Looking ahead of real time in hybrid component networks

While network simulation is a wider used technique for developing, testing and evaluating network protocols and applications, it is difficult and time-consuming to develop detailed models of the entire protocol stack and the application that is being evaluated. Hybrid component networks combine simulation models of protocols and applications with physical implementations of other protocols and/or applications, which can alleviate modeling complexity and build accurate models. However, hybrid component networks require that the simulated components execute faster than real time. Parallel simulation techniques can greatly improve the performance of hybrid component networks and facilitate evaluation of larger networks than would be feasible only with sequential execution. In parallel simulation, one of the most important properties to be exploited from the simulation is "lookahead" but noise of the previous studies has exploited the lookahead front the physical implementation of protocols or applications. We have developed Virtual Socket Interface (VSI), which exploits real time lookahead from the interactions between operational applications and simulated networks in GloMoSim, a parallel discrete event network simulator. This paper presents the implementation of VSI with the motion of real time lookahead and evaluates VSI for typical wireless ad hoc network scenarios. The results of the study demonstrate that the parallel situation with the real time lookahead greatly enhances the capability of hybrid component networks.

[1]  Mario Gerla,et al.  GloMoSim: A Scalable Network Simulation Environment , 2002 .

[2]  J. Broach,et al.  The dynamic source routing protocol for mobile ad-hoc networks , 1998 .

[3]  Charles E. Perkins,et al.  Ad-hoc on-demand distance vector routing , 1999, Proceedings WMCSA'99. Second IEEE Workshop on Mobile Computing Systems and Applications.

[4]  Zhen Liu,et al.  Evaluation of TCP Vegas: emulation and experiment , 1995, SIGCOMM '95.

[5]  Brian W. Unger,et al.  Applying parallel discrete event simulation to network emulation , 2000, Proceedings Fourteenth Workshop on Parallel and Distributed Simulation.

[6]  David A. Maltz,et al.  Dynamic Source Routing in Ad Hoc Wireless Networks , 1994, Mobidata.

[7]  R. M. Fujimoto,et al.  Parallel discrete event simulation , 1989, WSC '89.

[8]  Theodore S. Rappaport,et al.  Wireless communications - principles and practice , 1996 .

[9]  David B. Ingham,et al.  Delayline: A Wide-Area Network Emulation Tool , 1994, Comput. Syst..

[10]  R.M. Fujimoto,et al.  Parallel and distributed simulation systems , 2001, Proceeding of the 2001 Winter Simulation Conference (Cat. No.01CH37304).

[11]  David A. Maltz,et al.  A performance comparison of multi-hop wireless ad hoc network routing protocols , 1998, MobiCom '98.

[12]  Leo Monteban,et al.  WaveLAN®-II: A high-performance wireless LAN for the unlicensed band , 1997, Bell Labs Technical Journal.

[13]  Bruce Tuch,et al.  Development of WaveLAN®, an ISM band wireless LAN , 1993, AT&T Technical Journal.

[14]  Humayun Akhtar An overview of some network modeling, simulation and performance analysis tools , 1997, Proceedings Second IEEE Symposium on Computer and Communications.

[15]  Atsushi Iwata,et al.  Scalable routing strategies for ad hoc wireless networks , 1999, IEEE J. Sel. Areas Commun..

[16]  Mineo Takai,et al.  Parssec: A Parallel Simulation Environment for Complex Systems , 1998, Computer.

[17]  Randy H. Katz,et al.  Trace-based mobile network emulation , 1997, SIGCOMM '97.

[18]  Gordon S. Blair,et al.  A Network Emulator to Support the Development of Adaptive Applications , 1995, Symposium on Mobile and Location-Independent Computing.

[19]  W. Richard Stevens,et al.  TCP/IP Illustrated, Volume 2: The Implementation , 1995 .

[20]  Charles E. Perkins,et al.  Ad hoc On-Demand Distance Vector (AODV) Routing , 2001, RFC.