cidon and Inder Gopal. PARIS: an approach to integrated high-speed private networks. International Journal of Digital and Analog Cabled Systems, 1(2):77 -- 86, April-June 1988.

....the architecture of hardware based fast packet switches. The ability to implement cost effective high speed switches in silicon is (along with the advance of fiber optic technologies) the enabling technology behind the recent rise of broadband integrated networks in particular the ATM architecture [1, 2, 3, 4, 5, 6]. Fast hardware based switches have also been at the core of the emerging switched LAN (also termed switching hubs) products as well as the new generation of Gigabit routers [http: www.bbn.com magazine techwatch router.html] Current switching systems designs in high speed packet switching ....

I. Cidon and I. S. Gopal. PARIS: An approach to integrated high-speed private networks. International Journal of Digital and Analog Cabled Systems, 1(2):77--85, April 1988.

No context found.

I. Cidon and I. Gopal. PARIS: an approach to integrated high-speed private networks. International Journal of Digital and Analog Cabled Systems, 1(2):77 { 86, April-June 1988.

No context found.

I. Cidon and I. S. Gopal. Paris: An approach to integrated high-speed private networks. International Journal of Digital & Analog Cabled Systems, 1(2):77--86, April-June 1988.

First 50 documents  Next 50

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC