S. Keshav, "REAL: A Network Simulator," University of California at Berkeley, Berkeley, CA, USA, Tech. Rep., 1988.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....of mobile wireless networks. It shares the same implementation infrastructure as IRLSim since it is implemented in PARSEC as well. At the other end, there are more general simulation packages that target a wide range of protocols and environments. Such include ns (and its predecessor REAL [15]) OPNET [9] and others. Those high end simulation packages are usually distinguished by the level of abstraction they offer and the way they address the issue of scalability. Most come with their own simulation language and a set of assorted protocol libraries. Most high end simulators provide ....

S. Keshav. REAL: a network simulator. Technical report, University of California, Berkeley, Dec 1988.

....increases, the highest link bandwidth that can be simulated decreases. Moreover, in Dummynet, routing tables are associated with incoming links rather than nodes. Thus, the simulator will not know how to route packets generated by a router, since they do not come from any link OPNET [3] REAL [12], ns [2] SSFnet [13] PARSEC and GloMoSim [14] represent the traditional network simulation approach. In this approach, the thread supporting event scheduler, application programs that generate network traffic, utility programs that configure, monitor, or gather statistics about a simulated ....

S. Keshav, "REAL: A Network Simulator," Technical Report 88/472, Dept. of computer Science, UC Berkeley, 1988.

....reliable link service 4 We thank Rooftop Communications Corporation for donating the toolkit. Parsa and Garcia Luna Aceves Improving TCP Performance over Wireless Networks 17 from TULIP is TCP, which in our implementation, has been ported from the TCP Reno code contained in the REAL simulator [19]. Broadcast Radio (802.11 spec) WIRP UDP IP TCP Reno FAMA NCS TULIP Figure 7. Protocol stack of wireless node. The code of the Snoop protocol [7] was modified from the on line FreeBSD implementation 5 to run in CPT. Logically, Snoop is at the network layer as it monitors packets at the ....

S. Keshav. Real: a network simulator. Technical report, University of California, Berkeley, Berkeley, California, 1988. Technical Report 88/472.

....forwarding. WIRP control packets are treated by TULIP as any other UDP packet (i.e. as unreliable link packets) The only higher level protocol requesting reliable link service from TULIP is TCP, which in our implementation, has been ported from the TCP Reno code contained in the REAL simulator [20]. The code of the Snoop protocol [8] was modified from the on line FreeBSD implementation 1 to run in CPT. Logically, Snoop is at the network layer as it monitors packets at the incoming interface and then passes them to IP. The code was maintained in its original on line form and only when ....

S. Keshav. REAL: A network simulator. Technical report, University of California, Berkeley, Berkeley, California, 1988. Technical Report 88/472.

....round trip times to characterize the network traffic. However, when the network paths are asymmetric, the round trip times are insufficient to present an accurate picture of the differences in the forward and the reverse paths. Some of the well known freely available network simulators are REAL [1], x Sim [2] and Network Simulator (ns) 3] REAL (REalistic And Large) was initially developed from the NEST (Network Simulation Testbed) for comparing the fair queuing gateway algorithm with first come first served scheduling, and is used for various purposes. X Sim is a network simulator based ....

S. Keshav, "REAL: A network simulator," Technical Report 88/472, Department of Computer Science, University of California at Berkeley, CA,

....of the RED gateways. Specific questions about the most efficient implementation of the RED algorithm are discussed in Section 11. 5 A simple simulation This section describes our simulator and presents a simple simulation with RED gateways. Our simulator is a version of the REAL simulator [19] built on Columbia s Nest simulation package [1] with extensive modifications and bug fixes made by Steven McCanne at LBL. In the simulator, FTP sources always have a packet to send and always send a maximal sized (1000 byte) packet as soon as the congestion control window allows them to do so. A ....

Keshav, S., "REAL: a Network Simulator", Report 88/472, Computer Science Department, University of California at Berkeley, Berkeley, California, 1988.

....traffic and the mechanisms of the new applications on the web and to evaluate the performance of new protocols. To achieve these goals different approaches can be taken. One of them is to use simulations. A number of successful simulation tools have been developed and applied recently, e.g. REAL [7] or NS [11] However, the applications of simulation approaches are not adequate in several cases since they eliminate the impacts of different network mechanisms such as the operating systems. Another way to achieve our goal is to develop and use network performance measurement tools, which use ....

Keshav, S., "REAL: A Network Simulator", Technical Report 88/472, Department of EECS, UC Berkeley. 1988.

.... study machine architectures while capturing operating system overheads [5, 21] Other efforts have focused on protocol level simulation with the ability to run the actual protocol stack during simulation [6] and network level simulation with a focus on routing and end to end protocol performance [22, 23]. Most relevant to our work is architecture level and protocol level simulation. Since END executes on its own processor concurrently with the host, it avoids any intrusion on the host operating system (including the protocol stack) executing on the host CPU. More importantly, END utilizes actual ....

S. Keshav, "REAL : A network simulator," UCB CS Tech Report 88/472, University of California, Berkeley, December 1988.

....assumptions in order to quantify the throughput degradation of TCP connections because of the phenomenon of ack compression in two way environments. The rest of the studies are all simulation based. Some of the network simulators used are, modified version of OPNET, modified version of REAL [4], STCP [21] and some other custom developed simulation tools. 7 Evaluation and Results This section reviews the research issues and results from the studies discussed in this paper. We use the figureFig. 5 to illustrate the essential components affecting the performance of TCP over ATM. 16 ....

Keshav, S., "REAL: a Network Simulator," Report 88/472, Computer Science Department, Univ. of California at Berkeley, CA, USA. 1988.

....and language independent mechanism for generic simulations, and is suitable for remote simulation. However, this work is targeted solely for generic simulations and would have to be extended to be applicable to domain specific contexts. Mature network simulation frameworks such as REAL (Keshav 1988), ns 2 (Fall and Varadhan 1998) INSANE (Mah 1998) and x Sim (Brakmo, Bavier, Peterson, and Raghavan 1997) provide rich APIs and tools, and have been extensively used by researchers. However, even with support for web based simulation in REAL, its flexibility is limited because it does not provide ....

Keshav S. 1988. REAL: A Network Simulator. UCB CS Tech Report 88/472. University of California, Berkeley, CA.

....with a test FTP application are given in Section 4. 2. Implementing a transport layer in user space At the beginning, this implementation was highly influenced by the IDLInet [1] testbed code. IDLInet code was originally implemented on three platforms: REAL, a packet level network simulator [4], a network of PCs running MS DOS and a network of PCs running Brazil (Plan 9 for PC) 2] and later ported to the IRIX and FreeBSD operating systems [5] In the IDLInet project, a full protocol stack was implemented. A socket like interface and a new protocol family are provided by ulib to handle ....

S. Keshav, "Real : A network simulator," Tech. Rep. UCB CS Tech Report 88/472, University of California, Berkeley, Dec. 1988.

....of the FB Mechanism 4.1 Simulation Model, Test Topology, and Metrics We carried out simulations to investigate the performance of the proposed FB mechanism in comparison to other existing mechanisms. For the simulations we used the the REAL network simulator with modifications made by its authors [13]. In these simulations we compare the performances of the following six congestion control mechanisms: the UBR scheme, the UBR EB scheme, the UBR FB scheme, the UBR RR scheme, the UBR EB RR scheme, and the UBR FB RR scheme. The first three use FIFO scheduling, while the last three use RR ....

S. Keshav, "REAL: A network simulator," UC Berkeley Technical Report, 1988.

....drawn for a connection from mobile to fixed and a connection connecting two mobiles. It is also assumed that the receiver can always send out acknowledgment immediately for each data received without delay other than processing delay. The TCP used in the simulation is based on the REAL simulator [11]. Summarized in Table I are some parameters used for the simulations. The assumptions and parameters used seem specific, but general conclusions drawn in the following remain valid for different values of the parameters tested. Wired link capacity 0.5Mbps Round trip wired link propagation delay ....

S. Keshav, "REAL: a network simulator", Report 88/472, Computer Science Department, University of California at Berkeley, CA, 1988.

....as a LEO satellite network or other large scale systems) is infeasible. We performed most of our simulation studies using the UCB LBNL network simulator known as ns, now widely used as part of the VINT project [8] ns is a event driven simulator originally derived from the REAL network simulator [72]. The simulator has an object oriented architecture, and simulation objects are typically implemented as split objects: partly in C , and partly in MIT s Object Tcl (OTcl) 141] Such objects exist simultaneously in both language realms, and functionality can typically be added in either language ....

S. Keshav. REAL: A Network Simulator. Technical Report 88/472, University of California, Berkeley, 1988.

....network topology used for the performance testing of STTP can be seen in figure 3. Here, we have 1. n transmitting nodes and 1. n receiving nodes. In each case, node t1 transmits to node r1, node t2 to node r2 etc. Similar topologies are recommended by Keshav in his benchmark suite for the REAL [19] simulator. By varying the number of sources and the link capacity, we are able to create a wide variety of scenarios under which to test STTP, TCP Reno and TCP Vegas. Our reference protocol is TCP Reno, as a large amount of modelling and simulation has been conducted in the past, which provides ....

S. Keshav, "Real: A network simulator," Tech. Rep. 88/472, Department of Computer Science, UC Berkeley, 1998.

.... that yield three slow start phases in a cycle (see Figure 7 in Section 5) However in the vast majority of cases we find that the evolutions of W (t) include a single or two slow start phases per cycle, as shown in Figures 3 and 4, respectively (the figures were obtained with the REAL simulator [13]) In the next two sections, we analyze the case of a single slow start phase per cycle (Section 3) and the case of two slow start phases per cycle (Section 4) and we find the conditions when these cases occur. RR n3142 8 E. Altman, J. Bolot, P. Nain, D. Elouadghiri, M. Erramdani, P. Brown, D. ....

.... 1 N 2 N 3 T 1 T 2 T 1 T 2 T 3 and the average round trip is given by rtt = N 1 R 1 N 2 R 2 N 1 R 1 N 2 R 2 N 3 R 3 N 1 N 2 N 1 N 2 N 3 : 5 Simulation results We next present some results from extensive simulation work with the REAL simulator [13] done to validate the analytical results above. The differences between the simulation and the analytic models are i) the simulation model deals with discrete sized packets (while the simulation model deals with fluid flows) and ii) the simulation model incorporates the exact and complete TCP ....

S. Keshav, "REAL: A network simulator", see also http://www.cs.cornell.edu/Info/People/skeshav/

....preferential behavior is also absent in simulations with Random Drop instead of Drop Tail gateways. 3 1 2 4 SINK bandwidth 8000 kbps bandwidth 800 kbps GATEWAY FTP SOURCE FTP SOURCE d = 5 ms d d 1,3 2,3 3,4 100 ms Figure 3: Simulation network. Our simulator is a version of the REAL simulator (Keshav, 1988) built on Columbia s Nest simulation package (Bacon et al. 1988) with extensive modifications and bug fixes made by Steven McCanne at LBL. The gateways use FIFO queueing, and this section s simulations use Drop Tail on queue overflow. FTP sources always have a packet to send and always send a ....

....throughput ( 1.0 1.1 1.2 1.3 1.4 1.5 0 20 40 60 80 100 Figure 11: Node 1 s throughput with the DECbit scheme. roundtrip time is varied by varying 2 3 . The roundtrip time for node 1 packets is still 221.44 ms. These simulations use the implementation of the DECbit scheme in the REAL simulator (Keshav, 1988). Each simulation was run for 200 seconds. Figure 11 represents each 50 second interval (excluding the first 50 second interval) by a dot showing node 1 s throughput for that interval. The line shows node 1 s average throughput. For the simulations where the two roundtrip times differ by less than ....

[Article contains additional citation context not shown here]

Keshav, S., "REAL: a Network Simulator", Report 88/472, Computer Science Department, University of California at Berkeley, Berkeley, California, 1988.

....TCP performance evaluations have been conducted via simulation (e.g. FF96, Hoe96] A large variety of simulators for modeling internetworking protocols exist, and are currently used by researchers. Some of these tools are OpNet [Tec] x sim [BP96] the Network Simulator [Hey90] REAL [Kes88] ns [MF95] as well as specialized simulators written by various researchers for their own use. Picking a simulator to use is sometimes complicated. We encourage researchers to investigate various simulators and choose the one best geared towards the research at hand. For instance, one simulator ....

Srinivasan Keshav. REAL: A Network Simulator. Technical Report 88/472, University of California Berkeley, 1988.

No context found.

S. Keshav, "REAL: A Network Simulator," University of California at Berkeley, Berkeley, CA, USA, Tech. Rep., 1988.

No context found.

S. Keshav. REAL: A network simulator. Technical Report 88/472, University of California, Berkeley, 1988.

No context found.

S. Keshav, "REAL: A network simulator," Tech. Rep. 88/472, University of California, Berkeley, 1988.

No context found.

S. Keshav. REAL: A network simulator. Technical Report 88/472, University of California, Berkeley, 1988.

No context found.

S. Keshav, "REAL: A Network Simulator", Computer Science Department Technical Report 88/472, UC Berkeley, 1988.

No context found.

Srinivasan Keshav, "REAL: A Network Simulator," Computer Science Department Technical Report, University of California, December 1988

No context found.

Keshav, S., \REAL: A Network Simulator," Technical Report

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