An HLA-Based Approach to Quantify Achievable Performance for Tactical Edge Applications

Abstract : The DoD is pursuing an end-to-end, seamless, network-centric enterprise communications infrastructure to support a wide range of operating conditions and network topologies. Evaluating the achievable performance of this communications infrastructure, as it evolves, is essential to the user community in order to guide their ongoing requirements, design, and procurement activities. Tactical edge applications present significant challenges to network evaluation methods since they often include mobile ad-hoc networks (MANETs) that employ a wide range of platform types (ground-based, air-based, and satellite-based), traffic types (data, voice, video, and multimedia), delivery methods (unicast and multicast), offered traffic loads (kilobits/sec through megabits/sec), and numbers of nodes (from 10s to 1000s). The complexity exhibited by tactical edge applications typically demands the use of Modeling and Simulation (M&S) techniques, supported by high-fidelity models, to adequately quantify achievable performance on an end-to-end basis. However, these high-fidelity models often have very long runtimes, and restrictive limitations on scenario sizing. We investigate the application of the DoD High Level Architecture (HLA) and High Performance Computing (HPC) platforms to address the performance demands associated with analyzing tactical edge applications. A federation comprised of two Soldier Radio Waveform (SRW) federates and one Wireless Network after Next (WNaN) federate is developed and executed within an HPC environment at Aberdeen Proving Grounds (APG). High-fidelity OPNET models are used to represent the SRW and WNaN waveforms. Situational Awareness (SA) multicast traffic is delivered among the nodes represented within each of the three federates. Unicast traffic is exchanged between the SRW federates, in the presence of this SA background traffic, using the WNa

[1]  David S Alberts,et al.  Network Centric Warfare: Developing and Leveraging Information Superiority , 1999 .

[2]  Christopher D. Carothers,et al.  On deciding between conservative and optimistic approaches on massively parallel platforms , 2010, Proceedings of the 2010 Winter Simulation Conference.

[3]  Xinye Zhao,et al.  Basic Study on RTI Interoperability , 2010, 2010 Second International Conference on Computer Modeling and Simulation.

[4]  J. Steinman.  WarpIV Kernel : Real Time HPC-RTI Prototype , 2009 .

[5]  G. Amdhal,et al.  Validity of the single processor approach to achieving large scale computing capabilities , 1967, AFIPS '67 (Spring).

[6]  Christopher D. Carothers,et al.  Scalable RF propagation modeling on the IBM Blue Gene/L and Cray XT5 supercomputers , 2009, Proceedings of the 2009 Winter Simulation Conference (WSC).

[7]  R. Fujimoto.  Parallel and distributed simulation , 1995, Winter Simulation Conference Proceedings, 1995..

[8]  Yuan Shi.  Reevaluating Amdahl's Law and Gustafson's Law , 1996 .

[9]  Christopher D. Carothers,et al.  Scalable Time Warp on Blue Gene Supercomputers , 2009, 2009 ACM/IEEE/SCS 23rd Workshop on Principles of Advanced and Distributed Simulation.

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

[11]  Steve L. Ferenci,et al.  RTI Performance on Shared Memory and Message Passing Architectures , 1999 .

[12]  Kalyan S. Perumalla,et al.  μπ: a scalable and transparent system for simulating MPI programs , 2010, SimuTools.

[13]  Stuart Faulk,et al.  Productive petascale computing: requirements, hardware, and software , 2009 .