David W. Bauer’s research while affiliated with MITRE and other places

What is this page?


This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.

Publications (22)


Scalable RF propagation modeling on the IBM Blue Gene/L and Cray XT5 supercomputers
  • Conference Paper

December 2009

·

29 Reads

·

13 Citations

David W. Bauer

·

We present a performance analysis for a highly accurate, large-scale electromagnetic wave propagation model on two modern supercomputing platforms: the Cray XT5 and the IBM Blue Gene/L. The electromagnetic wave model is used to simulate the physical layer of a large-scale mobile ad-hoc network of radio devices. The model is based on the numerical technique called Transmission Line Matrix, and is implemented in a Time Warp simulation package that employs reverse computation for the rollback mechanism. Using Rensselaer's Optimistic Simulation System we demonstrate better than real-time, scalable parallel performance for network scenarios containing up to one million mobile radio devices, highly accurate RF propagation and high resolution, large-scale complex terrain.


Fig. 8. Image of a single wave transmission in the TLM model. As the Z-axis is truncated, the wave becomes donut-shaped as it continues to propagate in the XY directions.
Fig. 9. The GVT_interval and batch_size surface plot. GVT_interval had the largest effect on the performance of the TLM model.
Scalable Time Warp on Blue Gene Supercomputers
  • Conference Paper
  • Full-text available

June 2009

·

60 Reads

·

85 Citations

n this paper we illustrate scalable parallel performance for the TimeWarp synchronization protocol on the L and P variants of the IBM BlueGene supercomputer. Scalable Time Warp performance for models that communicate a large percentage of the event population over the network has not been shown on more than a handful of processors. We present our design for a robust performing Time Warp simulator over a variety of communication loads, and extremely large processor counts-- up to 131,072. For the PHOLD benchmark model using 65,536 processors, our Time Warp simulator produces a peak committed event-rate of 12.26 billion events per second at 10\% remote events and 4 billion events per second at 100\% remote events, the largest ever reported. Additionally, for the Transmission Line Matrix (TLM) model which approximates Maxwell's equations for electromagnetic wave propagation, we report a committed event-rate in excess of 100 million on 5,000 processors with 200 million grid-LPs. The TLM model is particularly challenging given the bursty and cubic growth in event generation. Overall, these performance results indicate that scalable Time Warp performance is obtainable on high-processor counts over a wide variety of event scheduling behaviors and not limited to relatively low, non-bursty rates of off-processor communications.

Download

An approach for the effective utilization of GP-GPUS in parallel combined simulation

December 2008

·

36 Reads

·

11 Citations

Proceedings - Winter Simulation Conference

A major challenge in the field of Modeling & Simulation is providing efficient parallel computation for a variety of algo- rithms. Algorithms that are described easily and computed efficiently for continuous simulation, may be complex to de- scribe and/or efficiently execute in a discrete event context, and vice-versa. Real-world models often employ multi- ple algorithms that are optimally defined in one approach or the other. Parallel combined simulation addresses this problem by allowing models to define algorithmic compo- nents across multiple paradigms. In this paper, we illustrate the performance of parallel combined simulation, where the continuous component is executed across multiple graphical processing units (GPU) and the discrete event component is executed across multiple central processing units (CPU).


Figure 1: Conceptual health care agency IT infrastructure. Computers in the inner-most section (green) have been sufficiently secured to contain patient data, and would form the basis for a Cloud Computing platform. 
Figure 2: 
Figure 5: Spatio-temporal domain scheduling performance for individual addresses. 
Figure 6: Decomposition of the time series.
An application of parallel Monte Carlo modeling for real-time disease surveillance

December 2008

·

68 Reads

·

3 Citations

The global health, threatened by emerging infectious dis- eases, pandemic influenza, and biological warfare, is be- coming increasinglydependent ontherapidacquisition, pro- cessing, integration and interpretation of massive amounts of data. In response to these pressing needs, new infor- mation infrastructures are needed to support active, real time surveillance. Detection algorithms may have a high computational cost in both the time and space domains. High performance computing platforms may be the best approach for efficiently computing these algorithms. Un- fortunately, these platforms are unavailable to many health care agencies. Our work focuses on efficient parallelization of outbreak detection algorithms within the context of cloud computing as a high throughput computing platform. Cloud computing is investigated as an approach to meet real time constraints and reduce or eliminate costs associated with real time disease surveillance systems.


Optimistic parallel discrete event simulation of the event-based transmission line matrix method

January 2008

·

29 Reads

·

14 Citations

Proceedings - Winter Simulation Conference

In this paper we describe a technique for efficient parallelization of digital wave guide network (DWN) models based on an interpretation of the finite difference time domain (FDTD) method for discrete event simulation. Modeling methodologies based on FDTD approaches are typically constrained in both the spatial and time domains. This interpretation for discrete event simulation allows us to investigate the performance of DWN models in the context of optimistic parallel discrete event simulation employing reverse computation for rollback support. We present parallel performance results for a large-scale simulation of a 3D battlefield scenario, 100 km and at a height of 100 m with a resolution of 100 m in the X-, Y-planes, and 10 m in the Z-plane for 754 simultaneous radio wave transmissions.


An Approach for Incorporating Rollback through Perfectly Reversible Computation in a Stream Simulator

July 2007

·

21 Reads

·

7 Citations

The traditional rollback mechanism deployed in optimistic simulation is state-saving. More recently, the method of reverse computation has been proposed to reduce the amount of memory consumed by state-saving. This method computes the reverse code for the model during rollback execution, rather than recalling saved state memory. In practice, this method has been shown to offer memory-efficiency without sacrificing computational efficiency. In order to support reverse codes in the model, events must continue to be preserved in the system until fossil collection can be performed. In this paper we define a new algorithm to support perfectly reversible model computation that does not depend on storing the full processed event history. This approach improves memory consumption, further supporting large-scale simulation.


Fig. 1. Amdahl's Law illustrates effective speedup.
Table 1 . Parallel Execution Performance of the Scan Statistic Model.
High Performance Computing for Disease Surveillance

January 2007

·

150 Reads

·

2 Citations

Lecture Notes in Computer Science

The global health, threatened by emerging infectious dis- eases, pandemic in∞uenza, and biological warfare, is becoming increas- ingly dependent on the rapid acquisition, processing, integration and interpretation of massive amounts of data. In response to these pressing needs, new information infrastructures are needed to support active, real time surveillance. Space-time detection techniques may have a high com- putational cost in both the time and space domains. High performance computing platforms may be the best approach for e-ciently computing these techniques. Our work focuses on e-cient parallelization of these computations on a Linux Beowolf cluster in order to attempt to meet these real time needs.


Eliminating Remote Message Passing in Optimistic Simulation

December 2006

·

6 Reads

·

2 Citations

Proceedings - Winter Simulation Conference

This paper introduces an algorithm for parallel simulation capable of executing the critical path without a priori knowledge of the model being executed. This algorithm is founded on the observation that each initial event in a model causes a stream of events to be generated for execution. By focusing on the parallelization of event streams, rather than logical processes, we have created a new simulation engine optimized for large scale models (i.e., models with 1 million LPs or more)


Table 1 : Stages of the BGP decision algorithm for route selec- tion.
Table 2 : Rocketfuel ISP Topology Parameters
Table 8 : Design 3: Analyze performance of protocol models under varying degrees of network stability and link weight management.
Detail of parameter space for the large-scale OSPF and BGP experiment designs.
Variation in the optimization of different perspectives. This table illustrates the tradeoffs made for each particular op- timization. Bold values are optimization results.
A Case Study in Understanding OSPF and BGP Interactions Using Efficient Experiment Design

January 2006

·

2,643 Reads

·

18 Citations

In this paper, we analyze the two dominant inter- and intradomain routing protocols in the Internet: Open Shortest Path Forwarding (OSPFv2) and Border Gateway Protocol (BGP4). Specifically, we investigate interactions between these two routing protocols as well as overall (i.e. both OSPF and BGP) stability and dynamics. Our analysis is based on large-scale simulations of OSPF and BGP, and careful design of experiments (DoE) to perform an efficient search for the best parameter settings of these two routing protocols


Eliminating remote message passing in optimistic simulation.

January 2006

·

53 Reads

·

1 Citation

This paper introduces an algorithm for parallel simulation capable of executing the critical path without a priori knowledge of the model being executed. This algorithm is founded on the observation that each initial event in a model causes a stream of events to be generated for execution. By focusing on the parallelization of event streams, rather than logical processes, we have created a new simulation engine optimized for large scale models (i.e., models with 1 million LPs or more). We have affectionately named our implementation Adirondack for the numerous streams contained in the Adirondack State Park in New York.


Citations (16)


... It does not assume a parametric model, but instead combines empirical p-values across multiple locations, days, and data streams to detect anomalies. A discrete event model is used by Beeker, Bauer and Mohtashemi [10] to simulate the spread of infectious diseases through an agent-based, stochastic model of transmission dynamics. The objective is to generate a benchmark from a network of individual contacts in an urban environment using publicly available population data. ...

Reference:

Extensions of the Scan Statistic for the Detection and Inference of SpatialClusters
Benchmark Data Generation from Discrete Event Contact Network Models
  • Citing Article

... The automated surveillance system assists the human operator in scene analysis and event classification by automatically detecting the objects and analysing their behaviour using computer vision, pattern recognition and signal processing techniques. Bauer and Mohtashemi (2008) proposed utilisation of cloud computing (http:// en.wikipedia.org/wiki/Cloud_computing) approach as a low-cost computational platform to address critical challenges and limitations in today's surveillance systems and introduced novel concept to facilitate disease search and SA. ...

Surveillance in the Cloud: A New Platform for Disease Search and Situational Awareness

... In this paper, we suggest an approach to perfectly reversible computation (PRC) which supports rollback in an optimistic simulation using only the current LP state and current event. The approach is defined in the context of the Adirondack stream simulator developed by Bauer and Carothers [2]. The remainder of the paper is organized as follows. ...

Eliminating Remote Message Passing in Optimistic Simulation
  • Citing Conference Paper
  • December 2006

Proceedings - Winter Simulation Conference

... Parallel discrete event simulation (PDES) is a computational technique used to simulate dynamic behaviors of complex systems, including internet and cybersecurity simulations, transportation and mobility applications, and hardware co-design simulations [24]. Despite the success of PDES modeling frameworks such as ROSS [4] and CODES [23], the simulation requirements and computational complexity are growing at an intractable rate. For instance, PDES requires four hours to simulate behaviors of a 4,096-node system over a 12 millisecond period [26]. ...

ROSS: A high-performance, low-memory, modular Time Warp system
  • Citing Article
  • November 2002

Journal of Parallel and Distributed Computing

... Constructing DBN metamodels involves selecting an optimal set of time instants at which simulation data must be collected to estimate the probability distributions for the simulation state, and conducting what-if analyses involves fixing the simulation state at a given point in time and updating the simulation state conditional probability distributions using the observed value of the simulation state. Another related technique involves metasimulation Bauer et al. (2004). They develop a simulation of internet and telephone networking protocols and propose the use of a random search approach to the design of experiments for finding key parameters and interaction patterns that affect key network performance measures. ...

A Case Study in Meta-Simulation Design and Performance Analysis for Large-Scale Networks.
David W. Bauer

·

Garrett R. Yaun

·

·

[...]

·

Shivkumar Kalyanaraman

... Subsequently, doctors can monitor patients to improve recovery and survival. Bauer and Mohtashemi [29] proposed a parallel Monte Carlo model using cloud computing as a method to not only meet real-time monitoring constraints in the medical field but also reduce or eliminate the costs associated with real-time disease monitoring systems. Meir and Rubinsky [30] introduced a new paradigm of medical technologies focusing on wireless technologies and cloud computing that were designed to overcome the growing cost of medical technology. ...

An application of parallel Monte Carlo modeling for real-time disease surveillance

... Bauer et al. [14] consider a combined continuous-discrete simulation and assign the continuous part the GPU and the discrete part to the CPU. The benchmark model PHOLD [46] is employed to explore different GPU configurations by varying the thread block size, the number of floating point instructions, the data transfer volume, and the communication pattern. ...

An approach for the effective utilization of GP-GPUS in parallel combined simulation
  • Citing Conference Paper
  • December 2008

Proceedings - Winter Simulation Conference

... In the context of simulating computer systems, a lot of models exist already. Examples include the simulation of network interaction (Szymanski, Liu, and Gupta 2003), (Bauer et al. 2003), (Kamal 2004) or memory hierarchies (Over, Strazdins, and Clarke 2005), (Iyer 2003), (de Farias et al. 2001). Common to these models is that they are typically reduced to certain aspects of interest. ...

Simulation of large scale networks III: ROSS.Net: optimistic parallel simulation framework for large-scale internet models.

... The topological properties of k-ary n-cubes have been explored in [23,24]. Due to the advantageous topological properties of 3-ary n-cube, Q 3 n such as symmetricity, pancyclicity, short message latency and easy implementation it has been utilised to build multicomputers such as the Cray XT5, Blue Gene/L supercomputers [25] and CamCube [26] systems. Embedding problem on 3-ary n-cubes is extensively studied on paths, cycles with faulty nodes and links [27,28]. ...

Scalable RF propagation modeling on the IBM Blue Gene/L and Cray XT5 supercomputers
  • Citing Conference Paper
  • December 2009

... In particular, seminal work by Chandy and Misra [12], Bryant [1], and Jefferson [22] laid the groundwork for parallel discrete event simulation, which was shown to run efficiently on supercomputers [4,5]. Previous traffic simulators that utilize PDES include those by Perumalla [33] and Yoginath [47], who used optimistic PDES for modeling traffic grids, though they evaluated their system on smaller-scale synthetic grid networks and used a different mechanism to mediate congestion among vehicles. ...

Scalable Time Warp on Blue Gene Supercomputers