Jain, R., Routhier, S.A., "Packet Trains - Measurements and a New Model for Computer Network Traffic", IEEE Journal on Selected Areas in Communications, Vol. 4, No. 6, September 1986, pp. 986 - 995.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....can capture an appropriate K value 13 N O N O N O N O N O N O N O N O N O N O F F O F F O F F O F F Figure 7: Packet data traffic that reduces the net signaling cost for RC. In this study, we combine the ETSI packet data model [4] with the ON OFF source model (also known as a packet train model) [7]. As shown in Figure 7, we assume that the packet data traffic consists of communication sessions, where the inter session idle period has an exponential distribution with mean 1= p2 . For general inter session idle periods, the variances of the idle period distributions have similar effects as ....

Jain, R. and Routhier, S.A. Packet Trains: Measurements and a New Model for Computer Network Traffic. IEEE Journal on Selected Areas in Communications, 4(6):986--995, 1986.

.... traffic, an interesting connection between the above two phenomena was established by Willinger and Taqqu et al. in [48] Based on a model originally introduced by Mandelbrot in [32] this paper claimed that the superposition of a large number of ON OFF sources (also known as packet trains [25]) whose ON periods ( train lengths ) and OFF periods ( intertrain distances ) exhibit heavy tails produces aggregate network traffic that exhibits self similarity. In fact, as both the number of sources and the block aggregation 12 size become large enough, the cumulative traffic approaches a ....

R. Jain and S.A. Routhier. Packet trains: Measurements and a new model for computer network traffic. In IEEE Journal on Selected Areas in Communications, volume 4, pages 986--995, 1986. 52

....that the superposition of several MMPP(n)s is again a MMPP(n) IPP can be viewed as a special case of the MMPP, while the MMBP(n) can be considered as the discrete time counterpart of the MMPP (n) but without the closure property. The concept of Packet trains model was first introduced in 1986[6]. It assumes that a group of packets travel together as a train(only need to make routing decisions for the locomotive) contrary to the car model , such as Poisson process which assumes that packets are independent of each other. The Packet train model can also be described by a two state Markov ....

R Jain, S. A.Routhier, "Packet Train-Measurements and a New Model for Computer Network Traffic", IEEE JSAC, Vol. SAC-4, No.6, Sep, 1986,pp986995

....by interaction between control and data traffic. The large tails that we observed in the delay distributions may in fact be due to some form of traffic synchronization that we do not completely understand. We think this is an issue that needs attention. In other measurements, Jain and Routbier [18] reported that successive packets on a Local Area Network belonged to the same end to end transmission entity. Moghul [28] has reported existence of similar localities at the level of processes. Feldmeier [7] has used locality of traffic for routing caches in gateways. Gussella [13] has studied ....

Jain, R., and S. Routhier, "Packet Trains: Measurements and a New Model for Computer Network Traffic," Tech Report MIT/LCSTM 292, Department of Elec- trical Engg, MIT, Nov 1985.

.... is typically one of the more expensive operations in handling inbound protocol segments [Fel90] Many recent protocol optimizations for the Transmission Control Protocol (TCP) Jac88] assume that a large component of TCP traffic is bulk data transfers, which result in packet trains [JR86]. If packet trains are prevalent, there is a high likelihood that the next TCP segment is en route to the same application (i.e. uses the same PCB) as the previous TCP segment. In these environments a very simple one PCB cache like those used in BSD systems yields very high cache hit rates. ....

Raj Jain and Shawn Routhier. Packet trains-- measurements and a new model for computer network traffic. IEEE Journal on Selected Areas in Communications, SAC-4(6):986--995, September 1986.

.... of services offered and the need to interoperate with several different protocols [1] Current internetworking devices execute the entire packet forwarding task for each packet arrival, despite much of the packet header information between arrivals is redundant due to network traffic locality[2] [3], 6] Consequently, there have been several attempts to take advantage of network traffic locality to reduce processing overhead at the LAN interconnection nodes, thereby improving transfer delay[3] 4] 5] Typically, such attempts involve caching the previously processed header and predicting ....

.... of the packet header information between arrivals is redundant due to network traffic locality[2] 3] 6] Consequently, there have been several attempts to take advantage of network traffic locality to reduce processing overhead at the LAN interconnection nodes, thereby improving transfer delay[3], 4] 5] Typically, such attempts involve caching the previously processed header and predicting that the next packet will require the same processing as the previous packet. In this paper, a new scheme is proposed for header prediction based upon the interframe gap between arriving packets. ....

[Article contains additional citation context not shown here]

Raj Jain & Shawn A. Routhier, "Packet Trains---Measurements and New Model for Computer Network Traffic," IEEE Journal on Selected Areas in Communications,Vol. SAC-4, No. 6, pp. 986995, September 1986.

....of X. If fi . 2, the distribution has in nite variance while if fi . 1; the distribution has in nite mean and in nite variance. The optimal value of fi is considered between 1 and 2 [29] One mechanism to generate this behavior is the use of the ON OFF model, proposed by Jain and Routhier [11] for Local Area Network (LAN) Wilinger, Taqqu, Wilson, and Leland [21] remedied some of the shortcomings of the ON OFF model and focused on conditions on the ON OFF model for it to generate self similar trac. Aggregating a large number of 0 1 processes whose ON or OFF periods are heavy tailed ....

R. Jain and S. A. Routhier. Packet trains: Measurements and a new model for computer network trac. pages 986 995. IEEE J. Select. Areas Commun.,vol. SAC-4, 1986.

....Folded Bus Simulation Points Folded Bus (No SIGs,a) Dual Bus Dual Bus w Reuse A B C A B C A B C Figure 10: Mean Delay Model and Simulation M J K SIG Overhead Available Slots Figure 11: The come right in model. M slots in a frame with J of those slots frame overhead. in [18]. The following symbols are used for the analysis. n i = Number of enqueued packets at the beginning of slot i. i = Number of arrivals generated in the ith slot. M = Frame length in slots. J = Number of overhead slots. The analysis begins by embedding a Markov chain at the beginning of ....

R. Jain and S. Routhier. Packet trains-measurements and a new model for computer network trac. IEEE Journal on Selected Areas in Communications, SAC4 (6), September 1986.

....and applications used for data transmission. Bulk data transfer activities, such as file transfer, generate bursts of packets between a particular pair of hosts. Interactive terminal traffic also generates highly correlated bidirectional traffic streams. Models such as the packet train model [8] or the packet tandem model [10] have been proposed to explain such behaviour. Such models have proven to be useful not only in the improvement of techniques for modelling workloads, but also in their application to network design. Several recent research efforts have identified the presence of ....

....[10] have been proposed to explain such behaviour. Such models have proven to be useful not only in the improvement of techniques for modelling workloads, but also in their application to network design. Several recent research efforts have identified the presence of network traffic locality [8, 9, 11], and suggested ways to exploit the locality property in the design and operation of computer networks [2, 6, 14] Exploiting locality, usually by caching recently referenced items, has been crucial in the success of virtual memory systems, multiprocessor systems, and distributed file systems, ....

[Article contains additional citation context not shown here]

R. Jain and S. Routhier, "Packet Trains -- Measurement and a New Model for Computer Network Traffic", IEEE Journal on Selected Areas in Communications, Vol. SAC-4, No. 6, September 1986, pp. 986-995. 20

.... traffic has been characterized at the aggregate level with significant discoveries that packet inter arrival times are not exponentially distributed, that there exist correlations in source destination pairs, and that Long Range Dependence (LRD) may be the primary determinant of queuing behavior [4, 10, 12]. The causes of LRD have been largely attributed to heavy tailed distributions of file sizes [15] and user behavior [3] at the application layer (Web browsing in the case of [3] TCP IP behavior has been extensively studied with the primary motivation of better tuning TCP for higher ....

R. Jain and S. Routhier, "Packet Trains - Measurements and a New Model for Computer Network Traffic," IEEE Journal on Selected Areas in Communications, Vol. 4, No. 6, pp. 986-995, September 1986.

.... traffic, an interesting connection between the above two phenomena was established by Willinger and Taqqu et al. in [48] Based on a model originally introduced by Mandelbrot in [32] this paper claimed that the superposition of a large number of ON OFF sources (also known as packet trains [25]) whose ON periods ( train lengths ) and OFF periods ( intertrain distances ) exhibit heavy tails produces aggregate network traffic that exhibits self similarity. In fact, as both the number of sources and the block aggregation size become large enough, the cumulative traffic approaches a ....

R. Jain and S.A. Routhier. Packet trains: Measurements and a new model for computer network traffic. In IEEE Journal on Selected Areas in Communications, volume 4, pages 986--995, 1986.

....Another difference is that the proposed approach attempts to capture the correlation among the known correlated traffic sources while still taking advantage of the likely non correlation (independence) of other traffic sources. Another example of a correlated traffic model is the packet train [22] model. However, it is not clear how this model can be used in the context of call admission control. An alternative approach that can be used to handle highly correlated 4 traffic is to transmit such traffic through available bit rate (ABR) or best effort service. The idea behind ABR service is ....

Raj Jain and Shawn A. Routhier. Packet trains---measurements and a new model for computer network traffic. IEEE J. Sel. Areas Commun., 4(6):986--995, September 1986.

....best effort traffic nor control messages yet. CBR traffic is synthetic, and is composed of connections randomly chosen from the following set of average bandwidth requirements: f64 Kbps, 1.54 Mbps, 55 Mbpsg. VBR traffic is semisynthetic. We model it with a pattern similar to the train packets in [8], but with timing parameters obtained from MPEG 2 video traffic. MPEG 2 video coding standard [13] encodes the video streams as a sequence of different frame types, I, P, and B, ordered with a prefixed and repetitive pattern, called GOP (Group Of Pictures) The GOP we use is IBBPBBPBBPBBPBB. I ....

R. Jain and S. Routhier. "Packet trains -- Measurement and a new model for computer network traffic," IEEE Journal on Selected Areas in Communications, pp. 986--995, September 1986.

....to characterize the secondary load (in terms of distribution of packet lengths) which is induced by a complex primary load representing e.g. the overlay of single video sources in a video server. Among others, our results would allow us to directly use and easily parametrize a packet train model [16] (with deterministic intertrain and intercar times) as a realistic description of the secondary load to be expected. 7 Validation of Our Transformer Approach in Secondary Load Characterization by Means of Analytical Modeling We now want to validate the accuracy of our method for secondary load ....

R. Jain, S.A. Routhier: \Packet Trains - Measurements and a New Model for Computer Network Trac", IEEE J. on Sel. Areas in Comm., Vol. SAC-4, No. 6 (1986), 986-995

....of this paper. The main emphasis is on the performance characteristics of the network like average setup time and blocking probability. Both of these parameters are especially important if burst switching [3] is applied. Computer traffic has a bursty behaviour that is the basis of the train model [9] and the basis of the traffic model of our simulation [8] Burst switching can result in a better utilisation. However, if different bursts have considerably different set up times and bursts are blocked within the call then there is less QoS guarantee for the whole connection. That is, the main ....

R. Jain, S. A. Routhier, "Packet Trains-Measurement and a New Model for Computer Network Traffic", IEEE Journal on Selected Areas in Communications, Vol. 4, No. 6, February 1986

....between the start of connections and the holding time parameters. 4.3.2 Class 2 Class 2 traffic is connection oriented without guaranteed bandwidth. We have chosen two models for class 2 connections: file transfer and variable bit rate voice. The models are based on the concept of packet trains [19]. Voice has been modeled as an ON OFF source with an average ON duration of 1.2 seconds [16, 20] The voice model can also be used as an aggregate of remote login, electronic mail, and database access connections. Packet trains have been shown to exist for TCP connections carrying those types of ....

....has been modeled as an ON OFF source with an average ON duration of 1.2 seconds [16, 20] The voice model can also be used as an aggregate of remote login, electronic mail, and database access connections. Packet trains have been shown to exist for TCP connections carrying those types of traffic [19]. The file transfer model is based on the one given in [20] with an average page size of 16 Kbytes. We chose a peak rate of 50 Mb s as a high estimate, and we made the train interarrival time large enough to allow for memory access delay. Both types of connections are full duplex, so source ....

[Article contains additional citation context not shown here]

R. Jain and S. A. Routhier, "Packet trains --- Measurements and a new model for computer network traffic," IEEE Journal on Selected Areas in Communications, vol. 4, no. 6, pp. 986--995, Sept. 1986.

....to characterize the secondary load (in terms of distribution of packet lengths) which is induced by a complex primary load representing e.g. the overlay of single video sources in a video server. Among others, our results would allow us to directly use and easily parametrize a packet train model [16] (with deterministic intertrain and intercar times) as a realistic description of the secondary load to be expected. 7 Validation of Our Transformer Approach in Secondary Load Characterization by Means of Analytical Modeling We now want to validate the accuracy of our method for secondary load ....

R. Jain, S.A. Routhier: \Packet Trains - Measurements and a New Model for Computer Network Trac", IEEE J. on Sel. Areas in Comm., Vol. SAC-4, No. 6 (1986), 986-995

....with the trace analysis and trac modeling, and then describe related work in the ow management without full per ow state. A. Analysis of Traces Understanding the characteristics and properly modeling trac pattern is important for network designers. However, as pointed out by Jain, et al. [6], Internet trac is not easily modeled, because it does not t classical Poisson process models. There are a number of research e orts to nd a good model [1] 6] 7] 8] 9] 10] 11] 12] 13] but none seems yet as clear cut and easy to handle as the classical Poisson models. B. NLANR ....

....the characteristics and properly modeling trac pattern is important for network designers. However, as pointed out by Jain, et al. 6] Internet trac is not easily modeled, because it does not t classical Poisson process models. There are a number of research e orts to nd a good model [1] [6], 7] 8] 9] 10] 11] 12] 13] but none seems yet as clear cut and easy to handle as the classical Poisson models. B. NLANR Data The NLANR is an organization concentrating on collection, analysis and research of Internet trac data. The organization builds infrastructure and performs ....

[Article contains additional citation context not shown here]

R. Jain and S. A. Routhier, \Packet trains - Measurements and a new model for computer network trac" IEEE Journal on Selected Areas in Communications, vol. 4, pp. 986-995, 1986.

....Wide Area Network Traffic Model, M 48 The degree of self similarity is expressed by the Hurst parameter H in equation 3.4. H varies between 0:5 and 1, where a larger value indicates a higher degree of self similarity. For a short range dependent process, such as the Poisson based models in [46, 51], the Hurst parameter will be approximately 0:5; thus, by (4) the correlation of a Poisson process will fall off as 1=m where m is called the aggregation level. Using the reference curves in Figure 3.14 we see that the correlation structure of the traces correspond to self similar processes with ....

....across the 100 ms interval, or use Poisson based interarrivals. However, the analysis in [80] shows that arrivals, when viewed at millisecond granularity, occur in bursts rather than a continuous flow. S models the short term arrival using the well known packet train model developed by Jain [51]. The packet train model considers packet arrivals as a sequence of bursts (so called trains) as opposed to independent events (so called cars) A train is defined as a series of arrivals, such that the elapsed time between any two packets does not exceed the maximum allowable intercar gap ....

R. Jain and S. A. Routhier. Packet Trains: Measurements and a New Model for Computer Network Traffic. IEEE Journal on Selected Areas in Communications, SAC-4(6):986--995, September 1986.

....estimation [59 61, 147, 148] Characterize the traffic in wormhole switched networks. The legacy from standard traffic theory strongly suggests that a Poisson model is appropriate [72,91] However, recent studies on buffered networks of various kinds argue that the Poisson assumption fails [77, 128]. Instead, the traffic has a self similar nature [101,151,156,159,160] This is attributed to high variability and long range dependence [42, 131] We conjecture that the traffic in wormhole switched networks has similar characteristics. Study if the GSMP view facilitates a parallel ....

JAIN, R., AND ROUTHIER, S. Packet trains - measurements and a new model for computer network traffic. IEEE Journal on Selected Areas in Communications 4, 6 (1986), 986--995. 37 37

....network node (i.e. LinkCost(T ) is used for all traffic) Regarding the workload, it is defined in terms of hsource node; destination nodei pairs. In each pair, the source produces data packets to be delivered to the destination. A source produces data packets according to a packet train model [13]. The workload consists of two parts, a delay sensitive workload and a throughput sensitive workload. For both parts, we use a uniform distribution of source destination pairs over the nodes of the network. Let parameter U (D) U (T ) denote the average number of sourcedestination pairs between ....

R. Jain and S.A. Routhier. Packet Trains - Measurements and A New Model for Computer Network Traffic. IEEE JSAC, 4(6):986--995, September 1986.

....than for the site s aggregate traffic to resemble that of another site. Another type of study analyzes the aggregate traffic seen on network links. These studies have focussed on the dynamics of packet arrivals on the link [FL91, LTWW94, PF95, WTSW95] the characteristics of packet flows [JR86, He90, CBP95], or on traffic patterns over particularly singular links, such as the trans Atlantic link connecting the U.S. and the U.K. CW91, WLC92] For link studies of local area networks [JR86, FL91, LTWW94, WTSW95] heterogeneity presents less of a problem than for those of wide area networks, because ....

....of packet arrivals on the link [FL91, LTWW94, PF95, WTSW95] the characteristics of packet flows [JR86, He90, CBP95] or on traffic patterns over particularly singular links, such as the trans Atlantic link connecting the U.S. and the U.K. CW91, WLC92] For link studies of local area networks [JR86, FL91, LTWW94, WTSW95], heterogeneity presents less of a problem than for those of wide area networks, because the latter encompass a much broader range of traffic sources and path characteristics than the former. Some wide area link studies attempt to address heterogeneity issues by analyzing traces from multiple ....

[Article contains additional citation context not shown here]

R. Jain and S. Routhier, "Packet Trains --- Measurements and a New Model for Computer Network Traffic," IEEE JSAC, 4(6), pp. 986-995, September, 1986.

....parameterize a simulation model of send side multiprocessor networking. ffl Non Poisson arrival processes. Recent work indicates that while some classes of network traffic are wellmodeled as Poisson processes, others are not [54] We plan to investigate how alternative models of packet arrivals [36, 54] impact our results. This will require modifying the simulation model s packet arrival process and re running the simulations. ffl Protocol thread scheduling heuristics. It may prove interesting to more carefully examine the conditions under which an additional protocol thread is scheduled. At ....

Raj Jain and Shawn Routhier. Packet trains: Measurements and a new model for computer network traffic. IEEE Journal on Selected Areas in Communications, 4(6):986--995, September 1986.

....components connected to it. In the current version, the user can choose between three different types of sourcesink pairs: file transfer (FTP) remote login (TELNET) and a simple workload (Simple Traffic) In FTP and TELNET, the source produces data packets according to a packet2 train model [Jain and Routhier, 1986]. Intuitively, each train corresponds to a connection. These workload models incorporate a flow control mechanism, and an acknowledgment with retransmission mechanism using roundtrip time estimates (see section 5.5 for details) The simple workload does not use the packet train model and has no ....

....components connected to it. In the current version, the user can choose between three different types of sourcesink pairs: file transfer (FTP) remote login (TELNET) and a simple workload (Simple Traffic) In FTP and TELNET, the source produces data packets according to a packettrain model [Jain and Routhier, 1986]. Intuitively, each train corresponds to a connection. These workload models incorporate a flow control mechanism, and an acknowledgment with retransmission mechanism using roundtrip time estimates. The simple workload does not use the packet train model and has no acknowledgment or retransmission ....

R. Jain and S.A. Routhier. Packet trains - measurements and a new model for computer network traffic. IEEE Journal on Selected Areas in Communication, SAC4 (6):986--995, September 1986.

....in the design of more efficient networks and provide a way to predict how added resources will affect performance. Currently there are five approaches being considered for modeling TCP IP traffic: Poisson related models [AMS82, HL86] Markov arrival processes [LCW94, Wan95] Packet Trains [JR86] . purely empirical models [DJ95] self similar distributions [LTWW93] These models predict very different trends in performance for increased number of data sources, increased buffer size, and node congestion. This causes skepticism about their usefulness in the design of future network ....

....distribution. The Packet Trains model, described by Jain, is essentially a two level Poisson model that introduces the concept of trains or bursts of packets. The arrival of trains is modeled with one 22 Poisson rate, and the arrival of cars within a train are modeled with a different rate. In [JR86], Jain validated parts of his model with data collected from a token ring network. However, aspects of the data could not be explained without modifications to the simple train theory. To this end, Jain extended the model with Tandem Trailer Trains (to describe RPC traffic) and the separation of ....

[Article contains additional citation context not shown here]

Jain, R. and S. Routhier. Packet Trains - Measurements and a New Model for Computer Network Traffic. IEEE Journal of Selected Areas in Communication 4, 6 (Sep) 1986, 986-995.

....4.2 Bursty Traffic Real network traffic is highly correlated from cell to cell [32] and so in practice, cells tend to arrive in bursts, corresponding perhaps to a packet that has been segmented or a packetized video frame. Many ways of modeling bursts in network traffic have been proposed [16] [21], 4] 32] Recently, Leland et al. 32] have demonstrated that measured network traffic is bursty at every level making it important to understand the performance of switches in the presence of bursty traffic. We illustrate the effect of burstiness on SLIP using an on off arrival process ....

Jain, R.; Routhier, S.A. "Packet Trains: measurements and a new model for computer network traffic," IEEE J. Selected Area Communications, Vol.4, pp.986-995, 1986.

.... of ATM switches, in terms of loss rate improved significantly when adding larger buffers [Gro96, RE96] Willinger et al. WTSW97] explain the self similar nature of Ethernet traffic observed on the packet level as the result of superposition of many ON OFF source (also referred to as packet trains [JR86]) where the length of the ON and OFF periods are drawn from heavy tailed distributions. Paxson and Floyd [PF95] show that Wide Area Network (WAN) traffic at the packet level is of asymptotic self similar nature, i.a. self similar behavior over large time scales. They propose a structural model ....

R. Jain and S.A. Routhier. Packet trains: Measurements and a new model for computer network traffic. IEEE Journal on Selected Areas in Communications, 4:986--995, 1986.

....Unix. This allowed use of the standard Unix data manipulation utilities to rapidly prototype utilities. 1.4 Related Work Studies of Networks and Protocols Related work includes studies to characterize the traffic of operational networks and systems built to collect data for such studies. In [10], Jain and Routhier developed a model for traffic on a token ring network using data collected by a monitoring system built by Feldmeier [6] More recently, Braden and DeSchon have developed a system NNStat [3] for gathering statistics from the Internet for traffic studies. All of these studies ....

Raj Jain and Shawn A. Routhier. Packet trains -- measurements and a new model for computer network traffic. IEEE Journal on Selected Areas in Communications, SAC-4(6):986--995, September 1986.

...., and a simple workload. However, we use only FTP workload because experience indicates that this is the major NSFNET application [17] The workload is defined in terms of source sink pairs which are attached to nodes. An FTP source produces data packets according to a packet train model [16]; intuitively, each train corresponds to a connection. This workload model incorporates a static send windowbased flow control mechanism and an acknowledgment with retransmission mechanism using roundtrip time estimates. Retransmissions can be due to node or link failures or buffer space ....

R. Jain and S.A. Routhier. Packet trains - measurements and a new model for computer network traffic. IEEE Journal on Selected Areas in Communication, SAC-4(6):986--995, September 1986.

....the response take longer time to come back and the sources do not generate new requests until the previous ones have been responded. This is more realistic than an open loop model where bursts are generated at a fixed rate regardless of the congestion in the network. The packet train model [25] used previously to model bursty behaviour is an open loop model. This model has the source sending out cars (having back to back packets) with an inter car time and trains having an inter train time. But, the source does not wait for any response from the destination. Note that the time between ....

R. Jain and S. Routhier, "Packet Trains - Measurement and a new model for computer network trafic," IEEE Journal of Selected Areas in Communications, Vol. SAC-4, No. 6, September 1986.

....requests) are generated depends upon the responsivity of the network. If the network is congested the response take longer time to come back and the sources do not generate new requests until the previous ones have been responded to. In an open loop traffic model like the packet train model [16], bursts are generated at a fixed rate regardless of the congestion in the network. Note that the time between two sets of requests (called a cycle time) is at least the sum of the time to transmit requests, the round trip time and the inter cycle time. Thus the idle time between two sets of ....

R. Jain and S. Routhier, "Packet Trains - Measurement and a new model for computer network trafic," IEEE Journal of Selected Areas in Communications, Vol. SAC-4, No. 6, September 1986, pp. 986-995.

....the polling design, incoming packets cannot be detected at all during any user or interrupt processing whatsoever, since the poll routine derives its flow of control from the idle loop. One could add polling points elsewhere in the operating system and try to take advantage of packet trains BibRef[36], but the polling rate needs to be quite high to catch a reasonable percentage of FDDI packets. Regardless, it is not unusual that a high performance workstation s operating system, given a single user workload, has all of its processes waiting, making the idleloop solution perform effectively. ....

R. Jain and S. Routhier, "Packet Trains: Measurements and New Model for Computer Network Traffic," IEEE Journal on Selected Areas in Communication, v4n6, 986-995, September 1986.

No context found.

Jain, R., Routhier, S.A., "Packet Trains - Measurements and a New Model for Computer Network Traffic", IEEE Journal on Selected Areas in Communications, Vol. 4, No. 6, September 1986, pp. 986 - 995.

No context found.

R. Jain and S. Routhier. Packet Trains: Measurements and a New Model for Computer Network Traffic. In IEEE J. Select. Areas Commun.,vol- ume 4, pages 1162--1167, 1986.

No context found.

R. Jain and S. A. Routhier, "Packet Trains - Measurements and a New Model for Computer Network Traffic," IEEE Journal on Selected Areas in Communications, vol. 4, pp. 986--994, Sept. 1986.

No context found.

R. Jain and S.A. Routhier, "Packet trains: measurements and a new model for computer network tra#c," IEEE J. Select. Areas Commun., Vol. 4 pp. 986-995, 1986.

No context found.

R. Jain and S. Routhier, "Packet trains -- measurements and a new model for computer network tra#c," IEEE J. Selected Areas in Communications, September 1986.

No context found.

R. Jain and S. A. Routhier. Packet trains---measurements and a new model for computer network tra#c. IEEE Journal on Selected Areas in Communications, SAC-4(6):986--995, September 1986.

No context found.

R. Jain and S. A. Routhier. Packet Trains - measurements and a new model for computer network tra#c. IEEE Journal on Selected Areas in Communications, SAC-4(6):986--995, September 1986.

No context found.

R. Jain and S. Routhier. Packet Trains: Measurements and a New Model for Computer Network Traffic. In IEEE J. Select. Areas Commun., volume 4, pages 1162--1167, 1986.

No context found.

R. Jain and S. A. Routhier, "Packet trains: measurements and a new model for computer network tra#c", IEEE Journal on Selected Areas in Communications, Vol.4:986-995, 1986.

No context found.

R. Jain and S. Routhier. Packet Trains - Measurement and a new model for computer network trafic. IEEE Journal of Selected Areas in Communications,, 1986.

No context found.

Raj Jain, Shawn A. Routhier, "Packet Trains-Measurements and a New Model for Computer Network Traffic", Integrated Broadband Networks, pp.157

No context found.

R. Jain and S. Routhier. Packet Trains - Measurement and a new model for computer network trafic. IEEE Journal of Selected Areas in Communications,, 1986.

No context found.

Jain, R. and Routhier, S. A. Packet Trains - Measurements and a New Model for Computer Network Traffic. IEEE Journal on Selected Areas in Communications, Vol. SAC-4, No. 6, September 1986, pp. 986-995.

No context found.

R. Jain and S. Routhier. Packet Trains - Measurement and a new model for computer networktrafi, IEEE Journal of Selected Areas in Communications,, 1986.

No context found.

R. Jain, S. Routhier. Packet trains: Measurements and a new model for computer network tra#c. IEEE Journal of Selected Areas in Communications, SAC-4(6):986--995, September 1986.

No context found.

R. Jain, S. Routhier, "Packet Trains: Measurements and a New Model for Computer Network Traffic," IEEE J. Select. Areas Commun., 4, pp. 1162-1167, 1986.

No context found.

R. Jain, S. Routhier, "Packet Trains: Measurements and a New Model for Computer Network Traffic," IEEE J. Select. Areas Commun., 4, pp. 1162-1167, 1986.

No context found.

RAJ JAIN,SHAWN A. ROUTHIER: "Packet Trains -- Measurements and a New Model for Computer Network Traffic" IEEE Journal on Selected Areas in Communications Vol Sac.4., 1986

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